Source Data (XLSX)

Today, Americans face a wide variety of challenges in our era of tumultuous transition. We are materially better off in many ways than in the past. But despite this real progress, there is a sense that our social fabric has seen better days. Leading thinkers have issued warnings that we are increasingly “bowling alone,” “coming apart,” and inhabiting a “fractured republic.” At the heart of those warnings is a view that what happens in the middle layers of our society is vital to sustaining a free, prosperous, democratic, and pluralistic country. That space is held together by extended networks of cooperation and social support, norms of reciprocity and mutual obligation, trust, and social cohesion. In short, it is sustained by what we do together.

The following report is the first product of the Social Capital Project—a multi-year research effort that will investigate the evolving nature, quality, and importance of our associational life. “Associational life” is our shorthand for the web of social relationships through which we pursue joint endeavors—namely, our families, our communities, our workplaces, and our religious congregations. These institutions are critical to forming our character and capacities, providing us with meaning and purpose, and for addressing the many challenges we face.

The goal of the project is to better understand why the health of our associational life feels so compromised, what consequences have followed from changes in the middle social layers of our society, why some communities have more robust civil society than others, and what can be done—or can stop being done—to improve the health of our social capital. Through a series of reports and hearings, it will study the state of the relationships that weave together the social fabric enabling our country—our laws, our institutions, our markets, and our democracy—to function so well in the first place.

This introductory report contains a broad overview of what we mean by “associational life,” discusses its importance, and provides an initial portrait of several long-term changes in the state of American associational life across the domains of family, religion, community, and work. Here are some key findings in each of those domains:

Family: Fewer living in families, no less time spent with our families, later marriage and childbearing, fewer children, more single parenthood 

• Between 1975 and 2011, the share of three- and four-year-olds cared for by a parent during the day declined from 80 percent to somewhere between 24 and 48 percent. But parents are spending no less time with their children overall.
• Between 1973 and 2016, the percentage of Americans age 18-64 who lived with a relative declined from 92 percent to 79 percent. The decline was driven by a dramatic 21-point drop in the percentage who were living with a spouse, from 71 percent to 50 percent. 
• In 1970, there were 76.5 marriages per 1,000 unmarried women aged 15 and older. As of 2015, that rate had declined by more than half to 32 per thousand.
• In 1970, 56 percent of American families included at least one child, but by 2016 just 42 percent did. The average family with children had 2.3 children in 1970 but just 1.9 in 2016. Among all families—with or without children—the average number of children per family has dropped from 1.3 to 0.8.
• Between 1970 and 2016, the share of children being raised by a single parent (or by neither parent) rose from 15 percent to 31 percent.
• Between 1970 and 2015, births to single mothers rose from 11 percent of all births to 40 percent.

Religion: Lower membership and attendance, fewer raised in a religious tradition, less confidence in organized religion

• In the early 1970s, nearly seven in ten adults in America were still members of a church or synagogue. While fewer Americans attended religious service regularly, 50 to 57 percent did so at least once per month. Today, just 55 percent of adults are members of a church or synagogue, while just 42 to 44 percent attend religious service at least monthly.
• In the early 1970s, 98 percent of adults had been raised in a religion, and just 5 percent reported no religious preference. Today, however, the share of adults who report having been raised in a religion is down to 91 percent, and 18 to 22 percent of adults report no religious preference.
• In 1973, two-thirds of adults had “quite a lot” or “a great deal” of confidence in “the church or organized religion,” and in another survey the same year, 36 percent reported “a great deal” of confidence in organized religion. By 2016, those numbers had fallen to 41 percent and 20 percent, respectively.

