For a while it looked like Hurricane Katrina would accomplish what the NAACP never could: reviving civil rights liberalism as a major force in American politics. There it was for the whole world to see: the United States was two nations, one rich, one poor and largely black, one driving away in the family SUV to sleep in the snug guest rooms of suburban friends and relatives, the other sunk in the fetid misery of the Superdome. Newsweek, echoing Michael Harringtons 1962 landmark book that ignited the War on Poverty, titled its Katrina coverage The Other America and warned the nation not to return to the old evasions, hypocrisies, and not-so-benign neglect of the problems of poverty, race, and class.
Though that liberalism revival only lasted for about five minutes, the post-Katrina insight was correct. There are millions of poor Americans, living not just in down-on-your-luck hardship but in entrenched, multigenerational poverty. There is growing inequality between the haves and the have-nots. And there are reasons to worry whether the American dream is within the reach of all.
But what two-America talk doesnt get is just how much these ominous trends are entangled with the collapse of the nuclear family. While Americans have been squabbling about gay marriage, they have managed to miss the real marriage-and-social-justice issue, one that affects far more people and threatens to undermine the American project. We are now a nation of separate and unequal families not only living separate and unequal lives but, more worrisome, destined for separate and unequal futures.
Two-America Jeremiahs usually nod at the single-parent family as a piece of the inequality story, but quickly change the subject to describeaccurately, as far as it goesan economy that has implacably squeezed out manufacturing jobs, reduced wages for the low-skilled, and made a wallet-busting college education crucial to a middle-class future. But one cant disentangle the economic from the family piece. Given that families socialize children for successor notand given how marriage orders lives, they are the same problem. Separate and unequal families produce separate and unequal economic fates.
Most people understand what happened to the American family over the last half-century along these lines: the birth control pill begat the sexual and feminist revolutions of the 1960s, which begat the decline of the traditional nuclear family, which in turn introduced the country to a major new demographic: the single mother. Divorce became as ubiquitous as the automobile; half of all marriages, we are often reminded, will end in family court. Growing financial independence and changing mores not only gave women the freedom to divorce in lemming-like numbers; it also allowed them to dispense with marriage altogether and have children, Murphy Brownstyle, on their own. (This is leaving aside inner-city teenage mothers, whom just about everyone sees as an entirely different and more troubling category.) Today, we frequently hear, a third of all children are born to unmarried women.
To put it a little differently, after the 1960s women no longer felt compelled to follow the life course charted in a once-popular childhood rhymefirst comes love, then marriage, then the baby carriage. Sure, some people got married, had kids, and stayed married for life, but the hegemony of Ozzie and his brood was past. Alternative families are just the way things are; for better or for worse, in a free society people get to choose their own lifestyles-bringing their children along for the ride-and they are doing so not just in the United States but all over the Western world.
That picture turns out to be as equivocal as an Escher lithograph, however. As the massive social upheaval following the 1960swhat Francis Fukuyama has termed the Great Disruptionhas settled into the new normal, social scientists are finding out that when it comes to the family, America really has become two nations. The old-fashioned married-couple-with-children model is doing quite well among college-educated women. It is primarily among lower-income women with only a high school education that it is in poor health. This fact may not conform to the view from Hollywood; movies from Kramer vs. Kramer to The Ice Storm to the recent The Squid and the Whale, not to mention unmarried celebrity moms like Goldie Hawn and moms-to-be like Katie Holmes, have helped reinforce the perception that elite women snubbing a conformist patriarchy were the vanguard of a vast social change. Now its pretty clear that this is a myth saying more about La-La Land than the reality of American family breakdown.
The most important recent analysis of that reality is The Uneven Spread of Single-Parent Families, a 2004 paper by Harvards David Ellwood and Christopher Jencks. The Kennedy School profs divide American mothers into three categories by education level: women with a college degree or higher; women with a high school diploma (including those with some college, whose trends look very similar to those with high school alone); and women who never graduated high school. The papers findings are worth pondering in some detail.
