Family Complexity, the Family Safety Net, and Public Policy

Family Complexity and the Fraying of the Family Safety Net

Changes in the American family that separate marriage and child-rearing—high percentages of children born outside of marriage, cohabiting relationships, the instability of marriages—have made families of the twenty-first century more complex than the idealized vision of a married mother and father living together with their biological children. Children may have multiple adults who act as parents at different points in their life. Grandparents are likely to live to see their grandchildren become young adults, so that there are more overlapping generations in families than ever before. And children have step and half siblings, some of whom live together for a time during childhood, and some of whom never share a household.

Americans’ more complicated family lives may be fraying the family safety net that protects vulnerable children. Why is this important? Many U.S. policies assume that family members will step in to help each other weather economic hardship and health crises. The family safety net has been reasonably effective, with grandparents helping to care for young children (Luo et al. 2012; Seltzer and Yahirun, forthcoming), and spouses and adult children providing care when family members develop infirmities in later life (McGarry 1998). The unpaid care that family members provide is economically valuable (Folbre 2012). Any threat to the viability of family care may be harmful for those in need.

The American public shares the public-policy assumption that family members will help each other. Americans report a greater responsibility to help children, parents, and grandchildren weather hardship than close friends or neighbors (Nock, Kingston, and Holian 2008). Yet willingness to help depends on just what kind of child, parent, or grandchild needs help. Stepparent-stepchild relationships may be problematic in adulthood. Adult stepchildren and their stepmothers are less likely to live together, less likely to live nearby, and less likely to move closer than adult biological children and their mothers (Seltzer, Yahirun, and Bianchi 2013). And adult stepchildren are less likely to assist their stepparents than are biological children (Pezzin, Pollak, and Schone 2008). Opinions about helping stepfamily members also are more variable than opinions about helping biological family members (Ganong and Coleman 1999; Rossi and Rossi 1990). Marriage matters. If a mother and stepfather divorce, many Americans think that a stepfather’s obligation to help support his stepchildren ends (Ganong, Coleman, and Mistina 1995); and later in life, adult stepchildren’s obligations to help an aging stepfather depend on whether he is still married to the child’s mother (Ganong and Coleman 1999; Ganong, Coleman, and Rothrauff 2009).

How viable the family safety net is depends on a common understanding about who is covered by the safety net. The family safety net is less durable when people disagree about who should help. The changes in U.S. families, with less marriage and fewer marriages lasting a lifetime, challenge the effectiveness of the family safety net because these demographic changes create ambiguity about who is in the family. Perhaps the greatest ambiguities arise when marriage is never part of the picture and when parents and children are linked through cohabiting relationships and relationships that do not even reach the point of living together. Americans are less likely to define cohabiting couples as a family, compared to married couples (Powell et al. 2010), and this may account for why people are less likely to approve of “doubling up” in a shared household to alleviate economic hardship when an adult child is cohabiting than when the child is married or single (Seltzer, Lau, and Bianchi 2012).

Complexity by itself need not be harmful, particularly when there are clear rules about who is in the family and who is not.¹ In fact, there do appear to be some rules in today’s complex families. Children typically stay with their mothers when their parents’ relationship ends, and men who live with a woman and her children are expected to help support these children even when they are not the man’s own offspring (Tach et al., this volume). But other rules are less commonly shared, for instance, rules about whether a biological father’s children from a first relationship should have priority over his children from a later relationship. Social scientists can contribute to debates about family complexity by doing more to learn how clear rules develop in social interactions and through the effects of laws and policies.

The family complexity addressed in this volume may be a challenge for public policy in part because the shifting definition of who is in the family is still working itself out in the minds of those forming these families. As a result, it is especially difficult for those implementing public policies to ask those seeking help from the state who their children are and who the children’s other parent is. Another, equally important challenge is that complex families of the type addressed in this volume already face economic disadvantages, so that even among those who share a clear understanding of family obligations, there may be too few resources available for the family safety net to provide much security.

Is Family Complexity a New Threat to the Family Safety Net?

