The most incisive guide to issues facing the American family today . . . An invaluable resource for anyone wishing to stay on the cutting edge of research on family trends.
-W. Bradford Wilcox
Associate Professor of Sociology, University of Virginia
It has been well documented that over the last few decades we have seen a substantial increase in the median age of marriage in the United States. In fact, we are currently at all-time historic highs in these trends. According to the Current Population Survey from the U.S. Census, the median age of marriage right now is nearly 29 years of age for men and 27 years of age for women. Given the trajectory of this trend over the last several years, we are quickly approaching a time in our culture where half of marriages will occur for individuals after the age of 30, quite different from what we have seen in previous generations.
It is generally assumed by many that this delay of marriage is a positive trend and that later marriage provides more time for maturation and preparation—especially in regards to economic or career readiness. For these reasons, delayed marriage is actively promoted by parents and others. However, some emerging evidence may challenge this assumption. Particularly when we consider patterns of family formation, family stability, and child well-being, there appear to be some significant tradeoffs associated with the delay of marriage. These trade-offs are significant enough, in fact, that the path many young adults are pursuing in an effort to be better prepared for lasting marriage and to have a successful family life is actually producing the opposite of what they intend.
Three Key Misunderstandings
Three key misunderstandings are leading to our societal miscalculation of the full implications of delayed marriage, particularly delayed marriage into the 30s. First, we tend to be interpreting this trend only through the experience of the college-educated. If we broaden out and look at the full social-economic spectrum, we start to see some very different patterns related to the delay of marriage. It is important to remember that those who are not college-educated make up the vast majority of our population. Unless we consider these less-educated segments of our society, we will never fully appreciate the consequences and impacts of delayed marriage.
The second misunderstanding, and perhaps the most important, is that the very term “delayed marriage” is an inappropriate name or designation for what is in fact happening. Our collective conversation about delayed marriage is missing the key point that what we are seeing is not “delayed marriage” as much as it is “re-sequenced marriage.” Yes, marriage is being delayed by many individuals; but sexual coupling and childbearing are not—at least not to the same degree. If marriage, sex, and childbearing were being delayed as a “whole package,” we would be seeing something very different. There is also a segment of society where we see a pattern of “marriage foregone”—not just delayed, but no longer a part of the life course.
The third misunderstanding is that we tend to fall into the belief that age equals maturation. During the teenage and childhood years, brain development and other biological processes of maturation are actively occurring. Thus, chronological and maturational age do tend to match up very well. But as people start to move into young adulthood, chronological age is no longer a guarantee of movement or progress in maturational age. The simple assumption that older is better, or that older means more prepared, actually starts to miss some of the most important factors that help people become ready for marriage and family life. We need to more fully develop a culture of marriage readiness that helps young people develop true “marital competence,” rather than arbitrarily selecting a chronological age and assuming that reaching that age makes one ready to be a spouse or parent.
Age of Marriage and Income Levels
Turning to some of the analyses my colleagues and I did for the Knot Yet Report, we can look at how age of marriage is related to income for men and women. We analyzed the American Community Survey (2008-2010) and divided the sample into three clusters or groups: (1) the less than high-school educated or the high-school dropout population, (2) the high-school graduate or community college population—what we sometimes refer to as the “Middle America” population, and (3) the college graduate and professional population (see Figure 1). We then divided those populations into age of marriage groups, including: (1) those who married under the age of 20, (2) those who married between 20 and 23, 3) those who married between 24 and 26, (4) those who married between 27 and 29, (5) those who married over the age of 30, and (6) those who never married. We compared those groups on their level of current personal incomes by the time they reached their mid-30s.
(Figure 1: Personal Income of 33-35 Year-Old Women, by Age at Marriage and Education, from The Knot Yet Report.)
Now we do see that when it comes to women’s personal income, there seems to be some direct advantages that occur, on average, with a delayed age of marriage. But where have the advantages primarily happened? Almost exclusively for college-educated women. If we look at the less than high school and the Middle America populations, we do not see nearly as much advantage or benefit coming from the delay of marriage—even though that is what is assumed will happen for everyone delaying marriage. And, even the advantages we see with the college-graduate population may come with some degree of tradeoff when we look at some of the other indirect factors of economic well-being. It should also be noted that the economic liabilities associated with earlier marriage for women are less pronounced if child-birth is delayed and educational trajectories are maintained. In fact, a further analysis of the matter suggests that the timing of childbirth is a better predictor of personal income levels for women in later life than is timing of marriage.