Forty-five years ago, there was only a small difference in the way American women went about the whole marriage-and-children question; just about everyone, from a Smith grad living in New Canaan, Connecticut, to a high school dropout in Appalachia, first tied the knot and only then delivered the bouncing bundle of joy. As of 1960, the percentage of women with either a college or high school diploma who had children without first getting married was so low that youd need a magnifying glass to find it on a graph; even the percentage of high school dropouts who were never-married mothers barely hit 1 percent. Moreover, after getting married and having a baby, almost all women stayed married. A little under 5 percent of mothers in the top third of the education distribution and about 6 percent of the middle group were either divorced or separated (though these figures dont include divorced-and-then-remarried mothers). And while marital breakup was higher among mothers who were high school dropouts, their divorce rate was still only a modest 8 percent or so.
That all changed in the decades following the 1960s, when, as everyone who was alive at the time remembers, the American family seemed on the verge of self-immolation. For women, marriage and children no longer seemed part of the same story line. Instead of staying married for the kids, mothers at every education level joined the national divorce binge. By 1980, the percentage of divorced college-educated mothers more than doubled, to 12 percentabout the same percentage as divorced mothers with a high school diploma or with some college. For high school dropout mothers, the percentage increased to 15 percent. An increasing number of women had children without getting married at all. So far the story conforms to general theory.
But around 1980, the family-forming habits of college grads and uneducated women went their separate ways. For the next decade the proportion of college-educated moms filing for divorce stopped increasing, and by 1990 it actually starting going down. This was not the case for the least educated mothers, who continued on a divorce spree for another ten years. It was only in 1990 that their increase in divorce also started to slow and by 2000 to decline, though it was too late to close the considerable gap between them and their more privileged sisters.
Far more dramatic were the divergent trends in what was still known at the time as illegitimacy. Yes, out-of-wedlock childbearing among women with college diplomas tripled, but because their numbers started at Virtually Nonexistent in 1960 (a fraction of 1 percent), they only moved up to Minuscule in 1980 (a little under 3 percent of mothers in the top third of education distribution) to end up at a Rare 4 percent.
Things were radically different for mothers in the lower two educational levels. They decided that marriage and children were two entirely unconnected life experiences. That decline in their divorce rate after 1990? Well, it turns out the reason for it wasnt that these women had thought better of putting their children through a parental breakup, as many of their more educated sisters had; it was that they werent getting married in the first place. Throughout the 1980s and nineties, the out-of-wedlock birthrate soared to about 15 percent among mothers with less than a high school education and 10 percent of those with a high school diploma or with some college.
Many people assume that these low-income never-married mothers are teen mothers, but teens are only a subset of unmarried mothers, and a rather small one in recent years. Yes, the U.S. continues to be the teen-mommy capital of the Western world, with 4 percent of teen girls having babies, a rate considerably higher than Europes. But that rate is almost one-third lower than it was in 1991, and according to up-to-the-minute figures from the National Center for Health Statistics, teens account for only about a quarter of unwed birthscompared with half in 1970. Today 55 percent of unmarried births are to women between 20 and 24; another 28 percent are to 25- to 29-year-olds. These days, it is largely low-income twentysomethings who are having a baby without a wedding ring. The good news is that single mothers are not as likely to be 15; the bad news is that there is now considerable evidence to suggest that, while their prospects may be a little better than their teenage sisters would be, they are not dramatically so.
Race has also added to misperceptions about single mothers. Its easy to see why, with close to 70 percent of black children born to single mothers todayincluding educated motherscompared with 25 percent of non-black kids. But blacks make up only 12 percent of the countrys population, and black children account for only one-third of the nations out-of-wedlock kids.
Tune out the static from teen pregnancy, race, and Murphy Brown, then, and the big news comes into focus: starting in 1980, Americans began to experience a widening Marriage Gap that has reached dangerous proportions. As of 2000, only about 10 percent of mothers with 16 or more years of educationthat is, with a college degree or higherwere living without husbands. Compare that with 36 percent of mothers who have between nine and 14 years of education. All the statistics about marriage so often rehashed in magazine and newspaper articles hide a startling truth. Yes, 33 percent of children are born to single mothers; in 2004, according to the National Center for Health Statistics, that amounted to 1.5 million children, the highest number ever. But the vast majority of those children are going home from the maternity wards to low-rent apartments. Yes, experts predict that about 40 to 50 percent of marriages will break up. But most of those divorces will involve women who have always shopped at Wal-Mart. [T]he rise in single-parent families is concentrated among blacks and among the less educated, summarize Ellwood and Jencks. It hardly occurred at all among women with a college degree.