Social historians sometimes admonish sociologists for claiming to have discovered new patterns that actually have roots in the past. That criticism would be valid concerning the recent “discovery” of family complexity. We are at risk of overstating the uniqueness of contemporary patterns of instability and complexity because we sometimes take the post–World War II era as the baseline, when in fact it was the most unusual time in the past century. The 1950s and early 1960s composed a period in which marital dissolution (from divorce and death combined) was relatively low and in which childbearing outside of marriage was uncommon. Young adults married earlier than at any other time in the twentieth century, and a greater proportion of them married than before or since (Cherlin 1992). Family patterns during this era were unusually uniform, as noted in an early article by Frank Furstenberg, a contributor to this volume, and colleagues (Modell, Furstenberg, and Hershberg 1976). The U-shaped pattern of greater complexity prior to the twentieth century, less complexity in the mid-twentieth century, and greater complexity in the early twenty-first century has led some historically oriented scholars to advance a “return to complexity” thesis to explain family change in Western nations during the twentieth century. This view is most forcefully argued by the sociologist Göran Therborn in his sweeping book, Between Sex and Power: Family in the World, 1900–2000 (2004).

But while chaos may be the same everywhere, complexity is not. There are important ways in which the pre-twentieth-century complexity of family life differs from the current complexity. First, the disruptions in marriages in the earlier period were primarily due to the death of a spouse rather than to a divorce. Therefore, marital dissolution did not lead to the kinds of ties across households that occur when a parent leaves the household, perhaps to start another family, but instead retained ties to children in the first household. In addition, the death of a parent has a finality that divorce does not provide. Children whose parent died could not retain a fantasy that their parents would reunite someday; in contrast, children whose parents divorce may experience feelings of “ambiguous loss” (Boss 2009) for a parent who is still psychologically present while physically absent. Studies suggest that the negative effects of experiencing a parental death are weaker than are the effects of experiencing a parental divorce (Amato and Keith 1991), perhaps due to the lack of marital discord that precedes divorce or to institutionalized sympathy and support for widowed parents and their children.

A second way in which the earlier complexity is different is in the legal status and social treatment of children born outside of marriage. In traditional Anglo-American family law, men who fathered children outside of marriage had no responsibility to support those children, and the children had no claim on the father’s wealth. It was as if children born outside of marriage were less than full citizens; in the term still sometimes used, they were “illegitimate.” In law, fathers and their children born outside of marriage were said to be “legal strangers” to each other. This status did not change until 1972 when, in two decisions, the U.S. Supreme Court ruled that a man could claim legal responsibilities for children born outside of marriage (Stanley v. Illinois 1972) and that such children could inherit wealth from their fathers (Weber v. Aetna Casualty & Surety Co. 1972). In the latter case, Justice Lewis F. Powell wrote that “visiting this condemnation on the head of an infant is illogical and unjust.” Through decisions such as these, the law moved toward recognizing the ties of parenthood, regardless of whether they were formed within marriage. Unmarried mothers were also stigmatized in the past, whereas today, with about 40 percent of children born outside of marriage, parents and children linked by nonmarital births are more generally accepted.

Third, instability and complexity have moved up the social class hierarchy. Unpartnered and cohabiting parents, for instance, used to be concentrated among the least advantaged groups. These are the parents who are most at risk for forming unstable and complex families. As recently as 1980, children whose mothers had not completed high school were far more likely to be living with an unpartnered or cohabiting mother than were children with more educated mothers. But over the next several decades, it became more common for children with mothers who had graduated from high school but had no four-year college degree—the middle of the educational distribution—to experience parental cohabitation or unpartnered living. By 2010, children with mothers in the middle of the educational distribution were looking more like children with mothers who had low education in terms of parents’ living arrangements. The family patterns in the middle, in other words, had partially converged with the patterns at the low end (Stykes and Williams 2013).

