Lane Kenworthy, The Good Society
May 2018

Historically, marriage has been as much a product of economic necessity and social norms as of love and friendship.¹ As societies get richer, economic circumstances and norms tend to change in ways that are likely to reduce the prevalence of marriage.

One such change is that women become better educated, more likely to be employed, and more likely to earn enough to live independently. Figure 1 shows the strong correlation between affluence and women’s educational attainment across countries. In addition, government benefits allow women with limited labor market prospects to survive without dependence on a husband. So for an increasing share of women, getting married, or remaining in an unsatisfying marriage, becomes a choice rather than a financial necessity.

A second change that comes with growing affluence is that people attach greater importance to individual freedom and choice. Most of us want the freedom to choose what to believe, how to behave, with whom to live, and so on. As material concerns become less pressing, this desire for freedom comes to the fore.² As a result, norms discouraging divorce, nonmarital cohabitation, and out-of-wedlock childbearing begin to weaken. Figure 2 shows, for instance, that in richer nations people are more likely to see divorce as justifiable.

It’s no surprise, then, that over the past half century the share of American adults who are married has declined. This is a potential problem for both adults and for children. For adults, having two rather than one in the household enhances economic security and boosts the likelihood of income gains. In the past generation median income has increased only for households and families with two earners. For those with a single earner it has been flat.³ And the risk that unemployment, sickness, or disability will result in significant income decline is much greater among households with only a single adult.⁴

It’s an even bigger problem for children. We have substantial evidence, first marshaled by Sara McLanahan and Gary Sandefur in the mid-1990s and steadily buttressed since then, that children who grow up with both of their original parents tend to fare better on a range of outcomes, from school completion and performance to crime to earnings and income to maintaining lasting relationships. This advantage holds compared to children whose parents never married, who married and then divorced, or who married, divorced, and remarried.⁵

How much have things changed? Is family decline more pronounced among particular groups? What can we do?

Skip to:

• Faltering families
• The US in cross-national perspective
• The class divide
• The future of families
• Summary

FALTERING FAMILIES

The share of American children not living with both biological parents and the share living in a single-parent household have each increased by 15 to 20 percentage points over the past half century, as figure 3 shows.

This is partly due to more married parents splitting up. As figure 4 shows, the divorce rate is a good bit higher now than half a century ago, though it has declined somewhat in recent decades. More important is the increase in having children without being married. Figure 5 shows that the share of kids born to unmarried parents rose from 5% in 1960 to 40% in 2007.

Why have out-of-wedlock births increased? It isn’t due to earlier sex or reduced use of contraception; teen births have declined since 1970 (see below). Instead, what’s changed is that fewer women are marrying and those who do marry are waiting longer to do so, as shown in figures 6 and 7.

THE US IN CROSS-NATIONAL PERSPECTIVE

In some rich democratic countries, an increasing share of couples who intend to stay together for the long-term have been substituting cohabitation for marriage. Comparing marriage patterns is therefore potentially misleading. Figure 8 shows a more useful measure: the share of 30-to-49-year-olds who are neither married nor cohabiting. The United States is toward the high end, but it doesn’t stand out.

The story for out-of-wedlock births is similar. While births to unmarried parents have risen sharply in the United States, this development isn’t exceptional. As figure 9 shows, the US is in the middle of the pack among the rich democratic countries both in the magnitude of its increase over time and in its current out-of-wedlock birth rate.

The US does stand out, however, when it comes to the share of children who don’t live with two parents, as we see in figure 10.

How can it be that America has such a large portion of its children living with just one parent when its out-of-wedlock birth rate is comparatively moderate? A key part of the answer is the shift in other countries from marriage to cohabitation, which increases the out-of-wedlock birth rate but doesn’t increase the share of kids in one-parent homes. Cohabitation has become more common in the United States, too, but American cohabitors are more likely to split up than their counterparts elsewhere. For example, a Swedish child born to cohabiting parents in the early 1990s had about a 60% chance that her parents would still be together fifteen years later. Her American counterpart had about a 20% chance.⁶

THE CLASS DIVIDE

The decline of marriage and of two-parent child rearing in America spans the socio-economic spectrum, but it is much less pronounced among those with a four-year college degree, as figures 11, 12, and 13 show.

Better-educated women now place considerable emphasis on a career, so they delay not only marriage but also childbearing, as figure 14 shows. This gives them time to get established and to find the right partner. As a result, among Americans with a college degree or better the decline in family has been muted. They are less likely to marry or stay married than their counterparts of half a century ago, and they are less likely, whether married or not, to remain together throughout their kids’ childhood. But these changes have been minor.