With men, however, we start to see a different pattern with age of marriage and personal income levels. As noted in Figure 2, we see a flat pattern in our lower-educated groups, but then among our college-educated population we see an interesting pattern emerge. While there is a gradual linear association between age at marriage and economic outcomes for women (particularly college-educated women), the association for men appears to be flatter and more curvilinear when we consider the never-marrieds. We actually see that men marrying in their mid to late 20s are earning just about as much as those who do not marry until their 30s. So even the direct association between age of marriage and personal income levels starts to tell a more complex story than we typically hear discussed in our social conversations on the matter. It appears that marriage may act as a motivator for men and encourages them to take greater advantage of their education and career opportunities.
(Figure 2: Personal Income of 33-35 Year-Old Men, by Age at Marriage and Education, from The Knot Yet Report.)
Age of Marriage and Family Formation Patterns
A fuller story, however, starts to emerge as we look at indirect measures of economic well-being and the benefits and harms of later marriage. Figure 3 depicts what my colleagues and I in the Knot Yet Report labeled “The Great Crossover.” This pattern becomes apparent when the median age of marriage is mapped on top of the median age of first birth in the United States. Approximately 20 to 25 years ago we experienced a significant demographic change in this country that has been rarely recognized and rarely talked about—at least not in any sustained cultural conversation. What we witnessed was the “crossing over” of the median age of marriage above the median age of first birth.
(Figure 3: The Great Crossover, from The Knot Yet Report.)
What this means is that the median age of birth in this country is now lower than the median age of marriage. This is the key piece of evidence that what is happening in our culture is not simply a delay of marriage, but rather a resequencing of marriage. In Figure 3, the bars at the bottom are the corresponding rates of non-marital childbirths in the country as a whole. We are rapidly approaching a tipping point in this country where more than half of children will be born to unwed mothers. As we recognize this association between age of marriage and non-marital childbirth, and the well-understood link between non-marital childbirth and family instability, and the equally well-understood link between family instability and child poverty, we get a clearer view of the impacts of this trend—particularly to the well-being of children.
Again, this phenomenon becomes even more significant as we look at different social-economic groups. Among women who are college graduates, the median age of marriage is about 27 years, and the median age of first childbirth is about 30 years. So the typical pattern is one where childbearing is delayed until after marriage—resulting in a 12% non-marital childbirth rate among college-educated women. This pattern is referred to as the “success sequence”—where education precedes or co-occurs with marriage and marriage precedes having children. Again, if this pattern were happening for everyone, we would be looking at a very different reality when it comes to family formation in America.
When we look at the less-than-high-school group, we see that the lines are completely reversed, and the difference is growing between them. That median age of first marriage is about 27 years of age, similar to their college-graduated counterparts, but the median age of first childbirth remains close to 20 years of age. This pattern leads to a non-marital childbirth rate of well over 80%.
The disheartening story extends when we look at the Middle America segment of America—the largest portion of our society, comprising roughly 60% of the population. Again, similar to their college-graduated counterparts, Middle America women have a median age of marriage of about 27 years; but their median age of first childbirth is now about 25 years. Between the 1970s and 1990s, the median age of first childbirth was slightly later than the age of marriage for this portion of our population. Now it has crossed over, and the number of Middle Americans following the success sequence has decreased—leading to a near 60% non-marital childbirth rate among this portion of the population.
What is clearly emerging when we examine age of marriage patterns in association with family formation patterns is that the disadvantages are occurring for the individuals and groups who can least afford these disadvantages. Personal income levels are not being substantially improved for less-educated Americans who delay marriage, but the risks of non-marital childbearing and single parenthood are being increased in ways that diminish the economic well-being of both children and adults.
Age of Marriage and Divorce
One of our longest-standing social beliefs, and the rationale behind many people’s encouragement for delayed marriage, is that later marriages are better marriages. Once again, older is better, right? Older individuals are consistently assumed to be more mature, better prepared, and more situated for the economic and social realities of marriage and family life.
