The Death (and Life) of Marriage in America

The story we think we know is that the institution of marriage is crumbling and on the brink of oblivion. The real story is much more complicated.

615 Wedding rings.jpg

National Marriage Week USA kicks off today, and for many people, a national booster movement for marriage could not come any sooner. The recession did a number on American matrimony, as you've surely heard. The collapse in marriage rates is cited as one of the most important symptoms -- or is it a cause? -- of economic malaise for the middle class. But the statistics aren't always what they seem, and the reasons behind marriage's so-called decline aren't all negative.

At first blush, the institution of marriage is crumbling. In 1960, 72% of all adults over 18 were married. By 2010, the number fell to 51%. You can fault the increase in divorces that peaked in the 1970s. Or you could just blame the twentysomethings. The share of married adults 18-29 plunged from from 59% in 1960 to 20% in 2010. Twenty percent!

What on earth is going on with these kids? Betsey Stevenson and Justin Wolfers tried to answer that question (among others) in their fantastic 2007 study "Marriage and Divorce: Changes and their Driving Forces."

The simplest summary of their findings is: It's really, really complicated. The full answer for the delay and decline of marriage would touch on birth control technology (which extends courtships by reducing the cost of waiting to get married), liberal divorce laws (which creates "churn" in the labor market by increasing divorces and new marriages), and even washing/drying machines (which both eliminate the need for men to marry lower-earning women to do housework and also free up women to work and study).

One important lesson from Stevenson and Wolfers is that, as much as it feels like things are changing very rapidly, a longer view on marriage trends reveals a more boring picture. If you pull back the lens, not to the 1960s but to the 1860s, the marriage rate and the divorce rate stick stubbornly to long-term trend lines.

Marriages-per-thousand people are declining, but slowly, after spiking in the 1940s. Divorces-per-thousand people are rising, but slowly, after spiking in the 1970s. Even in the Great Recession, which theoretically scared couples from investing in matrimony, we've seen "the same rate of decline that existed during the preceding economic boom, the previous bust and both the boom and the bust before that," Wolfers wrote.

Screen Shot 2012-02-06 at 1.12.38 PM.png

The median age of marriage for men and women is rising slowly into the high-20s. But that's not so unique compared to men's historical averages. What is new -- really, really new -- is the rising marriage ages for women.

The education revolution for women -- one of the happiest trends of the 20th century -- carries important implications for the marriage market. First, if women are going to college, more of their 18-22 years will be taken up by history classes rather than husbands. Second, when these women start earning money, fewer need to marry for financial reasons, which means they can afford to put off marriage. Thanks to birth control, a little bit of biotechnology has helped their cause by reducing the costs of being sexually active and single.


The economics of marriage suggest it's like any other investment. Women are more likely to get hitched when they see big potential gains from a union. That explains why the apparently monolithic Decline of Marriage is in fact two polar opposite trends. The first points toward the revitalization of marriage. The second points to decline.

First, for highly-educated or rich women, marriage rates are actually rising. It was once the case that a college degree was the equivalent of punching your spinster card. In the late 1800s, half of all college-educated women never married. But in the last 40 years, marriage rates have increased for the top 10 percent of female earners more than any other group, Michael Greenstone and Adam Looney found in a new report from The Hamilton Project. A 2004 American Community Survey also found college-educated women were 10 percentage points more likely to be currently married than women with less education.

Second, there is a parallel trend that really does look like the crumbling of marriage. Among less-educated and poorer women, marriage is in outright decline. The bottom half of female earners saw their marriage rates decline by 25 percentage points, Greenstone and Looney find and show in the graph above.

There is evidence that social and economic sorting in America creates clusters of people who match up by education, salary, and politics. As a result, perhaps poor or under-educated women are more likely to be matched with poor or less-educated men who offer a worse return on the marriage investment.

"Marriage has become the fault line dividing America's classes," Charles Murray writes in his newly book on the state of white America, Coming Apart. In a Wall Street Journal column based on the book, Murray compared those living in the country's educated white-collar world ("Belmont") and less-educated blue-collar world  ("Fishtown"). The divergence between Belmont and Fishtown begins at the alter, he argued:

Screen Shot 2012-02-06 at 6.20.01 PM.png
In 1960, extremely high proportions of whites in both Belmont and Fishtown were married--94% in Belmont and 84% in Fishtown. In the 1970s, those percentages declined about equally in both places. Then came the great divergence. In Belmont, marriage stabilized during the mid-1980s, standing at 83% in 2010. In Fishtown, however, marriage continued to slide; as of 2010, a minority (just 48%) were married. The gap in marriage between Belmont and Fishtown grew to 35 percentage points, from just 10.

The decline of marriage in Fishtown matters, Murray says, because we have an abiding national interest in seeing children raised in two-parent households. Murray might be overstating the importance of living with married parents, as opposed to couples in cohabitation (which is way up in the last 20 years). But if he's right, the following statistic should scare the heck out of you: In 1970, only 6% of births to undereducated "Fishtown" women were out of wedlock; by 2008, it had grown to a whopping 44%.


If Pride and Prejudice were written today, it might begin, "It is a truth universally acknowledged, that a single man in possession of a good fortune is practically an endangered species." In the last two years, this magazine has published two cover stories -- on the End of Men and the rise of the Single Lady -- that posited that "a marriage regime based on men's overwhelming economic dominance may be passing into extinction."

Marriage Extinction Watch is on high alert in the poorer segments of the country. In 2007, among women without a high-school diploma, just 43 percent were married. One reason was that the men they're mostly likely to marry are faring so poorly. The wages for all prime-age men have fallen by 40% in the last 40 years, after you account for the fifth of these men who have dropped out of the labor force. As the Hamilton Project reported last week, falling male earnings track closely to declining marriage rates. "At the bottom 25th percentile of earnings ... half of men are married, compared with 86 percent in 1970," Greenstone and Looney write.

We cannot know why millions of couples who might have been tying the knot 40 years ago aren't doing so today. Cohabitation is a factor. Divorce is a factor. But so is economics.

It might help to think of the marriage marketplace as, well, a marketplace. Historically, men and women have gone to the market to marry for the same reason that employers go the market to hire. They are looking to hook up with a productive "partner." As women are likely to seek partners in their income class, poorer women are more likely to be surrounded by men with low and falling fortunes, and more have chosen to forgo a union that could become a financial drain.


Screen Shot 2012-02-06 at 1.12.54 PM.png

If twentysomethings are responsible for inspiring "end of marriage" talk, let's praise seniors for keeping the institution alive by staying alive. As the graph above shows, today's Americans aged 60 and higher are as likely to be married as any other generation before them at that age. In fact, those over 65 are now more likely to be married as those in their 20s -- a moment that is unique in history. Adults are living longer, having fewer children, divorcing at a higher rate, and finding new partners, Stevenson and Wolfers write.

The graph above is a deceptively simple picture that says a lot about how the institution of marriage has changed in the last 130 years. First, it shows how unusually early twentysomethings married around 1960. This suggests that comparisons to that generation imply an exaggerated collapse. Second it shows that, at every age up to 60, today's Americans are less likely to be hitched than any generation before them. Third, it suggests that seniors are marrying close to their age. The shrinking gap between the ages of husbands and wives that helps to explain why couples are more likely to sort within their income group. Finally, it implies that even with rising divorces, the market for re-marriages is strong.

If somebody tells you nobody your age is getting married anymore, the optimistic rejoinder is: Just wait. Marriage isn't dead. It's just changed, and running a few years behind schedule.