For all the attention to the gender gap in politics, the lesser-noticed marriage gap is bigger.
While Donald Trump did better among male voters than female ones by 11 points, his support among married voters was 15 points higher than it was with unmarried voters. Married America voted for Trump by a solid margin (52 to 44 percent), while single America voted for Hillary Clinton in a landslide (55 to 37 percent).
These patterns have held for every presidential election since 1984, when exit pollsters started differentiating between married and unmarried voters. In 2008, for example, John McCain scored five points better among men than among women, and 19 points higher among the married than among the unmarried.
The causes of this marriage gap are probably both economic and social. Married couples are both more affluent and more religious than singles are. And those associations themselves have complex causes. Economic success makes people more marriageable, marriage seems to make men more economically successful, and older people are more likely to be both married and financially comfortable.
New research from family scholars W. Bradford Wilcox and Jon McEwan sheds more light on the relationship between marriage and voting — and on how Trump seems to have changed that relationship.
Wilcox and McEwan note, first, that the Republican advantage among married people extended to the county level: The higher a county’s ratio of two-parent to single-parent families, the more heavily it voted for Trump.
But the researchers also point out that the married and the unmarried voted more similarly in 2016 than they did in 2012. The marriage gap fell by six points. That’s the largest shift we have seen and the only decline. The marriage gap had steadily widened from 1984 (when married people favored Ronald Reagan by 10 points more than unmarried people) until 2012 (when Mitt Romney did 21 points better among the married).
The county data they examine tells the same story. Swing counties, the ones that flipped from supporting Barack Obama in 2012 to supporting Trump in 2016, had a higher proportion of single-parent families than the solidly Republican counties that supported Romney in 2012.
Wilcox and McEwan suggest that the decline of the married two-parent family in those counties may have made voters who live there more attracted to Trump. It may have reduced their sense of both economic and social security, and made them feel that what they value about America is slipping away.
The data suggest another possibility to this untrained eye. The marriage gap expanded as the parties became more ideologically polarized: as Republicans became more identified with economic and social conservatism, and Democrats with economic and social progressivism.
That probably wasn’t a coincidence. In the 2016 election, Trump dimmed those associations. His biography ran counter to the Republican Party’s social conservatism, and his platform to its economic conservatism. Voters, both married and single, may have responded to these cues. If you were a single, irreligious and economically downscale voter, you might have seen him as a fighter for you, in part because you didn’t think of him as a fighter for the religious right.
For several years before last month’s election, there was a lot of talk about the Republican Party’s demographic decline: It was doing better and better among whites, especially whites without college degrees, but they were a shrinking share of the population. Much the same could have been said about marriage: The percentage of Americans who are married is also declining. The preference of single Americans for the Democrats thus appeared to be a long-term threat to the Republican Party.
Since Trump won, fewer people have been talking about the inevitable extinction of the Republican Party on racial grounds. Or, at the very least, the date of that extinction has been pushed into the future. Trump’s modest gains among unmarried voters may provide an additional reason for Republican optimism.