When we look at the data from 20 years ago, that argument tended to make a bit more sense—but not nearly as late into the lifespan as people commonly believed. Figure 4 depicts data from the National Survey of Family Growth in 1995 showing the association between the age of marriage and the probability of divorce. From this analysis we see that as people got older, at least up until the early 20s, there was a significant association between age of marriage and divorce. Simply put, if one or both of the spouses was a teenager at the time of the marriage the risk of divorce was significantly higher than if both spouses were in their 20s. Because of this, we have a pattern of warning young people about the risks of marriage at such an early age. However, we often take this warning too far and forget to emphasize that the age benefit flattened out dramatically after the early 20s. The “older is better” pattern was only true to a point. But this was a useful conversation to be having when many people were marrying in their teenage years—a pattern that is not nearly as common today.
But look at what happens when we fast forward 20 years and look at the new data (see Figure 5). The Family Studies Institute released a study recently that analyzed more current data from the National Survey of Family Growth (2006-2010) that examines the trend patterns between age of marriage and the probability of divorce. What they found was that the risks patterns for divorce have changed significantly—particularly for people who marry after the age of 30. What was once largely a linear pattern of association now shows a clear curvilinear pattern, with couples marrying in their mid to late 20s having a significantly lower chance of divorce than couples marrying in their 30s and later. The “older is better” idea is significantly challenged when we look at the contemporary trends. And as previously discussed, nearly half of first marriages are now happening after the age of 30 in the United States. It is clearly time to shift our cultural conversation from the risks of teenage marriage and start to have a more thorough conversation about the potential risks of 30+ marriages.
Marriage Preparation Paradoxes
So what can be done about these issues, and how can we promote better marriage readiness among young people? The key lies in getting to the whys behind the whats in the patterns I have been describing. We must be able to explain, not simply describe, what is happening if we are going to make meaningful changes.
Part of the explanation for what has been happening with age of marriage patterns is that many young people today are preparing for marriage in ways that are actually producing the opposite of what they intend. I believe that what we have in our culture today is the emergence of “marriage preparation paradoxes.” Marriage preparation paradoxes are behaviors believed to increase one’s chances of marriage success, which actually, on average, diminish one’s chances of having a loving and lasting marriage. The key here is that these behaviors are not being engaged in as a departure from marriage or a giving up on marriage, but rather are being embraced by the rising generation because they are believed to strengthen marriages. For these patterns to change, the faulty logic that undergirds them must be exposed and corrected.
Perhaps the best prototype example of a marriage preparation paradox is the cohabitation paradox. The primary reason that young people, and their parents and families, give today for encouraging cohabitation prior to marriage is that it will be a “test drive.” In short, it is believed to be a way to lessen the risk and chance of a later divorce. In fact, many of our best and brightest minds in the social sciences back in the 1980s were claiming that we would see a huge reduction in the divorce rate because of the climb we were beginning to see in cohabitation. The belief was that cohabitation would act as a sort of Darwinian “survival of the fittest” mechanism that would weed out the weak relationships, and only the strong ones would ultimately survive into marriage—and thus divorce rates would decline. We now have 30 years of studies that have shown just the opposite: cohabitation before marriage has historically been associated with greater odds of divorce. And while some of the more recent studies have shown a possible weakening of this association, no study to date has ever shown cohabitation to have a protective factor against divorce.
Another example of a marriage preparation paradox is what I call the sowing wild oats paradox. I see examples of this paradox all the time in my research on young adults. Many young people, and their parents, refer to the young adult period of life as a time to sexually experiment—to have a variety of sexual experiences with a variety of people. The central logic behind this way of preparing for marriage is that young people need to do this to “get it out of their system” so they will be ready to “settle down” in marriage. A thorough examination of the data related to this paradox would require more time than I have available today, but suffice it to say that there is ample evidence that what is happening is the exact opposite. Instead of settling down, we see people getting worked up. Sexual experimentation before marriage does nothing to get such attitudes and behaviors out of your system; rather, it gets them into your system. Dozens of studies have shown that individuals with greater patterns of sexual promiscuity and more sexual partners actually have higher, not lower, chances of divorce when they marry. Again, a paradox—the logic does not work.