Community: Less time with neighbors, no less time with friends, less racial segregation, more class segregation, less trust generally and in institutions but no less trust in friends or local government, no less volunteering, less voting, mixed trends on political engagement

• Between 1974 and 2016, the percent of adults who said they spend a social evening with a neighbor at least several times a week fell from 30 percent to 19 percent.
• Between 1970 and the early 2010s, the share of families in large metropolitan areas who lived in middle-income neighborhoods declined from 65 percent to 40 percent. Over that same time period the share of families living in poor neighborhoods rose from 19 percent to 30 percent, and those living in affluent neighborhoods rose from 17 percent to 30 percent.
• Between 1972 and 2016, the share of adults who thought most people could be trusted declined from 46 percent to 31 percent. Between 1974 and 2016, the number of Americans expressing a great deal or fair amount of trust in the judgement of the American people “under our democratic system about the issues facing our country” fell from 83 percent to 56 percent.
• Between 1974 and 2015, the share of adults that did any volunteering who reported volunteering for at least 100 hours increased from 28 percent to 34 percent.
• Between 1972 and 2012, the share of the voting-age population that was registered to vote fell from 72 percent to 65 percent, and the trend was similar for the nonpresidential election years of 1974 and 2014. Correspondingly, between 1972 and 2012, voting rates fell from 63 percent to 57 percent (and fell from 1974 to 2014).
• Between 1972 and 2008, the share of people saying they follow “what’s going on in government and public affairs” declined from 36 percent to 26 percent.
• Between 1972 and 2012, the share of Americans who tried to persuade someone else to vote a particular way increased from 32 percent to 40 percent.

Work: Less time with coworkers off the job, little change in commuting time, more work among women, less work among men, more “alternative work arrangements,” part-time or part-year work no more common, longer job tenure, less union membership, more occupational licensing

• Between the mid-1970s and 2012, the average amount of time Americans between the ages of 25 and 54 spent with their coworkers outside the workplace fell from about two-and-a-half hours to just under one hour.
• The share of workers living and working in different counties was 26 percent in 1970 and 27 percent in the second half of the 2000s, and commuting time has risen only modestly since 1980.
• Between the mid-1970s and 2012, among 25- to 54-year-olds, time at work rose 4 percent. The story was very different for men and women though. Hours at work rose 27 percent among women. Among men, hours at work fell by 9 percent between the mid-1970s and 2012.
• Work has become rarer, in particular, among men with less education. From the mid-1970s to 2012, hours at work fell by just 2 percent among men with a college degree or an advanced degree, compared with 14 percent among those with no more than a high school education.
• Between 1995 and 2015, workers in “alternative work arrangements” (e.g., temp jobs, independent contracting, etc.) grew from 9 percent to 16 percent of the workforce.
• Since 2004, median job tenure has been higher than its 1973 level, indicating that workers are staying in their jobs longer than in the past.
• Between 1970 and 2015, union membership declined from about 27 percent to 11 percent of all wage and salary workers.

We conclude that rising affluence has made associational life less necessary for purposes of gaining material benefits, but that we have lost much by doing less together.

What We Do Together:
The State of Associational Life in America

– Michael Novak

– Robert Nisbet

Americans are living through a period of transition to a post-industrial society based on knowledge and services, one that has wrought immense social changes.³ Past changes of similar scale—first from the long pre-agricultural past to rural farming life, followed by our tumultuous transition to an industrial economy—have been accompanied by social dislocation and subsequent adaptation.⁴ But today we are struggling to constructively orient our politics in a forward-looking way; instead, much of our politics and discourse is oriented around nostalgia for a time that is never returning.⁵

To be sure, much is going well in America. Relative to many other countries, we hold an enviable position. Having emerged from the Great Recession, the nation enjoys relatively low unemployment and incomes that, while growing too slowly, are as high as they have ever been across the board.⁶ Educational attainment continues to increase; a higher share of Americans than ever before have a college degree.⁷ Most workers enjoy longer retirements, and overall life expectancy is at an all-time high.⁸ The internet and advances in mobile communications technology have made possible unprecedented and inexpensive access to the world’s knowledge. By these standards, it has never been a better time to be alive in America.

