Fodder for the Blame Game

by william h. frey

bill frey is a senior fellow at both the Milken Institute and the Brookings Institution, and author of Diversity Explosion: How New Racial Demographics Are Remaking America.

Published July 15, 2016.


Barring the appearance of a black swan, November's presidential election will pit the first major-party woman candidate against an opponent whose core backers are men. The electoral gender gap is thus likely to be bigger than ever before – but to whose advantage?

Women have been more partial than men to Democratic presidential candidates since 1980. In 2012, this proved decisive when 55 percent of women voted for Barack Obama, while 52 percent of men voted for Mitt Romney. But the gender divide will likely be greater this time around, as more women – and perhaps more men – cast ballots.

Actually, the electoral calculus is more nuanced. White women, as a group, are more likely to vote Republican, while racial minorities – both men and women – heavily favor Democrats. It's thus useful to look at segments of the white population where Democrats could fill out a majority.

A key distinction here is between unmarried and married white women. In 2012, the former – especially those with college degrees – voted Democratic, while white married women – especially those without diplomas – voted for the GOP.

Hence Clinton's success seems to hinge on two factors. First, she must raise the turnout of white, unmarried women. Their turnout rate has in the past been dwarfed by their married counterparts (58 percent to 72 percent in 2012).

Second, she must widen the preference gap between white, married women and their husbands. White married women already vote less strongly Republican than white married men, but this edge could be stretched by a campaign in which gender issues come to the fore.

Not only were women responsible for Obama's majority in the 2012 popular vote, they also played a big role in 11 of the 12 swing states he carried (the exception: North Carolina). This time, though, heavy turnout from Trump-inspired white men could reverse that dynamic. In fact, if both the turnout among all white males and their preference for a Republican candidate increased by 4 percent over 2012 (and nothing else changed), Trump would win in Ohio, Florida, Pennsylvania, Iowa and Colorado – enough to make the difference in the Electoral College.

But, of course, other things may well change. If white women's turnout and voting also shifted by 4 percent in the opposite direction, Trump would win only one of the five (Pennsylvania). For better or worse, we live in very interesting times.

charticle image correction

main topic: Demographics
related topics: Gender, U.S. Politics