Among less-educated women, in contrast, age at first childbirth hasn’t increased much. Because this gives them less time to mature personally and to find a partner with whom they are compatible, because the financial prospects of their partners tend to be weaker, because their partners more often have a preference for traditional gender roles, and because the presence of a child can heighten financial and interpersonal tensions, women with less education are less likely than their better-educated counterparts to stay with their child’s biological father. For these reasons, the decline in marriage, in happiness among those who are married, in sustained cohabitation, and in both-biological-parent child rearing is much sharper among the less educated.⁷

THE FUTURE OF FAMILIES

Will the traditional family unit — two adults in a longstanding relationship, with one or more children — survive? Could it thrive?

Scenario 1: Continued family decline

One possible scenario is continuation of the pattern of family decline that we’ve experienced in the past half century. The most relevant indicator is the share of children who grow up with both of their original parents. As figure 3 (above) shows, that fell from 75% in the 1960s to 55% in the 2000s.

If we turn to figure 13 (above), we see that the rise in the share of children who don’t grow up with both original parents has occurred mainly among those with less than a four-year college degree. Among the college-educated, this trend all but stopped after the 1970s. A variant of the decline scenario is continuation of this pattern — family decline among the less-educated coupled with stasis among those who have more education.

Then again, it’s possible that we’ve reached the end of the long decline of families. The divorce rate peaked in 1980 and has fallen since, and the share of children born out of wedlock has been flat since 2007 (figures 4 and 5). As a result, the share of children who grow up without both biological parents and the share living with one parent have been flat over the past decade (figure 3).

Scenario 2: Family revitalization

A second possibility is a revival of the family. How might this come about? What institutions and policies could facilitate it?

In the previous heyday of the stable family, the middle of the twentieth century, its most common form consisted of an employed male and a homemaker female, with an authority structure favoring the husband. This norm will no longer work for very many couples. Family revitalization hinges on shifting to institutions and policies that conform to the new reality of women’s equal (or superior) educational attainment, desire for employment, and expectation of sharing in housework and childcare.⁸

Figure 15 shows that among the world’s rich longstanding-democratic nations, some do quite well at ensuring that most children grow up in a two-parent home. However, a number of those countries — Italy, Spain, Portugal, Germany, Austria, and Japan — have accomplished this by holding onto the old male-breadwinner norm. One cost, as indicated by these countries’ position on the graph’s vertical axis, has been a very low fertility rate. Women in these countries have been less willing to have children.⁹

Most of the countries that have succeeded in achieving both family stability and reasonably high fertility — including France, Sweden, Denmark, Finland, and the Netherlands — are ones in which family norms have shifted in the direction of equality and sharing and in which supportive public policies such as childcare and paid parental leave have been implemented.¹⁰

What does this imply for the United States? What, specifically, could move us toward the family revitalization scenario?

In one view, the key to revitalizing family is to revitalize marriage.¹¹ While this may be a worthwhile aim, existing evidence suggests that the impact of proposed efforts is likely to be small.¹²

Some recommend restructuring government taxes and benefits to more strongly favor marriage, with a special focus on rewarding marriage among couples with low incomes. But improving the financial incentive for marriage was a key aim of the mid-1990s welfare reform, and it hasn’t worked. In fact, AFDC-TANF benefit levels have been falling steadily since the 1970s, with no perceptible effect on the decline of marriage among less-educated Americans.¹³

Another option is intensive marital counseling sessions and support services for vulnerable couples. The best evidence here comes from the Healthy Marriages Initiative, which ran from 2005 to 2010. A thorough review of the program found no noteworthy beneficial impact.¹⁴

A third marriage revitalization idea is an advertising and messaging campaign aimed at shifting the culture, perhaps coupled with enhanced dissemination of information. Similar campaigns helped reduce smoking and teen births. However, marriage is different. We have strong evidence that smoking and teen parenthood tend to be bad for the decision maker. The evidence is less overwhelming that avoiding or exiting marriage is bad for the person doing so. It tends to be harmful for children, but it may or may not be bad for the adults. Also, smoking is an activity, whereas marriage is a relationship. A partner or spouse who treats you poorly or cheats on you may be less tolerable than a daily routine without cigarettes.

Focusing on marriage, however well-intentioned, may be misplaced.¹⁵ Getting more low-income couples in their teens or early twenties who find themselves pregnant to decide to marry is unlikely to produce many lasting relationships. The shotgun wedding approach worked half a century ago because marriage was a financial necessity for many women and because women tended to have limited expectation of emotional fulfillment or shared decision making in a relationship. This has changed.¹⁶ If more couples in that position were to get married these days, there’s a good chance many of them would end up divorced.

