I Do, but Later

With the exception of the three decades following World War II—including the 1950s era of the (“Leave It to Beaver”) Cleavers and the (“Adventures of Ozzie and Harriet”) Nelsons—Americans were never ones to rush into marriage. While in most cultures, women have typically married in their teens and men a few years later, people in the United States and other Anglo countries have been notable for their leisurely approach to settling down. In 1900, the median age of marriage for women in the United States was a little over 23 and for men, around 26. In the past several decades, however, twentysomethings have been pushing marriage into even later years, taking us into entirely new demographic territory.


Let’s look at some of the numbers. The age of marriage has been rising steadily since 1970
(see Figure 1) and, in fact, in 1980 women passed the previous historical high, a benchmark reached by men ten years later. Today in the United States, as Hollywood has happily discovered, a greater proportion of twentysomethings are unmarried than ever before (see Figure 2). Much of that increase can be explained by delayed marriage. Back in the day—say, 1970—over 60 percent of women aged twenty to twenty-four and almost all (90 percent) of women aged twenty-five to twenty-nine had married; in 2010, those numbers had plummeted to 20 percent and about 50 percent. Men followed a similarly dramatic pattern. In 1970, almost half of men aged twenty to twenty-four were married, and a remarkable 80 percent of those twenty-five to twenty-nine had also settled down. By 2010, those numbers had plunged to slightly more than 10 percent and less than 40 percent (Figure 3). Over the past forty years, then, marriage in the early- or midtwenties has been going along the same path as the standard-shift car—not exactly a relic, but increasingly outdated.



When we tease apart this trend by education, we notice something striking, especially in the past decade. During that time, the percentage of college-educated women twenty-five to twenty-nine years old who were still single went from 46 to 55 percent (Figure 4). An increase like this makes researchers sit up and take notice, but it is not quite a demographic earthquake, especially because college-educated women have traditionally been more likely to postpone marriage than have other women.


The story for women without a college degree, on the other hand, does reach the level of a demographic headline. Between 1990 and 2000, the percentage of never-married women in their late twenties in two groups—high-school dropouts and those with a high-school degree and maybe some college—rose modestly by about 5 percentage points. Starting in 2000, however, the percentage of still-single women in both groups jumped more than 15 points. As a result, the age of marriage for women of all education levels converged near the same historically high mark; today, more than 50 percent of all women in this age group are not married. Whereas in the past, women from Vassar to the University of North Carolina were always known for marrying later than their less-educated sisters, that is no longer the case. Women doctors, teachers, medical technicians, or waitresses are now all equally likely to postpone marriage to their late twenties. Women doctors, teachers, medical technicians, or waitresses are now all equally likely to postpone marriage to their late twenties.

Unsurprisingly, men of all classes have also become members of the delayed marriage movement.12 The marrying habits of those without a college education are looking much more similar to those with a degree than they did a decade ago. Today, across all educational levels, almost two-thirds of men aged twenty-five to twenty-nine are unmarried. But in a moment, we’ll see why this class convergence does not signify anything remotely like class solidarity.

Some might see marriage delayed as proof that young people, being especially open to change, think marriage is obsolete, or that being naturally rebellious, they don’t believe in the institution anymore. Not at all. The large majority of young adults say they hope to marry someday. The large majority of young adults say they hope to marry someday.True, in the final quarter of the twentieth century, the number of high-school seniors who believed they’d wait five or more years after high school to get married grew significantly.13 But, as Figure 5 indicates, about 80 percent of young-adult men and women continued to rate marriage as an “important” part of their life plans; almost half of them described it as “very important.” In fact, in 2001–2002, 30 percent of twenty-five-year-old women wished they were already married, on top of the 33 percent who were. For men, it was comparable—19 percent wished they were married; another 29 percent were (see Figure 6).



The younger generation seems to have understood that there are some very good reasons for postponing what they genuinely wish. We find three big advantages from delaying marriage. First, as Figure 7 indicates, later marriage tends to mean richer women, especially among the college educated, even after controlling for other factors. By the time they reach their midthirties, there is an $18,152 difference in annual personal income between college-educated women who marry before age twenty and those who wait until thirty or later. Moreover, college-educated women who delay marriage enjoy markedly higher combined (household) incomes, as Figure 8 shows. Much of the gain is from the greater career focus and delays in motherhood that generally accompany later marriage. Indeed, for many women, the delay of their marriage has helped them adapt to the job and career uncertainties of today’s economy.



Marriage delayed carries another big social and personal benefit: it’s cut down the divorce rate. Studies have consistently shown that couples who marry before age twenty-five are more likely to find themselves in divorce court.14 Our own research based on data from the National Fatherhood Initiative Marriage Survey supports this conclusion: women who marry in their early twenties and especially in their teens are significantly more likely to end up divorced than those who marry in their midtwenties or later (see Figure 12 on page 20). Some people conclude that this finding implies that the older a couple is when they marry, the less likely it is that they will split up. This is true, but only up to a point. As divorce insurance, marriage after the midtwenties has diminishing returns.As divorce insurance, marriage after the midtwenties has diminishing returns; a twenty-five-year-old bride is at not much greater risk of splitting up one day than is a thirty-five-year-old bride.15 Still, discouraging early (especially teen) marriage has helped to drive the divorce rate down from its record highs in the early 1980s. At that point, experts estimated that about half of all first marriages were ending in divorce. Since then, the rate has been declining, and experts now put the number at closer to 40 percent.16 In general, couples who wait till their midtwenties or later enjoy more maturity and financial security, both factors that make it easier to sustain a lifelong marriage.

12 Our analysis of Decennial Census Public Use Microdata Samples, 1999-2000, and American Community Survey Public Use Microdata Samples, 2010.
13 Our analysis of Monitoring the Future, 1976–2010.
14 See, for instance, Evelyn L. Lehrer and Yu Chen, “Delayed Entry into First Marriage: Further Evidence on the Becker-Landes-Michael Hypothesis” (Chicago: University of Illinois at Chicago Department of Economics, 2013); Dana Rotz, “Why Have Divorce Rates Fallen? The Role of Women’s Age at Marriage,” Social Science Research Network working paper (December 2011), http://dx.doi.org/10.2139/ssrn.1960017.
15 Estimates by Lehrer and Yu (“Delayed Entry into First Marriage”) suggest that approximately 15 percent of typical brides twenty-four to twenty-six divorce in the first five years of marriage, compared to about 11 percent of brides aged thirty-three and older. By contrast, they estimate that 31 percent of teenage brides will divorce within five years.
16 See Andrew J. Cherlin, “Demographic Trends in the United States: A Review of Research in the 2000s,” Journal of Marriage and Family 72 (June 2010): 403–419. Cherlin estimates that the divorce rate for first marriages is between 40 and 50 percent; Rotz (“Why Have Divorce Rates Fallen?”) suggests that the divorce rate for first marriages is now lower than 40 percent.