The sexual chemistry paradox is an extension of this way of thinking. The current dating culture often emphasizes that two people should test their “sexual chemistry” before committing to each other. This type of compatibility is frequently mentioned as an essential characteristic for people to seek out in romantic relationships, particularly ones that could lead to marriage. Couples who do not test their sexual chemistry prior to the commitments of exclusivity, engagement, and marriage are often seen as putting themselves at risk of getting into a relationship that will not satisfy them in the future—thus increasing their probability of later marital dissatisfaction and divorce.
Two recently published studies call this assumption into question. My colleagues and I published the first study a few years ago in the American Psychological Association’s Journal of Family Psychology. This study involved a national sample of 2,035 married individuals who participated in the popular online couple assessment survey called “RELATE.” We found that the longer a dating couple waits to have sex, the better their relationship is after marriage. In fact, couples who wait until marriage to have sex report higher relationship satisfaction (20% higher), better communication patterns (12% better), less consideration of divorce (22% lower), and better sexual quality (15% better) than those who started having sex early in their dating. For couples in between—those that became sexually involved later in their dating, but prior to marriage—the benefits were about half as strong.
These patterns were statistically significant even when controlling for a variety of other variables, such as respondents’ number of prior sexual partners, education levels, religiosity, and relationship length.
The second study, by Sharon Sassler and her colleagues at Cornell University, also found that rapid sexual involvement has adverse long-term implications for relationship quality. Using data from the Marital and Relationship Survey, which provides information on nearly 600 low- to moderate-income couples living with minor children, their study examined the tempo of sexual intimacy and subsequent relationship quality in a sample of married and cohabiting men and women. Their analyses also suggest that delaying sexual involvement is associated with higher relationship quality across several dimensions.
They discovered that the negative association between sexual timing and relationship quality is largely driven by a link between early sex and cohabitation. Specifically, sexual involvement early in a romantic relationship is associated with an increased likelihood of moving more quickly into living together, which in turn is associated with lower relationship quality. This finding supports Norval Glenn’s hypothesis that sexual involvement may lead to unhealthy emotional entanglements that make ending a bad relationship difficult. As Sassler and her colleagues concluded,
Adequate time is required for romantic relationships to develop in a healthy way. In contrast, relationships that move too quickly, without adequate discussion of the goals and long-term desires of each partner, may be insufficiently committed and therefore result in relationship distress, especially if one partner is more committed than the other.
Again, the research shows a pattern of sexual restraint, wherein commitment precedes sex, creates the pattern that really lowers the risk of relationship dissolution.
All of this can be tied together into what I call the older is better paradox. Too many of our young people today are growing up with the view that marriage is a transition of loss, rather than a transition of gain. A number of years ago, I worked as a visiting scholar for the National Healthy Marriage Resource Center. We conducted focus groups all across the country; and in these focus groups the young twenty-somethings talked about what marriage would ultimately “take from them,” “what they would lose,” “what they would ultimately have to give up,” and “what they would have to stop doing.” Rather than the historical pattern we have seen where individuals, and society as a whole, view marriage as a transition of gain. That it is something that adds to our lives, allows us to start doing meaningful things, and gives us a better and richer life. This line of thinking is paradoxical as well, given that numerous studies have shown that getting married and staying married are linked to several aspects of individual health and well-being, such as better financial status, improved physical health, enhanced mental health, and higher sexual satisfaction. Therefore, as marriage is delayed in order to avoid the perceived losses associated with it, many young adults begin to miss out on these known benefits of marriage—creating once again a paradoxical outcome.
Fostering a Culture of True Marriage Readiness
How might we counter these marriage preparation paradoxes and foster a culture of true marriage readiness?
First, we must help to identify the optimal window of opportunity for forging enduring marriages. While the risks of teenage marriage have long been understood, the possibility of risks associated with 30+ marriages are just beginning to be understood. There is a need for more attention to later marriages as the national median age of marriage continues to increase. We need to find ways to help young people appreciate the curvilinear nature of outcomes associated with the age of marriage; this will help counter the risks of early marriage, but not unintentionally replace such risks with the newly identified risks associated with later marriage.