And yet, despite this real progress, there is a disorienting sense that our social fabric is frayed. We are wealthier in material terms than ever before, but leading thinkers have issued warnings that we are increasingly “bowling alone,”⁹ “coming apart,”¹⁰ and inhabiting a “fractured republic.”¹¹ At the heart of those warnings, to one degree or another, is the view that what happens in the middle layers of our society—what we do together in the space between the individual and the state—is vital to sustaining a free, prosperous, democratic, and pluralistic country. It is in that space where we are formed, where we learn to solve problems together, where we learn the “art of association”—a space held together by extended networks of cooperation and social support, norms of reciprocity and mutual obligation, trust, and social cohesion. As Yuval Levin puts it, the middle layers

While much is going well in America on a comparative and historical basis, our associational life today appears unhealthy in many ways. For example, between 1960 and 2015, the proportion of children under 18 living with only one or neither parent increased dramatically, from 12 percent to 31 percent.¹³ In 2015, over a third of parents—and half of fathers—said they spent too little time with their children.¹⁴ Work relationships, civic engagement, and religious communities have also been important elements of building and maintaining our social fabric. Yet, a steadily increasing share of prime-age males (ages 25 to 54) have dropped out of the labor force altogether.¹⁵ There have been uneven but decades-long declines in civic and community participation, especially of the face-to-face variety.¹⁶ Many measures of religious vitality have also slowly but steadily declined over the last several decades.¹⁷

We also appear to be losing faith in national institutions. The government, the press, and both organized labor and corporate leaders lack the confidence of the people, as reflected in several long-running surveys. For example, between 1958 and 2015, the Pew Research Center shows that public trust in the federal government fell from about 73 percent to about 19 percent.¹⁸ Gallup reports that Americans have very low confidence in many major institutions; less than 40 percent of respondents said they had a combined “great deal” or “quite a lot” of confidence in the medical system (39 percent), the presidency (36 percent), the U.S. Supreme Court (36 percent), the public schools (30 percent), banks (27 percent), organized labor (23 percent), the criminal justice system (23 percent), television news (21 percent), newspapers (20 percent), big business (18 percent), and Congress (6 percent). In most cases, current levels of confidence reflect a marked decline over the last several decades. These institutions that have so much influence over our lives appear sclerotic, unresponsive to modern needs, and opaque and inaccessible to many.

There are innumerable factors that have contributed to the challenges Americans face in this new era and to the anxiety stemming from the resistance of those challenges, thus far, to ready solutions. All are important, including the discrete economic issues Washington policymakers and pundits are most familiar with and therefore most comfortable discussing—taxes and spending, welfare, trade, employment, wages, and growth.

The Social Capital Project is a multi-year research project to investigate an equally important factor that is too often overlooked—the evolving nature, quality, and importance of our associational life. In other words, what we do together.

Through a series of reports and hearings, it will study the state of the relationships that weave together the social fabric enabling our country—our laws, our institutions, our markets, and our democracy—to function so well in the first place.

Why does the health of America’s associational life feel so compromised? Where is it compromised? What consequences have followed from declining social capital? Why do some communities have more robust civil society than others? What can be done—or can stop being done—to grow Americans’ stock of social capital? What will enable us to live better together?

The pages that follow first seek to define “associational life,” “social capital,” and related concepts, providing a brief intellectual history of these ideas. A broad overview of trends in the state of associational life follows. Our analyses will be organized by considering four domains of associational life—family, religion, community, and work—and we focus specifically on painting a picture of changes in what we do together. Finally, we preview some of the issues and topics we will be studying in the years ahead.

What Is “Associational Life”?

Many people in many eras have defined and described the importance of social relationships for sustaining a free, democratic, and prosperous society—using a variety of terms such as “civil society,” “mediating institutions,” “intermediate associations,” and the more recently popular “social capital.”¹⁹

There is little consensus about what exactly these terms include. For instance, some writers have included market relationships in their definition of associational life, while others have explicitly excluded them. Many have disagreed about the inclusion of family or politics under the umbrella of civil society. Some even distinguish between face-to-face interaction and those relationships not mediated by face-to-face contact (e.g., large, impersonal national membership organizations). Despite these important differences, the Social Capital Project will take a big-tent approach to these issues. We use “associational life” as shorthand for the web of social relationships through which we pursue joint endeavors—namely, our families, our communities, our workplaces, and our religious congregations.