When Americans began their family revolution four decades ago, they didnt tend to talk very much about its effect on children. That oversight now haunts the country, as it becomes increasingly clear that the Marriage Gap results in a yawning social divide. If you want to discuss why childhood poverty numbers have remained stubbornly high through the years that the nation was aggressively trying to lower them, begin with the Marriage Gap. Thirty-six percent of female-headed families are below the poverty line. Compare that with the 6 percent of married-couple families in povertya good portion of whom are recent, low-skilled immigrants, whose poverty, if history is any guide, is temporary. The same goes if you want to analyze the inequality problemstart with the Marriage Gap. Virtually all92 percentof children whose families make over $75,000 are living with both parents. On the other end of the income scale, the situation is reversed: only about 20 percent of kids in families earning under $15,000 live with both parents.
Princeton sociologist Sara McLanahan, co-author of the breakthrough book Growing Up With a Single Parent, has fleshed out the implications of the Marriage Gap for children in an important paper in Demographyand theyre not pretty. McLanahan observes that, after 1970, women at all income levels began to marry at older ages, and the average age of first marriage moved into the mid-twenties. But where mothers at the top of the income scale also put off having children until they were married, spending their years before marriage getting degrees or working, those at the bottom did neither.
The results radically split the experiences of children. Children in the top quartile now have mothers who not only are likely to be married, but also are older, more mature, better educated, and nearly three times as likely to be employed (whether full- or part-time) as are mothers of children in the bottom quartile. And not only do top-quartile children have what are likely to be more effective mothers; they also get the benefit of more time and money from their live-in fathers.
For children born at the bottom of the income scale, the situation is the reverse. They face a decrease in what McLanahan terms resources: their mothers are younger, less stable, less educated, and, of course, have less money. Adding to their woes, those children arent getting much (or any) financial support and time from their fathers. Surprisingly, McLanahan finds that in Europe, toowhere welfare supports for lone parents, as they are known in Britain, are much higher than in the United Statessingle mothers are still more likely to be poor and less educated. As in the United States, so in Europe and, no doubt, the rest of the world: children in single-parent families are getting less of just about everything that we know helps to lead to successful adulthood.
All this makes depressing sense, but when you think about it, the Marriage Gap itself presents a puzzle. Why would women working for a pittance at the supermarket cash registers decide to have children without getting married, while women writing briefs at Debevoise & Plimpton, who could easily afford to go it alone, insist on finding husbands before they start families? For a long time, social scientists assumed, reasonably enough, that economic self-sufficiency would lead more women to opt for single motherhood. And to listen to the drone of complaint about men around water coolers, in Internet chat rooms, on the Oxygen Network, and in Maureen Dowdworld, there would seem to be plenty of potential recruits for Murphy Browndom. Certainly when they talk to pollsters, women say that they dont think theres anything wrong with having a baby without a husband. Yet the women who are forgoing husbands are precisely the ones who can least afford to do so.
The conventional answer to the puzzle is this: in an economy marked by manufacturing decline, especially in cities, too many of the potential husbands for low-income women are either flipping burgers, unemployed, or in jailin other words, poor marriage material. But three facts raise doubts about this theory.
One, its not just unemployed men or McDonalds cooks who have become marriage-avoidant; working-class men with decent jobs are also shying from the altar. Two, cohabitation among low-income couples has been increasing; about 40 percent of all out-of-wedlock babies today are born to cohabiting parents. Why would there be a dearth of marriageable men, when there appear to be plenty of cohabitable fathers? And three, marriage improves the economic situation of low-income women, even if their husbands are only deliverymen or janitors. In a large and highly regarded study, the Urban Institutes Robert Lerman concluded that married, low-income, low-educated women enjoyed significantly higher living standards than comparable single mothers. Joe Sixpack may not be Mr. Darcy, but financially, at any rate, hes a lot better than no husband at all.
