Did either Donald Trump’s overt racism or faux economic populism get him elected president?

Two years later, the thesis of a thousand election postmortems is that the answer is “yes” to one or both of these questions.  Hence the endless think pieces about how Democrats can get enough of those precious white working class votes by adopting the populism (apparently without the faux part) while not pandering to the racism.

The (usually implicit) claim of these arguments is that Trump’s phony economic populism, plus his alt-right air raid siren racism/nativism were the key factors that proved more successful than Mitt Romney’s standard issue GOP Up With Rich People rhetoric, combined with the standard issue movement conservatism dog whistle racism/nativism.

Yet looking at the actual votes in the two elections, I’m struck by the extent to which Trump’s and Romney’s electoral performances were almost indistinguishable — not only in terms of total votes received, but in terms of the percentage of the vote they each got from almost every major demographic category.  Indeed, to the extent there were any differences, these differences were exactly the opposite of what differences one would expect to find if the open racism + economic populism as roadmap for success thesis was correct.

Let’s start with Real Americans:

Percentage of white voters who voted for Romney:  59%

Percentage of white voters who voted for Trump: 58%

White men for Romney: 62%

White men for Trump: 62%

White women for Romney: 56%

White women for Trump: 52%

Hmmm.  If you slice and dice the white vote some more, you do get some other differences: specifically, compared to Romney versus Obama, whites without a college degree were quite a bit more likely to vote for Trump than for Clinton, while the gap among college educated whites was fairly small. (College educated white women went 51% to 44% for Clinton). Still, what’s most striking is the overall stability of white voting patterns in the two elections.

Now — at least in terms of the “overt racism made the difference” argument — things get really weird:

Black men for Romney: 11%

Black men for Trump: 13%

Black women for Romney: 3%

Black women for Trump: 4%

Latino men for Romney: 33%

Latino men for Trump:  32%

Latino women for Romney: 23%

Latino women for Trump: 25%

These differences are all small, and push up against statistical significance, but nevertheless, Trump beats Romney in seven of eight categories among men and women African Americans and Latinos.

I don’t have comparable data for Asian-Americans, but in the “other” category which we can assume is dominated by Asian-Americans, Romney and Trump both got 31% of the vote.

Overall, Romney received 47.15% of the vote, as compared to 45.95% for Trump.  Trump, in short, got a smaller percentage of the vote than Romney because, in part, fewer white people voted for Trump than for Romney.  But really, the main point is that Trump’s support, both in terms of total votes and their distribution across the voting population, was practically identical to Romney’s.

The other part of that “in part” was the one real difference between the 2012 and 2016 votes, which is that the minor party vote more than tripled.  In 2012 minor party candidates got 1.48% of the total vote: Johnson, the Libertarian candidate got .99% of the vote, and Stein, the Green Party candidate, got .36%, with the rest going to a handful of others.  In 2016, Johnson got 3.28%, Stein got 1.07%, Evan McMullin (who really only campaigned in Utah) got 0.54%.  The total minor party vote was 5.04% — a 241% increase over 2012.

Hillary Clinton got exactly three percent (OK 3.01%) less of the total vote than Barack Obama received.  How much of that three percent disappeared into the abyss of the five million extra votes that minor party candidates received in 2016 beyond what they got in 2012? Of course the really critical question is how many minor party votes, or votes that weren’t cast at all because of the same sorts of motivations that lead people to waste their votes on vanity candidates like Johnson and Stein, would have gone to Clinton rather than Trump in the three rust belt states that cost Clinton the election?

But again, the bottom line appears to be that Republicans voted for Trump in exactly the same way they voted for Romney, with just a little bleed-off among white voters, specifically white women.  That certainly doesn’t fit with the thesis that either Trump’s faux populism or his open racism/nativism were the decisive factors in the election.  And that seems like both good and bad news.