Doing justice to the breadth of scholarship on civil society is beyond the scope of this paper, but we review briefly the contributions of a few key thinkers in order to provide historical context and to establish its contemporary relevance.

The foremost chronicler of what Americans do together was Alexis de Tocqueville, whose nine-month trip to the United States in the early nineteenth century resulted in his two-volume treatise Democracy in America.²⁰ Tocqueville was particularly interested in the ways in which Americans spontaneously organized themselves in the service of self-governance, thereby establishing widespread norms and habits of association. He observed that what gave the United States its unique character and strength was our proclivity to form associations of all kinds, and in so doing to see our interests in the shared interests of others. As Don Eberly puts it,

In the middle of the twentieth century, sociologist and intellectual historian Robert Nisbet wrote powerfully about the role of “intermediate associations” in a free society. By intermediate associations, Nisbet simply meant the social relationships and groups that play some functional role in our lives. In his 1953 book, The Quest for Community, he notes that the human need for associational life “will not be denied, for it springs from some of the powerful needs of human nature—needs for a clear sense of cultural purpose, membership, status, and continuity.”²² Nisbet was particularly worried that if people did not find belonging and purpose in human-scale intermediate associations of family, religion, and other forms of communal life, they would find it elsewhere—a grave prospect in the context of World War II, communism, and fascism.

In the late 1970s, the idea of the middle social layers was given attention by a group of writers and thinkers at the American Enterprise Institute.²³ The “mediating structures project,”²⁴ as it came to be known, sought to think through how we might use local social organizations to provide social services such as welfare, education, housing, and so on. The participants in that project, spanning much of the ideological spectrum, vehemently disagreed about the relationship between the state and mediating institutions, but the project helped carry the idea to a new generation.²⁵

Others have added to this long tradition by attempting to conceptualize and measure the causes, consequences, and importance of what we do together under the broad umbrella of “social capital.” The earliest known use of the term comes from 1916, when L. J. Hanifan first used it to describe “tangible substances [that] count for most in the daily lives of a people, namely goodwill, fellowship, mutual sympathy and social intercourse.”²⁶ Urbanist Jane Jacobs described social capital as neighborhood networks of mutual assistance and self-governance.²⁷ Sociologist James S. Coleman defined social capital as something realized in social networks that facilitated productive cooperation.²⁸ In short, these early definitions described social capital as a resource for solving collective action problems.

More recently, Robert Putnam revived the notion of social capital in his 1993 book Making Democracy Work.²⁹ In an article summarizing that book, he defined social capital as “features of social organization, such as networks, norms, and trust, that facilitate coordination and cooperation for mutual benefit.”³⁰ He later elaborated on the idea of social capital in his bestselling book Bowling Alone, in which he defined social capital in similar terms as “connections among individuals—social networks and the norms of reciprocity and trustworthiness that arise from them.”³¹ Widely lauded at the time, Bowling Alone took a broad view of the health of American life, showing general (though not unqualified) declines in various measures of associational life including political participation, community and civic involvement, religious participation, informal social connections, volunteering, and trust.