Second, we should make young people aware of the realities of dating and spouse availability. It is highly likely that some of the benefits of marriage at later ages are offset by less than ideal matching due to a diminishing dating pool. Too many parents and others convey the mistaken idea to their young adult children that marriage readiness and spouse selection is simply a matter of personal preference and preparation. The dating pool is dynamic and shifts across the life course, making high-quality matches with marriage- and family-centered people less likely later in life.
Third, we must emphasize the strength of twenty-something marriages. A notable finding across many datasets and dozens of studies is how well so many marriages started in the early to mid-20s are doing. This is particularly true when educational trajectories are maintained, as the benefits of college education occur whether the degree is obtained before or after marriage. Rather than simply becoming overly concerned about later marriages, the data suggests we should be more open to and supportive of earlier twenty-something marriages.
Fourth, we should be aware of young people’s desire for marriages. While some might see the delay of marriage as proof that young people think marriage is obsolete, the evidence does not support that conclusion. In the Knot Yet Report, we used the National Longitudinal Study of Adolescent Health to document that, in fact, by age 25 nearly two-thirds of women are either married (33%) or wish they were married (30%); and nearly half of men by age 25 are either married (29%) or wish they were married (19%). These figures should remind us that while age of marriage is associated with desired timing of marriage, it is not always a factor of choice. Many young adults are frustrated by the erosion of courtship in our culture and the difficultly they are experiencing in dating and getting married.
Fifth, we must give attention to factors other than age that impact marriage. After the teenage years, studies have shown that age of marriage is associated with marital outcomes, but it is not a particularly strong predictor of marital satisfaction or divorce proneness. We would do better to promote a greater understanding of the individual and couple factors that are strong predictors of marital quality and encourage young adults to pursue high-quality relationships when possible, rather than waiting for an arbitrarily selected age of marriage. Over 80 years of research on premarital predictors of marriage outcomes have shown that true marital competence or readiness involves helping young people develop the capacity to love and the capacity to communicate. Thus, the foundation factors of personal maturity, emotional readiness, commitment, forgiveness, religious devotion, sexual restraint, communication skills, and the management of conflict are far stronger predictors of marriage trajectories than is age at marriage. We should also stress the “success sequence” of family formation, which involves gaining maturity and education prior to, or alongside of, marriage and marriage prior to childbearing. It is time for the college-educated segment of our culture to start preaching what they practice when it comes to family formation patterns.
The Least Advantaged Among Us
Finally, I would like to reemphasize that for many individuals in our society, the so-called delay of marriage is actually transitioning to patterns of “marriage forgone” or “marriage lost.” According to the National Survey of Family Growth (2006-2008), 20% of men and 14% of women ages 40 to 44 report that they have never been married. This is a significant number, because over the years we have seen that the never-married rate after age 40 is very close to the lifetime never-married rate. Again, these trends are exaggerated among the less educated segments of our society. The never-married rate among high school drop-outs is nearly one in three of men and one in four of women; and among Middle Americans it is slightly lower, at one in four of men and one in five of women. When we get to the point that such a large portion of our culture is not experiencing marriage across the life course, that is a dramatic change.
Remember that it is those who are least advantaged who are being most impacted by these changes. We have to keep a view across the full range of our society to best understand and address these patterns. In fact, the decline in marital child-bearing and two-parent families is one of the primary drivers of the economic inequalities we see today. A recent analysis from the National Center for Family and Marriage Research observed,
Evidence supporting the theme of diverging destinies, defined as growing racial/ethnic and social class differentials in family behavior, often has focused on the disproportionate rise in non-marital fertility among the most disadvantaged individuals versus the quite stable, low levels of unwed childbearing among college graduates. Similarly, there appears to be a divergence in marriage trends, as growth in non-marriage has been greatest among the most disadvantaged.
Similarly, Dr. Raj Chetty, the Bloomberg Professor of Economics at Harvard University, recently examined economic well-being and intergenerational upward mobility in the United States as part of “The Equality of Opportunity Project.” One of his primary conclusions is that family stability plays a key role in children’s outcomes, noting:
. . . mobility is significantly lower in areas with weaker family structures, as measured, for example, by the fraction of single parents. As with race, parents’ marital status does not matter purely through its effects at the individual level. Children of married parents also have higher rates of upward mobility in communities with fewer single parents. Interestingly, we find no correlation between racial shares and upward mobility once we control for the fraction of single parents in an area.
