michael-zuckertThe Rover editors have posed the question: ‘to what extent was Donald Trump’s victory a win for the conservative movement?” My  answer: it depends. It depends on what you mean by “victory,” by “the conservative movement,” and finally by “Donald Trump.”

To start with the easiest of the three: Trump surely won the election in the Electoral College. I don’t want to get into a debate about the Electoral College, an institution pre-election Trump thought was horrible, but which president-elect Trump thinks is “genius.” Whichever Trump one agrees with, it is still worth noting that Hillary Clinton won the popular vote by 2.2 million and counting. This is not to impugn Trump’s victory, but it is to question the almost consensus view that Trump’s was a resounding win.

In addition to Clinton’s majority in the popular vote, the Democrats picked up a modest number of seats in both House and Senate, not what normally happens in a blowout election. The Trump victory looks large because it went contrary to expectations. But expectations are not the standard by which to judge the magnitude of an election outcome. It is worth noting, for example, that in the most battleground of the battleground states where Trump did astonish by winning far more than anybody believed possible (Florida, Michigan, Ohio, Pennsylvania, and Wisconsin), and in the major battleground state Clinton won (Virginia), Trump won by a net total of about 50,000 votes, not a blow-out when those states together had a two party total vote of upwards of 30 million.

Another figure to consider: Trump ran ahead of Romney’s 2012 vote total by about 1.5 million, but well behind Obama in the 2012 election by over 3 million votes. If we look at what actually happened in the battleground states, we get a sharper idea of what occurred to produce Trump’s victory. In 2012, Obama won all six of the states we are looking at; in 2016, Trump won five, all but Virginia. The pattern of voting in the six varied a good deal. Perhaps the most significant finding is that in the five states Trump won, Clinton fell below the Obama vote in 2012, and in only one of the five cases was Trump’s margin of victory greater than Clinton’s fall- off from Obama. This held true whether the total two-party vote in the state was greater than in 2012 (Florida, Pennsylvania.) or less (Wisconsin, Ohio, Michigan.). What appears to have been decisive in these cases was the fall-off for Clinton—either to third party candidates, or to stay-at-homes or abstentions on the top of the ballot, or to Trump switchers (the famed white working class voter). Either way, it does not appear to be a victory of or by the conservative movement, because those who voted for Obama and then did not vote for Clinton are not likely to be members of “the conservative movement.”

The exit polls, admittedly not completely reliable, give still more reason to think this election was not a matter of a conservative movement victory, for the percentage of voters who identify as conservatives is just the same in 2016 as in 2012—and these voters split their votes in just about the same proportions as in 2012—with liberals going 84 percent for Clinton (versus 86 percent for Obama), conservatives going 81 percent for Trump (versus 82 percent for Romney) and moderates 52 percent for Clinton (versus 56 percent for Obama). So far as there was change, it was among the moderates, who clearly shifted away from Clinton.

If we are looking for evidence of where Clinton’s fall-off came from, we might be surprised at the answer: though both Clinton and Trump show only small overall fall-offs from their party predecessor, the Clinton fall-off among African-Americans was rather large, five percent, with a Trump pick-up of two percent over Romney—a seven percent shift toward Trump. Clinton suffered a somewhat larger fall-off among Hispanics, six percent, with Trump picking up two percent over Romney, an eight percent shift toward Trump.

One other finding from the exit polls is important to note. Voters were asked when they made up their minds on whom to vote for. Perhaps the most striking shift in all the polling data is in the number of those who decided for whom to vote within the week before the election: in 2016, 20 percent of the voters did that; in 2012, only nine percent decided so close to the election. In 2012, roughly 50.5 percent of late deciders went for Obama, with 44.5 percent going for Romney. In 2016, strikingly, 48 percent went for Trump and only 44 percent for Clinton, a shift of about 10 percent toward Trump among the much larger pool of late deciders. It is difficult not to suspect a major causal role in this shift attributable to FBI Director James Comey’s late breaking letter on Clinton emails. The amount of late decision making may also be responsible for much of the error in the pre-election polling.

In any case, it is again fairly clear that this election was not a victory by the conservative movement. It remains to be seen whether it will be a victory for the conservative movement. That will depend on which part of the movement we have in mind when judging winners and losers and which of the many vague and contradictory Trump policies the administration actually pursues.

Michael Zuckert is a Nancy R. Dreux Professor of Political Science and faculty advisor to the Irish Rover.