In his book Trust, Francis Fukuyama argued for the importance to societies of non-familial sources of trust and cooperation, borrowing from James Coleman’s definition of social capital as “the ability of people to work together for common purposes in groups and organizations.”³² In a related 1995 book, The Great Disruption, he makes the case that the transition from an industrial economy to one based on knowledge weakened our social capital, which he defined as “a set of informal values or norms shared among members of a group that permits cooperation among them.”³³ Nevertheless, he argued, “social order, once disrupted, tends to get remade again,” because “human beings are by nature social creatures, whose most basic drives and instincts lead them to create moral rules that bind themselves together into communities.”³⁴

Since the mid-1990s, research about social capital has dramatically increased.³⁵ It has become a topic of interest to researchers around the world, precipitating extensive projects at the World Bank and the Organisation for Economic Cooperation and Development (OECD).³⁶ Nobel Laureate economist Elinor Ostrom has advocated the concept as “an essential complement to the concepts of natural, physical, and human capital.”³⁷ Many researchers have applied a variety of methods to understand aspects of social capital and the importance of related concepts to such topics as political life, social cohesion, volunteerism, work, civic participation, health, and happiness.³⁸

However, progress in the research has been slow due to a number of significant challenges. There are bright spots in the literature, but it generally suffers from challenges in definition, conceptualization, and measurement. As a recent National Academy of Sciences report notes,

“Social capital” has emerged as the most widely used and familiar shorthand for the concepts discussed here, but it remains a slippery term. It is not always clear, for instance, whether the dimensions associated with it constitute the thing “social capital,” are products that flow from social capital, or are lubricants that facilitate the development of social capital. We will not attempt to resolve these ambiguities. Instead, the project will use more or less precise terms as needed while, above all, emphasizing the importance of formal and informal social networks and associations that inhabit the space between the individual and large public and private institutions.

Why Is Associational Life Important?

In the intellectual tradition of studying associational life and its cousins, there emerge three key reasons why it is important.

First, the middle social layers are implicated in nearly every aspect of our lives, and therefore are critically important formative structures in which human development occurs. What we do together affects our character, capacities, deepest held moral commitments, and any number of other aspects of who we are.

Second, mediating institutions provide an important role in giving meaning and purpose to individual lives. “Meaning” and “purpose” are words that give hives to empirically minded social scientists, but nonetheless deserve our attention. Jointly pursuing common goals—prosaic or profound—draws people out of themselves, gives them a reason to get up in the morning, and to be responsive to the needs of others. When people lack the meaning and purpose derived from strong bonds and routine social attachments, they are more prone to alienation and atomization.⁴¹ Along these lines, David Brooks has argued, “The great challenge of our moment is the crisis of isolation and fragmentation, the need to rebind the fabric of a society that has been torn by selfishness, cynicism, distrust, and autonomy.”⁴²

The third reason our middle social layers are so important, especially today, is that they provide a useful means for discovering solutions to problems. The large institutions of our modern society, polity, and economy are often ill-equipped to address needs that are unique to the particular “circumstances of time and place.”⁴³ They are sometimes too far removed from local sources of knowledge and networks of trust, and they can be slow to adapt as problems evolve. Some can be out of touch with the values of specific places, breeding resentment and fueling regional polarization. As many analysts have concluded, decentralizing authority and decision-making capacity to our middle layers might go a long way to increasing America’s ability to address challenges incrementally through trial and error in ways that are much closer to the people and their varied situations.⁴⁴

An emphasis on the middle layers of our social life is no panacea for the many challenges and opportunities we face. But in an era where many of our conversations seem to revolve around the individual and large institutions, an emphasis on the space between them could bring many benefits. The rest of this report examines trends in various aspects of associational life. It will intermittently return to the basic question of why associational life matters.

How Has Associational Life Changed?

It is impossible for a report of this length to adequately survey the evidence on all the consequential ways that associational life has changed over the years. Instead, we have chosen to focus on available indicators fulfilling two requirements. First, we are interested in describing relatively long-term changes over some constant period of time. After reviewing the available evidence, we chose to focus on trends between the early 1970s and the 2010s—roughly the past 45 years. Second, we have chosen to focus as much as possible on indicators that illustrate what we do together, rather than look at any number of social or economic problems that might be presumed to reflect the state of associational life.

To organize the discussion, we assess changes in four domains of associational life, drawing on the work of previous scholars. Specifically, we take in turn changes in families, religious congregations, secular communities, and workplaces.

