A big part of the hope and fear partisans have about Donald Trump’s general election prospects is that his atavistic message of resentment toward elites and uppity women and minorities will not only win a big majority of white voters, but will boost white turnout as well. Sean Trende’s famous “missing white voters” hypothesis for one of Mitt Romney’s fatal defects suggests that marginal white voters tend to be the kind of people who voted for Ross Perot in 1992 — a peak year in white turnout. Trende has subsequently confirmed that the same kind of white voters have been turning out for Trump in the GOP primaries. But are there enough of them to make a difference, particularly given Trump’s big problems with minority voters?
At FiveThirtyEight, David Wasserman takes a close look at this question and arrives at an ambivalent conclusion:
The good news for Trump is that nationally, there’s plenty of room for white turnout to improve. If non-Hispanic whites had turned out at the same rate in 2012 that they did in 1992, there would have been 8.8 million additional white voters — far more than Obama’s 5 million-vote margin of victory. But before Democrats panic, here’s the catch, and it’s a doozy for Trump: These “missing” white voters disproportionately live in states that won’t matter in a close presidential race.
Between 1992 and 2012, white turnout dropped from 71 percent to 63 percent in the 38 non-Electoral College battleground states. There were huge double-digit declines in relatively Perot-friendly places such as Alaska, upstate New York and Utah. But in the 12 key battleground states, white turnout dropped more modestly, from 69 percent to 66 percent. There was virtually no white drop-off in Pennsylvania, and white turnout increased in New Hampshire and Virginia.
This makes sense if you think about it for a minute. If the “missing white voters” are basically marginal voters, they’re less likely to bother to vote in states where their votes “don’t count” in the sense of affecting the outcome. Meanwhile, the same voters are more likely to show up at the polls in highly competitive states where their votes do count, and where, moreover, they are the object of all the dark arts of base mobilization.
The bigger problem for Trump, as Wasserman notes, is that it’s by no means clear Trump’s going to win all the votes cast for Mitt Romney, much less add on many millions of marginal white voters in the right places. He’s got a real problem with college-educated white women, and in some polls isn’t doing all that well with college-educated white men. And that’s aside from the strong possibility that he’s going to do even worse than Romney with minority voters, who in any case are likely to continue to become a larger percentage of the electorate this November.
It all goes to reinforce the most important single insight I can offer to those all caught up in slicing and dicing the electorate: a vote is a vote, and running up the score in one demographic doesn’t mean squat if it’s offset by losses in another, especially in battleground states. And it’s another sign that Trump’s angry non-college-educated white men just aren’t large enough of a group to win the election for him, particularly if turning them out requires the kind of over-the-top borderline-racist-and-sexist histrionics that tend to mobilize the opposition as well.