Still, whatever the arguments against it, the no-marriageable-men theory is entrenched in policy circles and in the academy and is unlikely to go anywhere soon, so lets try another approach to the Marriage Gap conundrum. Instead of asking why poor and near-poor women have stopped marrying before having children, lets think instead about why educated women continue to do soeven though, in order to be accepted in polite company or to put food on the table, they dont need to.
One possible answer is especially pertinent to the Marriage Gap: educated women know that theyd better marry if they want their children to succeed academically, which increasingly is critical to succeeding in the labor market. The New Economy may have made single motherhood a workable arrangement for high-earning mothers in purely economic terms, but it made a husband a must-have in terms of child rearing. No one understands better than an Amherst or Stanford B.A. that her children will have to go to college one daythe bigger the college name, the betterif they are to keep their middle-class status. These women also understand how to get their kids college-bound. Educated, middle-class mothers tend to be dedicated to what I have called The Mission, the careful nurturing of their childrens cognitive, emotional, and social development, which, if all goes according to plan, will lead to the honor roll and a spot on the high school debate team, which will in turn lead to a good college, then perhaps a graduate or professional degree, which will all lead eventually to a fulfilling career, a big house in a posh suburb, and a sense of meaningful accomplishment.
Its common sense, backed up by plenty of research, that youll have a better chance of fully developing your childrenthat is, of fulfilling The Missionif you have a husband around. Children of single mothers have lower grades and educational attainment than kids who grow up with married parents, even after controlling for race, family background, and IQ. Children of divorce are also less likely to graduate and attend college, and when they do go for a B.A., they tend to go to less elite schools. Cornell professor Jennifer Gerner was baffled some years ago when she noticed that only about 10 percent of her students came from divorced families. She and her colleague Dean Lillard examined the records of students at the nations top 50 schools and, much to their surprise, found a similar pattern. Children who did not grow up with their two biological parents, they concluded when they published their findings, were only half as likely to go to a selective college. As adults, they also earned less and had lower occupational status.
To repeat the question: Why do educated women marry before they have children? Because, like high-status women since status began, they are preparing their offspring to carry on their way of life. Marriage radically increases their chances of doing that.
This all points to a deeply worrying conclusion: the Marriage Gapand the inequality to which it is tiedis self-perpetuating. A low-income single mother, unprepared to carry out The Mission, is more likely to raise children who will become low-income single parents, who will pass that legacy on to their children, and so on down the line. Married parents are more likely to be visiting their married children and their grandchildren in their comfortable suburban homes, and those married children will in turn be sending their offspring off to good colleges, superior jobs, and wedding parties. Instead of an opportunity-rich country for all, the Marriage Gap threatens us with a rigid caste society.
So what is it about the nuclear family that makes it work so well for children decades after Americans have declared it optional? The economists and sociologists who study these things often answer that question with some variation of what might be called the strength-in-numbers theory. Kids with two parents are more likely to have two incomes cushioning them during their developing years. More money means more stability, less stress, better day care and health care, more books, more travel, and, most of all, a home in a good school districtall of which lead to educational and, eventually, workplace success. A husband and wife can support each other if one is laid off or if the other wants retraining or more education. They can take turns caring for the children. Or if they can afford to, they can specialize: the woman (yes, its still almost always the woman) can take over as homework helper and soccer-team and church-group chauffeur, while the man earns a salary. According to the strength-in-numbers theory, then, two parents are better than one much the way two hands are better than one: they can accomplish more.
But this theory finally doesnt explain all that much. If two parents are what make a difference, then why, when a divorced mother remarries, do her childrens outcomes resemble those of children from single-parent homes more than they do those from intact families? Why do they have, on average, lower school grades, more behavior problems, and lower levels of psychological well-beingeven when a stepparent improves their economic standard of living?
You could posit that children in stepfamilies may well have suffered through their parents divorce or have had a difficult spell in a single-parent home. But what, then, do we make of cohabiting parents? Two cohabiting parents also provide few of the benefits for kids that married couples do. The Urban Institutes Robert Lerman has found that even when cohabiters resemble married couples in terms of education, number of children, and income, they experience more material hardshipthings like an empty pantry or no phone or an electricity shutoffand get less help from extended families when they do. And poverty rates of cohabiting-couple parents are double those of married couples. (Lermans study controls for education, immigration status, and race.)
