I, like many others, was shocked by the outcome of the 2016 Presidential election. So many polls and prognosticators had presumed Hillary Clinton’s success that when the results began to come in (I was at work at the time) I could hardly believe them. States like Virginia which had seemed to shift so strongly towards Obama’s powerful electoral coalition of young voters, ethnic minorities and working class voters became too close to call, all while Obama’s so-called firewall in the Midwest crumbled. Clearly my assumptions and the prevailing wisdom surrounding the election was flawed; indeed, the polls themselves were wrong (though mostly within the margin of error). Many of the predictions people made in the election were based on demographic changes that have taken place in the US, with some even declaring “demographics is destiny.” Obviously many of those assumptions regarding demographics were incorrect. I would like to look back on the election and critically examine some of the assumptions that were made about demographics, and also about how voters views of the candidates framed their decision making in surprising ways. Here are some of the narratives I’d like to critically examine from the 2016 election:
1. The minority vote didn’t shift dramatically in either turnout or proportion of support for Democrats like it was supposed to.
Before the voters went to the polls in November, there was a dominant narrative that minority voters, offended by Trump’s rhetoric about them would engender massive support for Hillary. This did not happen, and while people have focused on the identity politics of this election, according to the various exit polls Trump only gained +1% of white voters in this election compared to Mitt Romney. In fact, Trump performed better among non-white voters than his predecessor. 
(Source)
Hillary also failed to secure the same levels of turnout among minorities that Obama was able to in 2008 and 2012, which proved decisive in Midwestern states. In Michigan, turnout in mostly-black Detroit was depressed, and if Hillary had simply matched Obama’s turnout in Wayne County she probably would have won the entire state, where the margin was just 10,704 votes. The conventional wisdom that minority voters would vote uniformly and turnout in higher numbers against Donald Trump proved mostly untrue, though not completely so. I’ll speak more to this later on.
2. Many college-educated whites did switch support from Republican to Democrat, but not enough to offset the surge in support among non college-educated whites for Trump.
In a way my critique here relates to both the pre-election and post-election narratives. There was a long standing presumption that college-educated whites, who had traditionally supported Republican candidates would abandon the party and Hillary’s coalition of minority voters and college graduates would overwhelm Trump in swing states. In the end there was a shift against Trump among college-educated whites, but it was more than offset by the surge in support for him by non college-educated whites, who voted for him decisively:
(Source)
Hillary performed better among college-educated voters and higher income voters than Obama did in 2012, but lost a considerable amount of support among non college-educated voters and individuals who make less than $30,000 a year. While the narrative of overwhelming opposition to Trump from college educated whites didn’t come to fruition, I also have a problem with the post-election takeaway that Trump won with overwhelming support from white voters alone. According to the NYT exit polls Trump only performed slightly better among whites overall than Mitt Romney, which is supported by other exit polling. While the rhetoric surrounding the election focused heavily on race, the outcome proved that class and education were similarly important dividing lines. If the Republican Party has a problem with minority voters the Democrats have a similar problem with working-class white voters. I doubt leaders in either party are entirely happy with the strength of their respective electoral coalitions after this election.
3. “Sensible” Republican voters were supposed to defect en masse from the Republican Party to Hillary Clinton.
Trump was supposed to split the Republican Party and cause an end to the traditional Two Party system as we knew it. Voters didn’t see it that way, and in the end Trump held on to as many of his Party’s voters as Hillary, while also winning independent and Third-Party voters:
(Source)
On this, there was an argument among pundits that Trump’s personal temperament and would prevent him from receiving the full support of his Party. What exit polls show is that more people who viewed Trump unfavorably voted for him than did with Hillary Clinton:
(Source)
Part of what frustrates me with this narrative of division among Republicans is its persistence after the election. Journalists insist on asking elected Republicans how their fractured Party can unite around their president, when the election gives a pretty simple answer: Republican voters rallied around both Trump and other Republican candidates in the election, until Republican voters show signs of disunity it is hard to imagine their elected officials will.
4. Hillary gained support in red states, but not by enough to offset the surge of support in the Midwest for Trump.
Before the election we continued to read about the crumbling support for Trump in traditionally strong Republican states. In the end Hillary was unable to flip a single state Mitt Romney won in 2012, but if you look deeper at the numbers, she did perform better in some red states compared to Obama, due in part to changing demographics:
(Source and great interactive)
This is partially how Hillary was able to win the popular vote by nearly three-million votes while losing the electoral college. Turnout and support for Trump surged in the Midwestern battleground states while Hillary’s gains in places Texas and Arizona, while impressive, were unable to effect the outcome of the election.
Finally, this was a low turnout election. While there were more overall votes cast in 2016 than in 2012 this fell short of overall population growth during that period. Hillary was unable to turnout the same electoral coalition that secured Obama’s two election victories. Due to Trump’s unique support from working class and non college-educated whites in the Midwest he was able to win Pennsylvania, Wisconsin, and Michigan by just 77,744 combined votes, which ended up determining the outcome of the 2016 Presidential election.
I think it’s important for people from every political persuasion to accept when biases get in the way of objectivity and admit when our own assumptions get the better of us. For me, the outcome of the 2016 election challenged a lot of pre-conceived notions I held about politics in the US. Being able to critically examine my own biases is, in a way, therapeutic given the shock I felt after the results of the election.
Lee's lectern 