My wife and I are very different people. She’s from Texas; I’m from Manhattan. She likes Doctor Who; I like Star Trek. Her dad’s worked in ranching and the oil industry; my dad’s a novelist and screenwriter. Her family’s gatherings are WASP-y and polite; my family’s gatherings are loud and Jewish. But one thing about us is pretty similar: We both went to fancy colleges full of people with high SAT scores. And in that regard, we’re pretty typical.
It’s not that rare these days for people to marry outside their ethnic or religious group or to live in a different region of the country from where they grew up. Among college-educated people, in particular, the tendency is not so much to marry within your community as to marry within your educational cohort. Every once in a while you do see a college graduate married to a high school dropout (my parents, for example), but it’s quite rare. Since incomes are normally measured on the household level for statistical purposes, it matters quite a bit to big-picture national trends. In particular, “assortative mating” of this kind is a major driver of household-level income inequality.
How big? Jeremy Greenwood, Nezih Guner, Georgi Kocharkov, and Cezar Santos report in a new paper that if in 2005 the matching of husbands and wives had been random, the Gini coefficient—the most common summary measure of income inequality—would have been 0.34 rather than its real-world 0.43, a difference of almost 25 percent.
Of course, saying that inequality would be lower if marriages were randomized isn’t telling us much. It’s long been the case that a college-educated woman’s husband is much more likely to have a college degree than would be suggested by pure chance. But a number of closely related demographic factors have shifted dramatically over the past generation or two in a way that’s made assortative mating a much more potent economic force. And the nexus between marriage patterns and those other trends is an important motor of inequality.
If you go back to 1960, you’ll see that even though college-college married couples were disproportionate, they were also extremely rare—just 4.5 percent of all marriages. There just weren’t that many college graduates around back then. Eighty-two percent of married women hadn’t gone to college; 42.5 percent of married women hadn’t finished high school. Less than 40 percent of women, whether married or not, were in the labor force. And of course, those women who were in the labor force faced massive structural obstacles to obtaining high-paying jobs. These days there are more college-educated men than there used to be, many more college-educated women, and those women are more likely to be in the workforce. Aside from these family dynamics, the earnings gap between college graduates and less-educated workers grew enormously in the 1980s before leveling off in more recent years.
The combination of these trends has driven much more income inequality. The college-educated man of 2005 earns a larger premium over his working-class neighbor than was the case in 1960. He’s also much more likely to have a college-educated wife than he was in 1960; she is more likely to be working and is more likely to benefit from a college earnings gap in her favor.
On a public policy level, there’s no reasonable way to discourage these trends. The trend toward assortative mating isn’t amenable to “fixing” through the policy process, and the trend toward better educational and career opportunities for women is a change for the better. To the extent that the rise in income inequality is sometimes framed as a kind of nefarious conspiracy, the role of assortative mating will set minds at ease. In reality, a healthy chunk of the runup in inequality is a kind of side effect of benign or even positive social changes.
But there’s no need to take a blasé view of the trends, either. Not only is the prosperous man of the early 21st century much more likely to be married to a women who also earns a substantial income than was his predecessor of 50 years ago, but he also pays a lower income tax rate. Many of the drivers of inequality aren’t public policy choices or anything that we would want to reverse. But the sensible role of policy is to lean against possible negative side effects. During the past generation, we’ve mostly done the opposite.