In sum, the current, twenty-first-century complexity and the pre-twentieth-century version have at least three contrasting characteristics. First, marital dissolution by death in the nineteenth century cut the links between children and the departed parent, whereas marriage and cohabitation dissolution by separation in the current period preserve the links with nonresident parents and create the potential for additional family ties to form. Second, in the past, the lack of legal rights of “illegitimate” children and the lack of responsibilities their fathers had toward them allowed men to ignore children they fathered outside of marriage, whereas family law today requires fathers to acknowledge and support such children. Third, whereas nonmarital fertility was concentrated among the lower class in the past, and was therefore easier for most Americans to ignore, nonmarital fertility in the twenty-first century is common among people in the middle of the educational distribution. As a result of these differences, today’s complex family ties are both more visible and more likely to involve relationships that extend across households than was the case in the past.

Much of the growth in complex families in recent decades is tied to the growth of cohabitation and, in particular, the sharp increase over the past few decades in cohabiting relationships that include children. Of late, nearly all of the growth in the percentage of children born outside of marriage has occurred among women who were cohabiting at the time of birth (Kennedy and Bumpass 2008). Manning, Brown, and Stykes (this volume) show that within each of the family types that they consider (simple families, those with only sibling complexity, those with only parent complexity, and those with both types of complexity), the proportion of children living with a cohabiting couple has increased since 1996. Berger and Bzostek (this volume) studied the 1979 and 1997 cohorts of the National Longitudinal Surveys of Youth (NLSY) and reported that one of the most striking findings is the increase over time in the proportion of cohort members who had ever lived with a cohabiting partner and that partner’s children by age 30 from 1 percent in the earlier cohort to 8 percent in the later cohort. Tach et al. (this volume) found that for fathers in the NLSY-97 sample, the probability of having multiple-partner fertility was nearly twelve times higher for those with a nonmarital birth than for those having children within marriage only.

The cohabiting relationships that these authors are studying differ from the image that many Americans have of middle-class cohabitation, involving young adults who live together for a year or two prior to marrying and who postpone childbearing until after marriage. Cohabiting relationships that include children tend to occur among the non–college educated; and they often lack the long-range planning that is implicit in premarital cohabitation among young college graduates. One study found that in nearly 40 percent of cases in which a woman was cohabiting at the birth of her first child, the cohabiting relationship began after the woman became pregnant—it was a response to a pregnancy rather than a precursor to it (Rackin and Gibson-Davis 2012). The prevalence of these hasty, unplanned, postconception cohabiting relationships may help to explain why the original finding of the Fragile Families and Child Wellbeing Study—that the birth of a child constituted a “magic moment” for the many cohabiting couples that could potentially be used to encourage marriage—led to disappointment, as few of the couples actually married (McLanahan 2011).

Family Complexity and Grandparents, an Understudied, Underappreciated Family Resource

Concern about family complexity rightly focuses attention on vulnerable children and who is available to care for them and launch them into adulthood. Debates about available caregivers, however, ignore an important member of children’s families—their grandparents (but see Dunifon, Ziol-Guest, and Kopko, this volume; Kalil, Ryan, and Chor, this volume). Throughout the course of the twentieth century, children became increasingly likely to grow up with their grandparents (Uhlenberg 2005). Thus, over many of the decades during which marriage and childbearing have become such separate dimensions of family life, a new constant has appeared. At the same time, the economic well-being of the elderly has improved due to growth in Social Security, with poverty rates much lower for those age 65 and older than for those under age 18 (Seltzer and Yahirun, forthcoming). Grandparents are valuable members of the family safety net, not just for their economic resources, but also for their help with child care and during family crises (Cherlin and Furstenberg 1992; Laughlin 2010). Grandparents view the help they provide as a family responsibility (Harrington Meyer 2012), but there is little evidence of whether grandparents view all of the grandchildren from their offspring’s complex families as full-fledged family members.

