Where is the Family Going? An American Perspective

The central thesis of this essay is that a large and growing “opportunity gap” separates rich kids and poor kids in the Americas, most clearly in the United States but also in Latin America, largely because of increasing class divergence in family structure. This trend, I argue, poses grave social, economic, political, and (above all) moral questions.¹

A word first about my approach to this topic. I use the term “social capital” to refer to social networks within which all of us are embedded and which contribute value to our lives in many ways. Of all those networks (from church congregations to social clubs to neighborhoods), the most fundamental is the family.² In this sense I agree completely with the premise of this conference that family is a relational good. Similarly, my essay with its foundation in empirical description is deeply in accord with the Pope’s admonition to “look at the reality of the family today in all its complexity, with its lights and its shadows”,³ though like the Holy Father, I will in the end move beyond empirical description to moral reflection.

Next, a prefatory word about my methods. My own research and expertise have focused almost entirely on the United States, and hard data about families are more widely available for the United States, so my conclusions about North American families rest on much firmer footing. On the other hand, in this essay I’ve been asked to include evidence from Latin America, and to a striking extent the core trend that I describe appears throughout the Americas. Of course, this pattern is a “first cut” judgment, with much complexity both within and between the two continents. For example, in some parts of both North and South America black slavery has had a supremely important impact on family patterns, but in other parts of the two continents slavery was less important. But while that complex and controversial topic itself deserves much more attention, time and space constraints here preclude me from delving into it.

In short, I here seek to simplify broad empirical trends rather than digging deeply into the details. I beg your pardon for that simplification, of which I am deeply conscious. I do so for three reasons: (1) time; (2) the search for common themes in this setting; (3) most important, the central empirical trend I focus on has profound moral significance – and that is the focus of this conference.

The essay is divided, roughly speaking, into four parts. I begin by focusing on marital patterns in the United States, and then in Latin America. The second section of the essay adds children to the marriage, so to speak, by focusing on parenting. The third session widens the scope to encompass the broader community, such as friends and neighbors, mentors, and last but not least, religious institutions. Fourth and finally, I reflect briefly on the larger questions of “why?” and “so what?”.⁴

My most important conclusion will turn out to be surprisingly simple: a fundamentally similar pattern of growing class divergence in family structure – what I will call a “two tier” family pattern – appears throughout the Americas. That trend toward a two-tier family structure is part of a larger and unfortunate economic, social, and moral trajectory, as I will describe later. Against that common backdrop, there are obviously differences within and across national boundaries. In fact, before turning to the evidence of commonalities, I begin with what is probably the most important single difference between North and South American marital patterns – the relative frequency of marriage, divorce, and cohabitation.

Marriage has been very common throughout much of North American history. Divorce briefly became somewhat more common in the years immediately after World War II, but sixty years ago most United States families consisted of a breadwinner dad, a homemaker mom, and kids: a stable, Ozzie-and-Harriet-style union.⁵ Divorce was uncommon, and births outside of marriage were rare in all social strata – 4 percent overall in 1950, although the rate was slightly higher among the economically disadvantaged.⁶ Although today this family structure is often considered “traditional”, historians of the family have demonstrated that in fact it did not predominate throughout United States history.⁷

Two social norms helped make the Ozzie-and-Harriet family possible: (1) a strongly patriarchal division of labor, coupled with widely shared prosperity that allowed most families to get by on one male income, and (2) a strong norm against out-of-wedlock births, so that pre-marital pregnancy was typically followed by “shotgun” marriage.⁸ As a result, most baby-boomers in the United States were raised by both biological parents. In the 1970s, however, as the boomers themselves were coming of age, that family structure suddenly collapsed, in what demographers agree was the most dramatic change in family structure in United States history. Pre-marital sex lost its stigma almost overnight; shotgun marriages sharply diminished, and then virtually disappeared; divorce became epidemic; and the number of kids living in single-parent families began a long, steady ascent.⁹

Those who have studied this change in family structure don’t agree on exactly what caused it, but most agree that these factors contributed: 

– Sex and marriage were delinked with the advent of the birth-control pill.¹⁰

– The feminist revolution transformed gender and marital norms.