Many have argued that family stability plays a key role in children’s outcomes . . . To evaluate this hypothesis, we use three measures of family structure . . . (1) the fraction of children living in single-parent households, (2) the fraction of adults who are divorced, and (3) the fraction of adults who are married. All three of these measures are very highly correlated with upward mobility . . . The fraction of children living in single-parent households is the single strongest correlate of upward income mobility among all the variables we explored . . . 
These divergent destinies deserve our attention and should focus our efforts on promoting the marriage formation processes and family structures that undergird economic prosperity and upward mobility. Only when we start fostering these true foundations of economic readiness across the entire social-economic spectrum will we be preparing the rising generation for the economic realities they will face.
Dr. Jason S. Carroll is a Professor of Marriage and Family Studies in the School of Family Life at Brigham Young University, and a Fellow at the Wheatley Institution. This essay is adapted from a presentation given at World Congress of Families IX, October 27-30, 2015, in Salt Lake City, Utah.
 K. Hymowitz, J.S. Carroll, W.B. Wilcox, K. Kaye, Knot Yet: The Benefits and Costs of Delayed Marriage in America (2013), a commissioned report sponsored by the National Campaign to Prevent Teen and Unplanned Pregnancy, the RELATE Institute, and the National Marriage Project, p. 12, available at www.twentysomethingmarriage.org.
 J.S. Carroll, S. Badger, and C. Yang, “The Ability to Negotiate or the Ability to Love?: Evaluating the Developmental Domains of Marital Competence,” Journal of Family Issues 27 (2006): 1,001-32.
 Hymowtiz et. al., 15-16.
 Ibid., 7.
 Ibid., 8.
 Tim B. Heaton, “Factors Contributing to Increasing Marital Stability in the United States,” Journal of Family Issues 23 (2002): 392-409, doi: 10.1177/0192513X02023003004, at 407.
 N. Wolfinger, “Want to Avoid Divorce? Wait to Get Married, But Not Too Long,” Institute for Family Studies, July 16, 2015, available at http://family-studies.org/want-to-avoid-divorce-wait-to-get-married-but-not-too-long/.
 S.M. Stanley, G.K. Rhoades, and H.J. Markman, “Sliding vs. Deciding: Inertia and the Premarital Cohabitation Effect,” Family Relations 55.4 (October 2006): 499-509, doi: 10.1111/j.1741-3729.2006.00418.x.
 D.M. Busby, B.J. Willoughby, and J.S. Carroll, “Sowing Wild Oats: Valuable Experience or a Field Full of Weeds?” Personal Relationships 20 (2013): 706-18, doi: 10.1111/pere.12009.
 D.M. Busby, J.S. Carroll, and B.J. Willoughby, “Compatibility or Restraint?: The Effects of Sexual Timing on Marriage Relationships,” Journal of Family Psychology 24 (2010): 766-74, doi: 10.1037/a0021690.
 S. Sassler, F.R. Addo, and D.T. Lichter, “The Tempo of Sexual Activity and Later Relationship Quality,” Journal of Marriage and Family 74 (2012): 708-25, doi: 10.1111/j.1741-3737.2012.00996.x.
 N.D. Glenn, “A Plea for Greater Concern About the Quality of Marital Matching,” in A.J. Hawkins, L.D. Wardle, and D.O. Coolidge (eds.), Revitalizing the Institution of Marriage for the Twenty-First Century (Westport, CT: Praeger, 2002), 46-58.
 Sassler et. al., 710.
 Hymowitz et. al., 15.
 Carroll et. al., “The Ability to Negotiate or the Ability to Love?”
 Hymowitz et al., 14.
 W.D. Manning, S.L. Brown, and K.K. Payne, “Two Decades of Stability and Change in Age at First Union Formation,” Journal of Marriage and Family 76 (2014): 247-60, doi: 10.1111/jomf.12090.
 R. Chetty et. al., “Where is the Land of Opportunity? The Geography of Intergenerational Mobility in the United States,” Quarterly Journal of Economics 129.4 (2014): 1,552-1,623, at 1,558, doi: 10.1093/qje/qju022.
 Ibid., 1,616.