Others take an alternative approach to the question of why children growing up with their own two married parents do better than children growing up without their fathers. Its not marriage that makes the difference for kids, they argue; its the kind of people who marry. Mothers who marry and stay married already have the psychological endowment that makes them both more effective partners and more competent parents. After all, weve already seen that married mothers are more likely to be educated and working than single mothers; it makes sense that whatever abilities allowed them to write their Economics 101 papers or impress a prospective boss or husband also make them successful wives and mothers. Many low-income mothers may not have the skillsor, some would argue, the IQthat would get them their B.A. or a good job, and this lack makes them less likely both to marry or stay married and to raise successful children. Parents with limited cultural and material resources are unlikely to remain together in a stable marriage, Frank Furstenberg, a famed family researcher, wrote in Dissent last summer. Because the possession of such psychological, human and material capital is highly related to marital stability, it is easy to confuse the effects of stable marriage with the effects of competent parenting.
The problem with this theory is that it merely tiptoes up to the obvious. There is something fundamentally different about low-income single mothers and their educated married sisters. But a key part of that difference is that educated women still believe in marriage as an institution for raising children. What is missing in all the ocean of research related to the Marriage Gap is any recognition that this assumption is itself an invaluable piece of cultural and psychological capitaland not just because it makes it more likely that children will grow up with a dad in the house. As societys bulwark social institution, traditional marriagethat is, childbearing within marriageorders social life in ways that we only dimly understand.
For one thing, women who grow up in a marriage-before-children culture organize their lives around a meaningful and beneficial life script. Traditional marriage gives young people a map of life that takes them step by step from childhood to adolescence to college or other work trainingwhich might well include postgraduate educationto the workplace, to marriage, and only then to childbearing. A marriage orientation also requires a young woman to consider the question of what man will become her husband and the father of her children as a major, if not the major, decision of her life. In other words, a marriage orientation demands that a woman keep her eye on the future, that she go through life with deliberation, and that she use self-disciplineespecially when it comes to sex: bourgeois women still consider premature pregnancy a disaster. In short, a marriage orientationnot just marriage itselfis part and parcel of her bourgeois ambition.
When Americans announced that marriage before childbearing was optional, low-income women didnt merely lose a steadfast partner, a second income, or a trusted babysitter, as the strength-in-numbers theory would have it. They lost a traditional arrangement that reinforced precisely the qualities that they-and their men; lets not forget the men!needed for upward mobility, qualities all the more important in a tough new knowledge economy. The timing could hardly have been worse. At a time when education was becoming crucial to middle-class status, the disadvantaged lost a reliable life script, a way of organizing their early lives that would prize education and culminate in childbearing only after job training and marriage. They lost one of their few institutional supports for planning ahead and taking control of their lives.
Worst of all, when Americans made marriage optional, low-income women lost a culture that told them the truth about what was best for their children. A number of researchers argue that, in fact, low-income women really do want to marry. They have white picket dreams, say Kathryn Edin and Maria Kefalas in Promises I Can Keep: Why Poor Women Put Motherhood Before Marriage, and though the men in their lives cannot turn those dreams into reality, they continue to gaze longingly into the distance at marriage as a symbol of middle-class stability and comfort. What they dont have, however, is a clue about the very fact that orders the lives of their more fortunate peers: marriage and childbearing belong together. The result is separate and unequal families, now and as far as the eye can see.
As family experts find themselves surrendering to their own research and arguing more and more that marriage is central to the overall well-being of children, they often caution that it is not a cure-all. Is Marriage a Panacea? is the illustrative title of a 2003 article in the scholarly journal Social Problems, and you know the answer to the question without reading a page. No, shrinking the Marriage Gap may not be a magic potion for ending poverty or inequality or any other social problem. But its hard to see how our two Americas can become one without more low-income men and women making their way to the altar.
Marriage may not be a panacea. But it is a sine qua non.