An alternative approach to family renewal is to try to get more women with limited education to delay childbirth until their mid-to-late twenties, when they are in a better position financially and have had time to find a desirable partner.¹⁷ One way to achieve that is via increased educational attainment. At the moment, many women with little education view their prospects in work as so dim that they are eager to move quickly to what they perceive as the other key source of fulfillment in life: having a child.¹⁸ More schooling will improve their opportunity for a decent-paying, fulfilling job and thereby heighten the influence of employment in their calculations about when to have children.

Greater availability of stable jobs along with higher wages would help, particularly if supplemented by a more generous Earned Income Tax Credit for those without children. This too would enhance the attractiveness of employment for women with limited education, and it would make more less-educated men attractive as long-term partners or husbands.¹⁹

Another strategy for delaying childbirth is more direct. We could achieve some reduction in the incidence of unplanned pregnancy via greater use of long-acting, reversible contraceptives (LARCs) such as intrauterine devices (IUDs) and implants. As Isabel Sawhill explains, these types of contraceptives “change the default from getting pregnant (if you do little or nothing to prevent it) to not getting pregnant until you take deliberate steps to to do so. The beauty of LARCs is that they do not depend on someone making a rational decision in an irrational state. They do not require remembering to take a pill or to get a prescription refilled. They are forgiving of well-documented human frailties (like a desire for sex). As a result, LARCs have been found, in practice, to be about forty times more effective than condoms and twenty times more effective than the pill at reducing the incidence of unplanned pregnancies.”²⁰

Is there reason to hope that any of this would work? One encouraging trend is the decline in teen births in recent decades, shown in figure 16. We don’t know exactly why this has happened, but it suggests that progress is possible.²¹

Finally, convincing a larger share of less-educated Americans to embrace gender-egalitarian values might help.²² Women are already moving in this direction, so many less-educated men may have little choice if they wish to have a long-term family. Gender-egalitarian values may have an additional advantage in promoting lasting relationships. Partnerships based on traditional gender roles can work, but the expectations about contributions are fairly rigid, so there’s a greater chance the harmony will be disrupted if, say, the man struggles in the labor market or the woman decides she wants a career. A partnership based on gender-egalitarian expectations is likely to be more flexible — and hence better suited to adapt to the many unexpected changes and challenges life throws at us.

Scenario 3: Substitutes for family

If we don’t succeed in revitalizing family, that doesn’t necessarily mean half of America’s kids are doomed. For one thing, it’s conceivable that other stable units, such as extended groups of adults, might replace the traditional family and do a decent job at ensuring income stability for adults and intellectual and social development for children.²³

Or public policy might take on some of the functions we used to rely on families to carry out. High-quality affordable early education — childcare and preschool — is likely to be especially helpful for families with a single adult or limited labor market capability. Early education allows parents to be employed and very likely helps to reduce gaps in the skills and capabilities of children from different family backgrounds.²⁴

An alternative to early education is a cash grant or tax benefit that would allow a family to either pay for good-quality childcare and preschool or keep one parent at home.²⁵ This is what Finland does. Norway used to have a home-care allowance as well, and since 2008 some municipalities in Sweden also have offered one. This approach offers parents more choice, but it has two drawbacks. First, it leads mothers to stay out of the labor market longer, which hurts their long-run employment, promotion, and earnings prospects, a problem that could be particularly pronounced in the American context.²⁶ Based on an in-depth study of 160 women with limited education, Kathryn Edin and Maria Kefalas conclude that, because they see their labor market prospects as poor, these women view children as the key source of meaning and fulfillment in their lives.²⁷ Offering a subsidy for staying home might further weaken this group’s labor market attachment.²⁸ Second, for children in less advantaged households, low income tends to be just one of multiple barriers to opportunity. For these children, early education is likely to be more helpful than added household income in boosting capabilities.²⁹

SUMMARY

Families play a vital role in ensuring economic security, raising living standards, and enhancing opportunity. Yet the American family has weakened considerably, mainly among those with less education. Gone are the days when a couple who gets pregnant and has a child in their late teens or early twenties stays together through thick and thin. Economic necessity no longer exerts the same influence, the stigma attached to out-of-wedlock birth and divorce has faded, and expectations of satisfaction from a partnership are higher. Women with college degrees are delaying children until they find a partner with whom they have a decent shot at long-term harmony and happiness. When they have kids, they are now only a little less likely to stay together than their grandparents were. In contrast, many women with less education still get pregnant and have a child before their mid-twenties. Whether they marry the father or not, the relationship seldom lasts. Many of these children do not grow up in a household with both of their original parents.

The keys to altering this trend, and thereby facilitating a revitalization of families, appear to be increases in women’s education, better financial prospects for those at the low end of the labor market, affordable good-quality childcare, and greater spread of gender-egalitarian values.