– Millions of women, in part freed from patriarchal norms, in part driven by economic necessity, and in part responding to new opportunities, headed off to work.

– The end of the long post-war boom began to reduce economic security for young working-class men.

– An individualist swing of the cultural pendulum produced more emphasis on “self-fulfillment”.¹¹

The collapse of the traditional family hit the black community earliest and hardest, in part because that community was already clustered at the bottom of the economic hierarchy. That led observers to frame the initial discussion of the phenomenon in racial terms, as Daniel Patrick Moynihan did in his controversial 1965 report, “The Negro Family: The Case for National Action”.¹² But it would turn out that white families were not immune to the changes, and with the benefit of hindsight it’s clear that from about 1965 to 1980, both white and black families underwent a massive transformation.

During this period of seemingly anarchic change, it was possible to imagine that marriage and family were on their way to extinction. But the upheaval in family structure in the 1970s produced a different and unexpected outcome – a bifurcation into two very distinct family patterns. In the 1950s all social classes had largely followed the Ozzie-and-Harriet model, but the two family types that appeared after the 1970s were closely correlated with class. The result was a novel, two-tier pattern of family structure that is still with us today.¹³

In the college-educated, upper third of United States society, a “neo-traditional” marriage pattern has emerged. It mirrors the 1950s family in many respects, except that both partners now typically work outside of the home, they delay marriage and childbearing until their careers are underway, and they divide domestic duties more evenly. The result is something like Ozzie-and-Harriet – except that Harriet is now a lawyer or a social worker, Ozzie spends more time with the kids, and on two incomes, they can afford a few more luxuries. These neo-traditional marriages are more egalitarian in the gender division of labor, and they have become nearly as durable as the 1950s model, as divorce rates among this upper third have retreated from the peaks of the 1970s.¹⁴ For the children of these families the news is good, as we shall see: the way they are being raised leads to many positive outcomes.¹⁵

In the high school-educated, lower two thirds of the population, by contrast, a new, more kaleidoscopic pattern began to emerge in which childbearing became increasingly disconnected from marriage, and sexual partnerships became less durable. In this model, dubbed “fragile families” by the sociologist Sara McLanahan and her collaborators, a child’s parents may never have been married or even stably connected to each other.¹⁶

Even if the parents were married at the time of the child’s birth, that marriage was frail, as divorce rates in this social stratum continued to rise. Because both parents likely moved on to other partners, with whom they also had children, even family units with two adults often included step-parents and step-siblings. More common, of course, were single-parent families, when one parent jumped or got pushed off the marital merry-go-round.¹⁷

A mother’s age at first birth has historically been relatively low in the United States, compared to other countries, but relatively high compared to Latin America. Age at first birth rose steadily and substantially in the United States from about 21 in 1970 to about 25 in 2000, but this aggregate number is misleading, because it hides a sharply growing class disparity. College-educated women in the United States now typically delay childbearing and marriage until their late twenties or early thirties, about six years later, on average, than their counterparts a half century ago. High school-educated mothers, by contrast, typically have their first children in their late teens or early twenties, slightly earlier than their counterparts in the 1960s, and ten years earlier than college-educated moms today. See Figure 1, the first in a series of “scissors charts” that will appear in this essay, each showing a statistically significant divergence in trends between upper and lower class parents and children.¹⁸ Delayed parenting helps kids, because older parents are generally better equipped to support their kids, both materially and emotionally, giving a significant boost to children from better-off families.

Unintended births: High school-educated women don’t aspire to have more children than college-educated women, but research shows that the former typically start having sex earlier, use contraception and abortion less often, and have more unintended or semi-intended pregnancies.¹⁹ These class-linked differences are widening. According to the sociologist Kelly Musick and her colleagues, the most plausible explanations for this class discrepancy include the mothers’ ambivalence about pregnancy, the erosion of their self-confidence by low education and economic distress, and perhaps differential access or attitudes to abortion. Access to contraception doesn’t seem to explain the pattern.²⁰

Whatever the reasons, children of less-educated parents are increasingly entering the world as an unplanned surprise (complete or not, pleasant or not), while children of more educated parents are increasingly entering the world as a long-planned objective. That difference is very likely to affect both the financial and the socioemotional resources available for raising those kids.

