In 2004, the late Sara McLanahan published a landmark article called “Diverging Destinies: How Children Are Faring Under the Second Demographic Transition.” The paper was the first scholarly attempt to propose that the decline of the two-parent family in the United States since the 1960’s was intensifying the already unequal life chances for poor and more advantaged children. The insight encompasses an irony that continues to perplex social policy debates: Post 1960’s changes in the family which promised people—especially women—greater personal freedom and liberation from traditional constraints were making inequality worse.
Armed with another twenty years of data, Melissa Kearney, an economist at the University of Maryland, has now revisited the subject in “The ‘College Gap’ in Marriage and Children’s Family Structure,” a working paper recently published in NBER. Her primary findings won’t surprise anyone keeping track of the scholarship on families and children, but she is able to expand and refine our understanding of the trends that McLanahan saw were creating a dangerous disparity in national well-being—disparities that since then have grown and hardened into a seemingly intractable socioeconomic reality.
McLanahan’s article, based on surveys from 1960 to 1990, showed that while most college-educated women continued to raise their children in two-parent homes, that was no longer the case for the least-educated women. A sharp rise in single motherhood among that latter group during those decades was limiting the future of children who were already at a social-economic disadvantage. A third group of what, for simplicity’s sake, we’ll call lower-middle-class women (those with a high school degree and perhaps a year or two of college) had a modest shift towards single motherhood but for the most part continued to marry and establish traditional two-parent families. Since that time, as Kearney demonstrates in her new paper, the lower-middle-class family has all but collapsed. While the children of the least- and highest-educated mothers continue to live in the same general family arrangements as they did in 1990, the percentage of their working-class peers growing up in two-parent families fell from 83 percent to 60 percent. They now resemble their poorer and least-advantaged sisters more than they do their college-educated peers. In this respect, Kearney’s paper adds to the considerable literature on the “hollowing of the middle class”; the middle class is dwindling while the ranks of the lower skilled and affluent grow further apart.
Taking advantage of a growing body of disaggregated data since “Diverging Destinies” was published—as well as a mounting interest in racial gaps—Kearney also delves into family differences between identity groups. At one end of the spectrum, a strong majority (77 percent) of White children and an even larger share of Asian children (88 percent) live with their married parents. In the middle are the 62 percent of Hispanic children living with their married parents, and at the low end are 38 percent of Black children (all 2019 numbers). The “college gap,” as the author calls it, holds for all four of the largest racial and ethnic groups, though there are notable differences between them. Black children are by far the least likely of the four groups to live with married parents, but they have a substantially better chance of doing so if their mother has a college degree: 60 percent of Black children with college-educated mothers have both parents in the house compared to a mere 30 percent for both the other Black education groups. Like the population as a whole, children of the lower-middle class in all identity groups saw the biggest decline in two-parent households between 1980 and 2019. This pattern holds for Asians, as well, but mothers of all education levels are far more likely to be married than mothers of other racial groups: 83 percent of the children of the least-educated Asian mothers, for instance, are living with their married parents, compared to 60 percent of White and Hispanic children and 30 percent of Black children. It’s a good guess this reflects the higher percentage of immigrants among the least-educated Asians, since immigrant parents are generally more likely to be married than American-born parents.
It’s reasonable to infer from several findings in Kearney’s paper that poor and lower-middle-class children now experience less familial care and social support that extends well beyond nonresidential fathers. Unlike in 1980, when close to two-thirds (64 percent) of single mothers had been divorced, in 2019, a majority (52 percent) of unpartnered mothers, or what Europeans call “lone mothers,” have never married in the first place. That’s especially true for Black mothers; more than two-thirds (70 percent) of Black unpartnered mothers have always been single.
The distinction between divorced and never-married parents turns out to be highly consequential for children, since men who were never married to their child’s mother are less likely to pay child support and, perhaps even more importantly, to be actively engaged in their children’s lives. The only place where divorce remains the primary driver of single parenthood is among the college educated, a group already better resourced in other respects. In general, the share of both married and all partnered adults has declined. Many observers of American family life had assumed that, even if marriage was becoming less popular among adults of childbearing age, parents would live and raise their children together just as unmarried parents do in much of Western Europe. But Americans haven’t adopted the continental model. While cohabitation remains fairly common, more children live in single-mother households than with cohabiting parents here. Nor are extended kin likely to take the place of missing fathers: 67 percent of children living with an unpartnered mother have no other adult in the house. Also worth noting is that the shift from single motherhood predominantly due to divorce in the early decades of family decline to never-married single motherhood today likely signals an overall decline in male connection to family life.
So how do we explain the collapse of the lower-middle-class family since 1980? Kearney makes a strong case for the theory first proposed by William Julius Wilson in his seminal study of the 1980’s Black inner city, The Truly Disadvantaged. Wilson speculated that as industrial jobs moved from cities into truck-accessible industrial parks, they left behind a population of jobless, low-skilled men, many of them Black. With no earnings and no prospects for employment, these men were far from desirable mates; they were not, as Wilson put it, “marriageable.” Many poor women decided they were just as well off trying to manage on their own, with the help of government benefits and, with luck, better off family members.
By 1980, a similar dynamic was changing the domestic calculations of working-class women. Lower-middle-class women were themselves taking jobs to compensate for the stagnating wages of their husbands. In the 1970’s, they earned only a small share of the family budget, but over the next decades their median paychecks grew, at least compared to their one-time breadwinning spouses. This was in contrast to college-educated women who continued to earn only about two-thirds of what their husbands did. As working-class men’s fortunes declined, women’s earnings were becoming as great as or even higher than their potential spouses. The household income of lower-middle-class, two-parent families has risen only 8 percent since 1980; among high school dropouts, married-couple incomes have actually declined. Kearney cites continuing evidence that even in the soulmate era, marriage remains an economic partnership, one that now fails to impress the lower-middle class. Women are reluctant to marry men who earn less than they do. No wonder that, in 2020, only 54 percent of men with a high school degree and 53 percent of men without one were married, or that marriage rates are lower in areas where women tend to earn more than men, widening both neighborhood and aggregate inequality.
Still, Kearney shows that, though the marriageable men theory offers a reasonable origin story for children’s disparate destinies, it’s missing a crucial part of the plot. In a 2018 paper, Kearney and co-author Riley Wilson looked at marriage and fertility trends in the 2000’s as the fracking boom flooded a number of de-industrialized areas with relatively high-paying jobs. The logic of the marriageable men thesis would predict that more men and women in those places would marry, and fewer babies would be born to single mothers. That’s not what happened. Yes, non-college-educated men had higher earnings, both in an absolute sense and relative to women. And yes, more babies were born where fracking improved the economic situation of childbearing adults. But marriage rates didn’t budge, while more children were born to single mothers. This is in sharp contrast to a coal boom in the 1970’s and 80’s. In that earlier case, when working-class men earned higher incomes, they got married and only then had children. It appears that in the 1970’s and early 80’s, marriage before children remained pretty much the expected norm among working-class couples, at least when men were well employed; by the 2000’s, that norm no longer held. Marriage has become irrelevant to lower-middle-class women’s decisions about childbearing.
The most disturbing part of this is that, even in the lower-middle class, children growing up with married parents are more likely to go to college and earn a high income at age twenty-five—that is, to be upwardly mobile. For that to happen—and to see inequality and immobility decline across all classes—children will need something more than higher household spending. “Reversing the decline in married-parent families for children,” as Kearney concludes, “will likely require both economic and social changes.”
Editor’s note: This article was first published by the Institute for Family Studies.