Families Together

We come into the world met with the joyous smiles of family, and we leave it with family mourning the loss. In between, if we are fortunate, we forge our strongest social connections with children, spouses, parents, siblings, and other relatives—with family. All subsequent social capital investment begins with the bonds between parents and children. These bonds provide purpose to parents' lives, happiness, and a connection to previous and future generations. Through parents, children receive physical sustenance and emotional support, acquire language and other skills and dispositions, and form values, beliefs, and aspirations. The development of secure connections with parents facilitates the formation of secure attachments to other people—including future partners and children, in whom parents’ stock of social capital is reinvested. Social capital obtained and created within the family is practically a necessity for all other forms of associational life.

So much of what we do together occurs in families into which we are born or that we choose to create in adulthood. The institution of marriage has weakened significantly in recent decades, but vast majorities of young Americans still hope to get married someday.⁴⁵ We also derive benefits from a range of other family relationships—not just with our children, spouses, and parents, but our brothers, sisters, grandparents, grandchildren, in-laws, cousins, aunts, and uncles. In many communities, these relationships assume special importance, as with the strong role grandparents often play in black communities or the deep extended family connections in Latino families. The happiness our family relationships impart is the most vivid illustration of the importance of healthy associational life.

Married couples, for example, report higher life satisfaction than single adults.⁴⁶ While people predisposed to be happy are presumably likelier to get married in the first place, evidence suggests that marriage has a positive causal effect on life satisfaction. Around the world, the well-known midlife dip in life satisfaction is moderated by being married.⁴⁷

Indeed, vast bodies of research find that married couples and their children tend to have better outcomes than single people and the children of single parents. Unfortunately, methodological problems render most—though not all—of these studies unpersuasive.⁴⁸ However, if we stipulate uncontroversially that happy marriages provide at least some of the beneficial outcomes to adults and children that the research claims is produced by “marriage,” we can avoid, for now, divisive debates over what public policy should or should not do to promote marriage. And the research is clear that on average, marriage is associated with innumerable positive outcomes.⁴⁹

Research on extended family’s effects suffer from similar analytical challenges but also accord with the intuition that healthy family relationships impart benefits.⁵⁰ It requires little insight to note that adult children provide help to their aging parents, and parents provide support to their adult children in the form of financial help, advice, and child care.

Trends in Family Associational Life

Closeness of Family Connections

Americans do not appear less likely to live near their parents in adulthood than in the past. A useful indicator is the share of adults living in their birth state. Research has shown that living in one’s childhood state in adulthood is strongly correlated with living near a parent.⁵¹ Between 1970 and 2015, the percentage of native-born Americans age 25 to 54 who lived in their birth state (and who thus tended to live near their parents) stayed about the same, rising from 63 percent to 64 percent.⁵² The share of prime-age adults with children living near the children’s grandparents was flat at 64 percent.

According to the General Social Survey, 38 percent of adults in 1974 said they spent a social evening with relatives several times a week or more, while 39 percent did in 2016.⁵³ It is clearly the case that children are less likely to be cared for during the workweek by a parent. For example, between 1975 and 2011, the share of three- and four-year-olds cared for by a parent during the day declined from 80 percent to somewhere between 24 and 48 percent.⁵⁴ This profound change reflects the increase in work among mothers in recent decades.

Percent of Births Occurring to Married Women, 1940-2015 ^{Source: 1940-2006 are from Solomon-Fears (2008), Table A-1; 2007 from Martin et al. (2010a), Table 18; 2008 from Martin et al. (2010b), Table 15; 2009 from Martin et al. (2011), Table 15; 2010-2013 from Solomon-Fears (2014), Table 1; 2014 from Hamilton et al. (2015), Table 15; 2015 from Martin et al. (2017), Table 15.}

However, evidence from time-use surveys suggests that both mothers and fathers are nonetheless spending more time with their children than in the past.⁵⁵ Even though more mothers are working, school-age children are not around during the day anyway during the school year. Parents appear to have otherwise compensated for any additional time they spend at the workplace versus the past.⁵⁶