Today, non-marital births to college-educated women remain low (about 7 percent) and have risen only slightly since the 1970s. Among high school-educated women, however, they have risen sharply over the last thirty years and now make up more than half of all the births (57 percent in 2007) in this group.²¹ (See Figure 2, another scissors chart).

Divorce: The divorce rate in the United States, having more than doubled in the 1960s and 1970s, peaked around 1980 and then began to taper off. That broad national pattern, however, concealed another significant class divergence, for the divorce rate among college-educated North Americans fell significantly after 1980, whereas it continued to rise among their high school-educated counterparts, even as marriage itself was becoming less common in that stratum of society.²² By 2000 in the United States the ratio of divorced to married people was nearly twice as great among high school-educated people (roughly 24 per hundred) as among college graduates (14 per hundred), and by 2008-2010 the gap had grown further (roughly 28 per hundred to 14 per hundred).²³

Cohabitation: At all levels of contemporary United States society, as throughout the West, cohabitation (an unmarried couple living together) has become common in recent decades. But unlike Western Europe, among younger North Americans it rarely amounts to “marriage without a license”. Although about two thirds of all marriages in the United States nowadays follow a period of cohabitation, the average cohabitation in the United States lasts about 14 months and generally does not end in marriage.²⁴ Cohabitation patterns also increasingly differ according to class. The percentage of high school-educated women who had ever cohabited doubled in the two decades after 1987, from about 35 percent to about 70 percent, while the percentage among college-educated women during that same period rose only from 31 percent to 47 percent.²⁵

College-educated, cohabiting couples in the United States seldom have children, but when pregnancy does occur, it tends to derive from a stable relationship, and a stable marriage is the likely outcome.²⁶ Among their high school-educated counterparts, by contrast, cohabitation is generally not a way station to permanent partnership. Children are often born to less educated cohabitating couples, but such cohabitation does not typically lead to marriage, nor do the partnerships generally last. Low-income men and women have children while searching for a long-term partner, not after they have found one. Nowadays, in short, most high school-educated women cohabit; most college-educated women don’t, and those who do, rarely have children.

Demographers use the term multi-partner fertility to describe the emergence of the complex, impermanent structure characteristic of less-educat ed United States families today – “blended families”, as family counselors describe them.²⁷ In such situations many different adults live in a single household, sometimes coupling, sometimes not, with impermanent partnerships and children from many different parental pairings, often with little or no contact with their biological fathers. Most commonly, children of multiple men with one wife are all living together. This residential pattern is pathological, at least from the point of view of children. It is much more common among less educated families in the US, and it is rapidly growing.

From this complex confusion at least one pattern emerges clearly: Many kids, especially from less-affluent, less-educated backgrounds, live without their fathers, and poorly educated men are less likely to be part of the lives of their children. Figure 3 portrays this aspect of the two-tier system, showing how many men of fathering age (15-44) have any biological children with whom they do not live, and of those non-resident fathers, how many have essentially no contact with their children. Compared to college graduates, high school-educated men are four times more likely to father children with whom they do not live, and only half as likely to visit those children.²⁸ In short, less educated men are far more likely to be absent dads. All these changes in family structure have produced a massive, class-biased decline in the number of children raised in two-parent families during the past half century or so. As Sara McLanahan and Christine Percheski summarize, “In 1960, only 6 percent of children in the United States lived with a single parent. Today over half of all children are expected to spend some time in a single-parent family before reaching 18 ... Children with mothers in the bottom educational quartile are almost twice as likely to live with a single mother at some point during childhood as children with a mother in the top quartile”.²⁹ Figure 4 summarizes the growth of this remarkable gap, from 13 percent points in 1953 to 57 percentage points in 2012.³⁰ The growth in unplanned pregnancies and non-marital births that I have described is concentrated among women aged 25-34, so this trend is mostly not about teen pregnancy. Of all unwed births in the United States nowadays, more than three quarters are to post-teen adults, and that share is growing.

Most important for our present purposes, then, marital and parental stability is high and rising for the upper third of United States society, but low and falling for the lower two thirds. The class gap embodied in the two-tiered United States family structure is not merely large, but has been steadily growing for nearly half a century.