Grandparents with more grandchildren spend less time with each grandchild than those with fewer grandchildren (Uhlenberg and Hammill 1998). As with other aspects of family life, there are striking educational differences in how many grandchildren a person has. One way to characterize these differences is by considering how many of an older person’s offspring have produced grandchildren, that is, how many sets of grandchildren a person has. By the time they reach old age (75 and older), older persons who did not complete high school are about twice as likely to have four or more sets of grandchildren as older persons who are college-educated (Seltzer and Yahirun, forthcoming). This suggests that the demand for grandparent assistance is likely to be greatest in families in which the oldest generation is least able to help. That grandparents still provide assistance even when they are themselves disadvantaged points to the durability of this often ignored family tie (Hughes et al. 2007)

Family Complexity, Cohabitation, and Family Policy

The central role that cohabitation is playing in the rise in family complexity suggests that laws and public policies should be developed that pertain to it. Developing these policies is challenging because of the contested political terrain around marriage and the lack of consensus among the American public about whether cohabiting relationships are family relationships. Currently, there is little family law that applies to cohabiting couples (Cherlin 2012). One solution would be to extend marriage laws to some or all cohabiting relationships. A panel of family law experts recommended this course in 2002; their proposal would have treated cohabiting partners who met criteria such as living together for two or three years, intermingling their finances, or assuming joint responsibility for a child, as if they were married (American Law Institute 2002). But the proposal met with opposition from all sides. A group of promarriage scholars and legal experts argued that officially recognizing cohabiting relationships as equivalent to marriage would erode the status of actual marriages and reduce young adults’ incentive to marry (Wilson 2006). A more left-leaning group of scholars argued that one of the important benefits of cohabitation was precisely its unregulated state: it provided a refuge for couples who did not want to be bound by the responsibilities of marriage. The latter argument has been more common in Europe; for instance, the Civil Solidarity Pact, a type of registered partnership in France, allows either partner to leave the relationship at any time without legal action (Martin and Théry 2001).

A solution may be to legalize only the aspects of cohabiting relationships that are most relevant to caring for dependents. Lifshitz (2012) proposes to extend legal regulation of the protection of children to cohabiting partners; but he would preserve autonomy by not imposing the idea of a community, with its support for equal division of property, no matter which partner earned it and when. He would also preserve the legal right of immediate exit from a cohabiting union. This degree of regulation and recognition could still prove unsatisfactory to both the promarriage camp (it erodes the unique status of marriage) and the pro-alternative-relationships camp (it constrains cohabiting partners too much). But the share of children who are living in low- and moderate-income cohabiting-partner-based families is growing so large that public policies that ignore them will be increasingly inadequate to protect children and their primary caregivers in the years ahead.

The instability and complexity that many parents and children experience also raise the issue of how to decide which relationships should receive priority in law and public policy. Child support policy has clearly placed a priority on biological fathers’ support for existing children from prior relationships, requiring that efforts be made to identify fathers at the birth of a child and to ensure that the father continue to pay part of his income in support of the child no matter how his later life develops. This priority fits with the general sentiment that men should take responsibility for the children they father and that they cannot evade that responsibility merely because they have ended (or never had) a relationship with the child’s mother. But recent studies suggest that this is not the view of low-income fathers and their current partners. For example, Nelson and Edin (2013) find that among low-income men who have children with more than one partner, the dominant view is that fathers should provide support to their current partner and to their partner’s children first and then, if anything is left, provide support to children who are living with previous partners. It is as if the men are engaged in a variation of the child-swapping pattern observed by Furstenberg (1988), in which fathers are replaced by stepfathers as men with primary responsibility for providing for the children in the household. As long as this disjuncture exists between what public policy says fathers should prioritize and what low-income fathers and their current partners think, it will be difficult for policies to succeed in encouraging an optimal level of support for all children living in complex families. Even with greater clarity about what fathers and other family members owe one another, children’s material needs may not be met by family members alone. The family safety net, frayed or strong, can only stretch as far as its members’ financial resources allow. That complex families are more likely among those most economically vulnerable points to the importance of combining public support with private, family support.

Biographies

Andrew J. Cherlin is the Benjamin Griswold III Professor of Sociology and Public Policy at Johns Hopkins University. In 1999, he was president of the Population Association of America; and in 2009 he received the Taeuber Award from the Population Association of America in Recognition of Outstanding Accomplishments in Demographic Research.

Judith A. Seltzer is a professor of sociology and the director of the California Center for Population Research at the University of California, Los Angeles. Her research investigates changes in U.S. family patterns and the consequences of family change for inequality.