Marital choices are exacerbating household income inequality

“HERE’S what nobody is telling you: Find a husband on campus before you graduate,” wrote Susan Patton, a human-resources consultant, in 2013. In an infamous letter to the editor of Princeton’s student newspaper, Ms Patton warned female students at the university that they will “never again be surrounded by this concentration of men who are worthy of [them]”. Critics responded harshly. Ms Patton recalls that she was branded “a traitor to feminism, a traitor to co-education and an elitist”.

Economists might offer yet another critique of Ms Patton’s letter: it was largely unnecessary. It is clear to academics that people tend to marry spouses with similar levels of education. They also know that “assortative mating”, as the practice is called in the jargon, is exacerbating income inequality. In America, Britain, Denmark, Germany and Norway, they have found that household income would be more evenly spread if couples were less keen to marry similar sorts.

Less clear, however, is whether or not assortative mating is on the rise. Answering this question is hard, because the ratio of educated men to educated women has shifted over time. That university-educated men are now more likely to marry university-educated women may not show a change in spousal preferences. It may simply reflect the increased number of women with degrees.

The latest iteration of the debate comes in a recent paper by Pierre-André Chiappori and Bernard Salanié, of Columbia University, and Yoram Weiss, of Tel Aviv University, which argues that assortative mating is indeed growing. They note that in the mid-20th century households were primarily concerned with divvying up chores. Since then, inventions like the washing machine and frozen food have meant that people can spend less time on housework. At the same time, computers have increased the demand for skilled labour. The authors argue that parents worried about their children’s future now have to focus on raising the brightest youngsters possible, and one of the surest ways to have bright children is to marry a bright spouse. By building an economic model which takes into account these shifting preferences and testing it against census data, the authors conclude that Americans born in 1972 do indeed have a stronger preference for better-educated partners than those born in 1943.

One implication of assortative mating is that most estimates of the returns on investment in a university education err on the low side, as they fail to take spouses’ earnings into account. Research from the Federal Reserve Bank of New York finds that the annualised return on investment for a four-year bachelor’s degree in America rose sharply between 1980 and 2000 but has since stabilised at around 15%. Our calculations show that, if a spouse’s income is added to a person’s own, the returns to higher education have increased steadily since 2000, now reaching 18%. Similar patterns hold for both men and women. Gender inequality in lecture halls has faded; household income inequality has widened.