Let us now turn to a necessarily more cursory look at family trends in Latin America.

Historically, both marriage and divorce have been rarer in Latin America than in the US. The most important reason for the low divorce rate, in fact, is the low marriage rate, since by definition, a union that never existed cannot be dissolved. Moreover, unlike in the US, married couples in Latin America are likely to be Catholic and thus to be constrained by the Church’s strictures on divorce. So comparing divorce rates between the two continents is less meaningful (or at least much more complicated) sociologically. That said, the low divorce rate is modestly rising in Latin America. As in the United States, class and ethnic differences in family structure are likely to be important in Latin America, as illustrated in the prize-winning Mexican film Roma. Unfortunately, I have found no good continent-wide data on class differences in marriage and divorce, so I am unable to pursue this important theme here.

Cohabitation has long been much more common in Latin America, for many complicated economic and cultural reasons that I am unable to explore here. Today cohabitation is already 4-5 times more common than in the United States.³¹ Moreover, cohabitation is rising very rapidly throughout Latin America, and marriage (already less common than in the US) is declining. Furthermore, the class gap in cohabitation in Latin America is both high and rising, just as in the US.³² As in the US, well-educated women in Latin America are increasingly treating cohabitation as a precursor to marriage rather than as an alternative.³³

Unmarried single mothers and non-marital births have historically been much more common in Latin America than elsewhere in the West (including the United States), because marriage never became a universal institution in Latin America. As two leading Latin American demographers emphasize, “a significant majority of men have had a weak sense of commitment to their children and wives”.³⁴ In recent decades, non-marital births in Latin America have increased yet more. Moreover, the class gap in the rate of unmarried motherhood has been rising, at least in the countries for which we have good data.³⁵

In short, even though the frequency of unmarried motherhood is higher in Latin America than in the United States, the fundamental trends in the class disparity of non-marital births are very similar. Among non-college-educated Latin American women, birth and child-rearing are increasingly taking place outside formal marriage, while this is less true among their college-educated counterparts. The growing class divergence in the context within which children are raised is essentially identical in the United States and Latin America.

The age at which women in Latin America have their first children has historically been relatively low compared to North America – roughly speaking, in the mother’s early twenties or even younger.³⁶ On average, age at first birth has been relatively stable over time in Latin America.³⁷ On the other hand, just as in the United States, that stability is misleading, because the education gap for age at first birth is very large, and that gap is sharply growing.³⁸ Just as in the United States, poorer, less educated Latin American women still have children at a very young age, while their university-educated counterparts (like their counterparts in the US) increasingly have their children much later in life. Once again, surface differences in family patterns (in this case, age at first birth) between the United States and Latin America turn out to hide a more fundamental similarity – a widening class divergence that leaves working class kids at a growing disadvantage compared to their counterparts from upper-middle class families.

I earlier described the complex family pattern that North American demographers term “multi-partner fertility”. Latin American demographers use a somewhat similar concept, “household complexity”, to refer to a household with a high proportion of members, especially young children, who are not directly related to the head of household. As in the United States, that phenomenon is growing in Latin America. However, this measure has a different sociological meaning south of the Rio Grande, partly because so many more men there are entirely detached from their families, and partly because other members of the extended family (grandmothers, adult daughters, aunts, and cousins) are often included in the household as a kind of social safety net for family members coping with homelessness and poverty.To be sure, that latter pattern is not unknown in the US, but it is qualitatively much more important in Latin America. So we cannot assume that the social pathology associated with North American “multi-partner fertility” is necessarily characteristic of Latin American “household complexity”.

To summarize our argument to this point, despite many differences within and between the two American continents, the available evidence clearly suggests the emergence of a two-tier family structure in recent years, most clearly in the United States but also in Latin America, as upper-class families become stronger and more stable, whereas working class families become weaker and less stable. It is important to acknowledge that because formal structures are not deterministic, parental love can blossom in the most adverse circumstances. Nevertheless, as we shall now see, this two-tier structure increasingly constrains the opportunities available to poor kids.

