Despite the substantial efforts of the Democratic party, the opposition or indifference of a large majority of the Republican intellectual and political establishment, and despite the confident predictions of an army of political analysts (fom old-school cable talking heads to new-school “big data”-style forecasters like Nate Silver) Trump pulled off astonishing upset victory over Hillary Rodham Clinton.
So instead of looking to explain—and take credit for—the Hillary victory that most Estonians went to sleep expecting, all three groups—media, Democrats, and Republicans—are instead struggling to answer the simple question: why?
Despite the occasional warning that Brexit might serve as a cautionary tale, the vast majority of observers expected a Clinton victory. Even Republican columnists had already written elegies about how to recover from a dramatic Trump defeat in the days prior to the election; (perhaps most hubristically, defeated primary candidate John Kasich scheduled a press conference about the future of the Republican party…for Thursday; presumably that speech will no longer take place as intended.)
So the first reaction both in the US and around the world is “why couldn’t our experts predict this?” And for once—and unlike in the recent Estonian case—the Electoral College or other arcane features of the American election system can’t be blamed, given the size of Trump’s victory in the popular vote as well. The scale of the predictive failure shouldn’t be exaggerated—nearly all the national polls except for that of the LA Times got the result wrong, while Nate Silver’s FiveThirtyEight (which got the last two elections right) closed by predicting a 66% chance of a Clinton victory.
One compelling partial explanation that’s been suggested is an American manifestation of the so-called “Shy Tory” phenomenon in Britain, in which voters reluctant to tell pollsters about their support for controversial parties abandon that bashfulness when in the privacy of the polling booth.
Another, of course, is in the turnout predictions for key constituencies in each party were substantially off—whether in overstating that of millennial or ethnic minority voters in the Democrats’ case, or understating that of rural white voters for the Republicans.
Indeed, when Democrats ask “why,” one answer will also lie in turnout—specifically, their failure to mobilize their base to the extent that they had done in previous campaigns. Despite a superior “ground game” of volunteers (and paid consultants) on the ground in key states relative to those of the GOP, the level of enthusiasm for Hillary—weakened partially by the unexpectedly bitter primary battle with Vermont Sen. Bernie Sanders—was far lower among African-Americans and young people than predicted. While these groups did not flock to Trump as feared by some, the inherent lack of appeal in what many saw as a “lesser of two evils” candidate—one dogged by continuing allegations from mishandling classified information to shifting policy positions to simple “untrustworthiness”--undoubtedly discouraged many from heading to polling stations. Given Trump’s extremely controversial statements about Mexicans and Mexican-Americans, one would have expected his numbers to drop substantially among Latinos; however, with the exception of Florida (a state whose Cuban-American demographic has been historically becoming less Republican over time as the divisions of the Castro era fade), Trump actually exceeded Mitt Romney’s share of the Hispanic-American vote in other battleground states, notably Nevada.