Evidence does indicate, however, that parents and children are less likely today to participate with each other in activities at home than they were in the past.⁵⁷ Claude Fischer speculates that this circle can be squared if parents are spending more time with children outside the home, such as at restaurants and stores, play dates, and extracurricular activities.⁵⁸

Family Formation and Dissolution

Another indication of diminished family connections is that the number of adults living in families has declined over the past 45 years. Between 1973 and 2016, the percentage of Americans age 18-64 who lived with a relative declined from 92 percent to 79 percent.⁵⁹ The decline was driven by a dramatic 21-point drop in the percentage who were living with a spouse, from 71 percent to 50 percent. Adults who would have been married in the early 1970s were instead cohabiting (possibly as a parent, up from less than one percent to 9 percent); living alone (up from 6 to 11 percent); living with other relatives (possibly just their own child, up from 8 to 12 percent); living with roommates (up from under 2 to 4 percent); or living in a parent’s household (slightly up from 13 to 14 percent). Between 1973 and 2016, among 25- to 34-year-olds, the share who were living with a spouse dropped from 78 percent to 41 percent, and the drop among women 18-24 was from 44 percent to 9 percent.⁶⁰

Marriage rates have plummeted over the past several decades. In 1970, there were 76.5 marriages per 1,000 unmarried women aged 15 and older. As of 2015, that rate had declined by more than half to 32 per thousand.⁶¹ Americans are getting married later, and more of us are not marrying at all. In 1970, the median age at first marriage was 21 among women and 23 among men. By 2016, those medians were 27.5 and 29.5—both higher by six and a half years.⁶² Between 1970 and 2015, the share of Americans aged 50 to 54 who had never married rose from 6 percent to 14.5 percent.⁶³

These trends have numerous causes, including rising educational and employment opportunities for women, increasing affluence, and the sexual revolution. Marriage is simply not viewed as being as necessary as it once was. Increasingly, long-term romantic relationships involve couples who are unmarried but living together. Among women aged 19 to 44 who married between 1965 and 1974, just 11 percent had cohabited with their husband prior to marriage. That was true of 66 percent of women 19-44 who married between 2005 and 2009.⁶⁴

An important reason for rising cohabitation is the increasing fragility of marriage, which has led more couples to “try out” relationships with each other before committing to an institution that has seen rising failure rates over time. The divorce rate was 15 per 1,000 married women in 1970. It rose throughout the 1970s, as no-fault divorce spread throughout the country, then fell thereafter to about 18 per 1,000 in 2010, still higher than in 1970.

However, this decline was in part a product of baby boomers aging out of high-divorce life stages into low-divorce ones. If the distribution of married women across ages had been the same in 1970 as it was in 2010, the divorce rate would have risen steadily, from about 10 per 1,000 in 1970 to 18 per 1,000 in 2010.⁶⁵

These trends can also be interpreted in terms of Americans perceiving marriage as less necessary. The spread of no-fault divorce reflected demand for easier exits from marriage; between 1973 and 2016, the share of adults saying divorce should be easier to obtain rose from 32 percent to 38 percent.⁶⁶ We might expect that as rising divorce removed unhappy couples from the stock of married families, the remaining husbands and wives would be more satisfied with their marriages. However, between 1973 and 2016, the share of them reporting being in a “very happy” marriage actually fell from 67 percent to 60 percent.⁶⁷

In contrast to the non-elderly population, living arrangements were much more stable among Americans aged 65 and older. In 1973, 71 percent lived with a relative, and 70 percent did in 2016.⁶⁸ Elderly women became more likely to live with a spouse and less likely to live with another relative, while elderly men became more likely to live alone and less likely to live with a spouse or other relative. These changes reflect longer lifespans. Since women tend to marry older men, greater longevity results in more marriage for women (who do not become widowed as early as in the past), more living alone for men (who are more likely to outlive their younger wives than in the past), and less reliance on other family members.