Parenting practices mediate between family structure and what happens to children, so it’s important for our purposes to examine trends in parenting. Unfortunately, in the time available to me I’ve been unable to discover systematic evidence on how parenting practices have evolved in Latin American in recent decades, still less on how those trends might have differed across the social hierarchy. So unfortunately this section of my essay must rely almost exclusively on evidence from the United States.

In what follows, I will review the best evidence of these national patterns in parenting, and I will explore what difference these patterns make for children’s prospects. I begin with a close focus on the latest scientific research on brain development in young children – research that clarifies exactly what aspects of parenting help and hurt most in terms of a child’s cognitive and socio-emotional development. I then zoom back to a wide-angle view of class differences in parenting practices nationwide over the last several decades, and will explore how and why those class differences have grown, to the relative disadvantage of poor kids.

Recent research has greatly expanded our understanding of how young children’s early experiences and socioeconomic environment influence their neurobiological development, and how, in turn, early neurobiological development influences their later lives. These effects turn out to be powerful and long-lasting. “Virtually every aspect of early human development”, write the authors of a landmark study by the National Academy of Sciences, “from the brain’s evolving circuitry to the child’s capacity for empathy, is affected by the environments and experiences that are encountered in a cumulative fashion, beginning in the prenatal period and extending throughout the early childhood years”.³⁹ The bottom line: early life experiences get under your skin in a most powerful way.

Healthy infant-brain development requires connecting with caring, consistent adults. The key mechanism of this give-and-take learning is termed by specialists in child development “contingent reciprocity” (or more simply, “serve-and-return” interaction).⁴⁰ This interaction is classically illustrated when a parent, while reading to a toddler, points at pictures and names them and the child is encouraged to respond. The brain, in short, develops as a social organ, not an isolated computer.

Intellectual and socio-emotional development are inextricably intertwined from an early age. Research has shown that so-called “non-cognitive” skills (grit, social sensitivity, optimism, self-control, conscientiousness, emotional stability) are very important for life success. They can lead to greater physical health, school success, college attendance, employment, and lifetime earnings, and can keep people out of trouble and out of prison.⁴¹

So on the positive side of the ledger, the child’s interaction with caring, responsive adults is an essential ingredient in successful development. On the other side of the ledger, neglect and stress, including what is now called “toxic stress”, can impede successful development.⁴² Summarizing the results of many studies, the Nobel Prize-winning economist James Heckman writes, “Early adverse experiences correlate with poor adult health, high medical care costs, increased depression and suicide rates, alcoholism, drug use, poor job performance and social function, disability, and impaired performance of subsequent generations”.⁴³ Kids at any socioeconomic level can encounter such adverse experiences, of course, but those who grow up in low-income, less-educated families are at considerably greater risk.⁴⁴ The fundamental social significance of the neurobiological discoveries that I’ve just summarized is that healthy brain development in American children turns out to be closely correlated with parental education, income, and social class.⁴⁵ Class-based differences in parenting style are well established and powerfully consequential. The ubiquitous correlation between poverty and child development (both cognitive and socioemotional) is, in fact, largely explained by differences in parenting styles, including cognitive stimulation (such as frequency of reading) and social engagement (such as involvement in extracurricular activities).⁴⁶

But what about trends in parenting over time? Reliable indicators are hard to find, because persuasive measurement requires repeated, identical surveys over many years. But there is one exception: family dinners. Whether or not a family breaks bread together at the end of the day may seem a trivial issue of pop sociology, but in fact, trends in family dining turn out to tell a revealing story. Jane Waldfogel has shown that (even after controlling for many other factors) family dining is a powerful predictor of how children will fare as they develop.⁴⁷

From the mid-1970s to the early 1990s, as Figure 5 shows, family dinners became rarer in all social echelons, as families struggled to manage the new scheduling complexities of having two working parents. In the mid-1990s that steady waning of opportunities for family conversation was suddenly halted among college-educated parents, but it continued uninterrupted among high school-educated families.⁴⁸ The result is another of the scissors charts that appear throughout this essay – a growing gap in childhood experience between kids from well-educated, affluent backgrounds and kids from less-educated, impoverished backgrounds. Family dining is no panacea for child development, but it is one indicator of the subtle, but powerful investments that parents make in their kids (or fail to make). And it’s worth pairing investments of time with that other powerful form of investment – investments of money.

