LF: Trump Isn’t Winning Enough White Voters

There’s a belief, which I don’t share, that the growing share of nonwhite voters in the population, particularly Latinos, is giving Democrats an enduring advantage in winning elections. The theory — known to some as the “Emerging Democratic Majority” — works only if voting patterns stay the same and Republicans don’t gain among either minority voters or white voters. Republicans could very well add to their share of white voters, which is why I think Donald Trump can win the presidential election this year: He could run up the score with white voters, even if nonwhite voters don’t like him.

It’s also why I found a new analysis by Nate Cohn of The New York Times so interesting. Cohn looked at census data, voter files and pre-election polls and found that “more white, older working-class voters went to the polls in 2012 than was found by exit polls on Election Day.” In other words, Trump might have a somewhat friendlier electorate to work with than we thought.

That’s good news for Trump. But here’s the bad news for his campaign: The evidence so far suggests Trump isn’t taking advantage of that possibility. Four years ago, using a similar method to Cohn’s, I argued that Mitt Romney could win with a lower percentage of whites in 2012 than many thought. Of course, Romney didn’t pull it off; just because something is possible doesn’t mean it will happen. And polls show Trump isn’t in a stronger position than Romney to pull in enough white voters to win.¹

Trump has trailed Hillary Clinton in every national poll for roughly the last three weeks. He’s led in only three of 34 polls since knocking Ted Cruz and John Kasich from the race in early May. In fact, the only two pollsters who had Trump ahead and have released a more recent poll (Fox News and Rasmussen Reports) now show him trailing by 3 and 4 percentage points, respectively.

One big reason Trump is trailing — by an average of 4 to 6 percentage points, depending on which aggregator you use — is because, despite all the bluster, he isn’t doing any better than Romney did among white voters. According to Cohn’s estimate, based on pre-election surveys, Romney beat President Obama by 17 percentage points among white voters. To win, Trump would need to improve on Romney’s margin by a minimum of 5 percentage points if the electorate looked exactly the same as it did in 2012 and every other racial group voted in the same manner as it did in 2012. Here’s a table showing Trump’s lead over Clinton in the seven live interview polls that broke out results by race and were conducted after Trump eliminated Cruz and Kasich:

Trump’s edge among white voters since securing the nomination
		SUPPORT WITH WHITE VOTERS
POLLSTER	START DATE	TRUMP	CLINTON	DIFFERENCE
CBS/NYT	May 13	50%	38%	+12
Fox	May 14	55	31	+24
NBC/WSJ	May 15	52	36	+16
ABC/Post	May 16	57	33	+24
Quinnipiac	May 24	50	33	+17
IBD/TIPP	May 31	48	36	+12
Fox	June 5	49	33	+16
Average				+17

Not surprisingly, Trump led Clinton overall in the two polls in which he had the biggest advantage with white voters, leading Clinton in that group by 24 percentage points in both. But, if you consider the other polls, Trump is winning white voters by an average of 17 percentage points, matching exactly Romney’s margin from four years ago. That’s not good enough, especially considering that the 2016 electorate will probably be more diverse than 2012’s. Trump probably needs to do even better than a 22-point lead among white voters, or he will have to pull in more minority voters than Romney did in order to win.

Trump could still improve his standing among white voters enough to take the election. There’s plenty of time.² And Trump is already doing far better than Romney among white voters without a college degree, as Cohn noted. But in politics, as in physics, every action has a reaction. As Amy Walter of the Cook Political Report has pointed out, Trump is also doing considerably worse than Romney among white voters with a college degree. That makes sense, given Trump’s direct appeals to whites who do not have a high level of education, and his penchant for shunning intellectuals.

To be fair, Trump hasn’t been dealt the strongest hand. The incumbent Democratic president is becoming increasingly popular, and the economy isn’t in a downward spiral when looking across different metrics. In other words, Trump has limited electoral capital to work with. If the economy becomes worse, the president becomes less popular or Clinton has another scandal, then Trump’s chances will improve. Until that happens, Trump is an underdog, and he’s not yet running up the score enough with white voters to win.

Footnotes

Note: Polls demographically weight their estimate of the electorate in a similar fashion to Cohn’s method. That is, there’s no reason to believe the polls are underestimating Trump. ^
Obviously, he could also collapse among white voters too. The chance of that happening would be improved if Clinton is able to rally mostly white Bernie Sanders supporters after he drops out. ^