Fertility

Along with delayed and declining marriage has come delayed and declining childbearing. Between 1970 and 2015, the average age at which women first give birth increased from about 21 years to about 26 years.⁶⁹ In 1970, 56 percent of American families included at least one child, but by 2016 just 42 percent did.⁷⁰ The average family with children had 2.3 children in 1970 but just 1.9 in 2016. Among all families—with or without children—the average number of children per family has dropped from 1.3 to 0.8.⁷¹

The decline in fertility appears to reflect a diminished interest in having children. Between 1972 and 2016, the share of adults with four or more children fell from 25 percent to 15 percent. The share who said the ideal number of children was four or more fell from 28 percent to 15 percent.⁷²

Family Instability

The subject of family breakdown—the declining share of children living with two biological parents—has generated controversy for much of the past 50 years. The crux of the debate turns on a number of questions. Do the typically better outcomes of children with two parents reflect the importance of having two parents or simply the more advantageous attributes and circumstances of those with intact marriages? Would the children of single parents do better if their actual parents—not hypothetical ones—stayed together? What are the chances that the trend in single parenthood can be reversed? Are the benefits of promoting two-parent families worth the cost in the form of possible stigmatization of single parents and their children?

These are questions the Social Capital Project will be exploring in future papers. For present purposes, we assert only a proposition that we take to be uncontroversial: healthy family relationships are valuable and constitute the most fundamental sphere of associational life. The fact that so many children today grow up in disrupted families is a cause for great concern to the extent that we can imagine a world in which many more children are raised by two happily married biological parents. Progressives and conservatives may disagree about what it would take to produce more healthy marriages—more economic opportunity? public campaigns to promote marriage?—but the goal of doing so need not be ideological or controversial.

Between 1970 and 2016, the share of children being raised by a single parent (or by neither parent) rose from 15 percent to 31 percent.⁷³ Over half of the children of high school graduates with no postsecondary education live with a single parent, and three in five children of parents without a high school diploma.⁷⁴ In part because of this educational disadvantage, more than three in five African American children live with a single parent, though single parenthood has increased sharply among non-Hispanic whites and Hispanics as well.⁷⁵ These estimates assess living arrangements at a point in time, but a majority of American children can expect to live with a single parent at some point before reaching the age of sixteen.⁷⁶

Single parenthood has risen both because of an increase in divorce and due to a rise in unwed childbearing. Between 1970 and 2015, births to single mothers as a share of all births rose from 11 percent to 40 percent.⁷⁷

Much of the increase in unwed childbearing is due to rising cohabitation. Fifty-eight percent of unmarried mothers are cohabiting at the time of their child’s birth.⁷⁸ However, cohabiting relationships tend to be much less stable than marriages. Half of children born to cohabiting parents will see their mother’s relationship to one or more men break up by their third birthday, compared to just 13 percent of children in married-parent families.⁷⁹

From a number of perspectives, then, families seem to associate together no less than in the past. They spend comparable time with their families, despite the increase in work among mothers, and appear to live as close to family members as before. But Americans spend less time in families today, reflecting the decline in marriage. They also have fewer children, which seems to reflect falling demand for them. Instead, Americans are marrying and having children later than in the past and cohabiting more. These trends reflect increasing individualism and pursuit of non-familial ends. While we may be no worse off individually for marrying and having children later, these trends may have reduced social cooperation to the extent that family life promotes community engagement.

More to the point, the decline in marriage partly reflects the weakness of the institution, as marriages have been increasingly likely to dissolve over time and the remaining marriages appear to be less happy than in the past. And these trends have almost certainly not been benign for children. Many single parents do as good a job or better as many married parents at investing in their children. But it is hard not to conclude that if we had managed to shore up marriage these past decades, children today—to say nothing of parents—would be much better connected to valuable family ties. Increasing family disconnection is of particular concern considering the role that family plays as the foundation for all other relationships.