Parents from all social backgrounds nowadays invest both more money and more time in raising their kids than was true a generation ago. However, college-educated parents have increased their investments of both money and time much more rapidly than less affluent parents – and not just at the dinner table. Because affluent, educated families have not only more money but also more time (because they typically split childcare between two parents), they have been able to increase their investments much faster than poor parents (usually single moms). As a result, the class gap in investments in kids has become wider and wider.

On average, parents from all socioeconomic strata have increased their spending on childcare and education over the past five decades. But that spending, always somewhat unequal, has become steadily even more unequal over the decades. (See Figure 6.) The increasing gap is concentrated in spending on private education and childcare, but a class gap in spending is also visible for music lessons, summer camp, travel, school supplies, books, computers, extracurricular activities, recreation, and leisure.⁴⁹ These differences in parental investment, in turn, are strong predictors of children’s cognitive development.⁵⁰

Parents at all educational and income levels are spending more time with their kids nowadays than their counterparts did a half-century ago. However, as we saw above with money, the increase is much greater among college-educated parents than among high school-educated parents. Figure 7 (like Figure 6, yet another scissors chart) shows trends in the time that parents from different educational backgrounds have spent on developmental care for infants aged 0-4.⁵¹ By 2013, however, the average infant or toddler of college-educated parents was getting half again as much quality time every day as the average infant or toddler of high school-educated parents. What are kids from less-educated homes doing when they are not getting personal attention from their parents? The answer, overwhelmingly, is TV and the internet. In short, rich kids get more face time, while poor kids get more screen time. So kids from affluent, educated homes get the best of both worlds – more monetary investment (because their parents can afford it) and more time investment (because their two parents are able to make it a priority) – whereas kids from lower class homes get the worst of both worlds.

The everyday hassles of parenting are stressful: cleaning up after the kids, managing multiple schedules, whining, lack of privacy, and lack of time for self and partner. Everyday stress levels vary across families, of course, but a vast body of research links parental stress with less sensitive, less responsive parenting, and thus with bad outcomes for kids. Stressed parents are both harsher and less attentive parents.⁵² As Laura Bush once observed in a 2007 White House discussion of the growing class gaps among American kids, “If you don’t know how long you’re going to keep your job, or how long you’re going to keep your house, you have less energy to invest in the kids”.⁵³ The widening economic cleavage in America exacerbates the parenting gap both directly and indirectly (via the effects on family structure we discussed earlier in this essay). The disadvantages facing poor kids begin early and run deep, and are firmly established before the kids get to school.

This growing class gap is only enhanced by what happens to kids once they enter school, because whom you go to school with matters a lot. Regardless of their own family background, kids do better in schools where the other kids come from affluent, educated homes. This pattern appears to be nearly universal across the developed world.⁵⁴ “The social composition of the student body is more highly related to achievement, independent of the student’s own social background, than is any other school factor”, wrote sociologist James Coleman. In fact, Coleman coined the term “social capital” to describe this effect of school-based social networks on student achievement.⁵⁵

What kids from affluent homes and neighborhoods bring to school encourages higher achievement among all students at those schools. But the opposite is also true: the disorder and violence that kids from impoverished homes and neighborhoods bring to their schools discourages achievement for all students at those schools. The American public school today is a kind of echo chamber in which the advantages or disadvantages that children bring with them to school have effects on other kids.

A second fundamental fact in the United States is that rich and poor parents (and children) are increasingly living in separate neighborhoods, and there is some reason to believe that the same is true in Latin America and Europe.⁵⁶ This residential sorting by income over the last 30-40 years has shunted high-income and low-income students into separate schools. The growing class segregation of our neighborhoods and thus of our schools means that middle-class kids hear mostly encouraging and beneficial echoes at school, whereas lower-class kids hear mostly discouraging and harmful echoes.

Since the growth of neighborhood segregation appears to be nearly universal in today’s world, and since the “spillover” effects of that segregation also seem to be universal, there is reason to believe that these factors combined are fostering greater inequality of opportunity in both the United States and Latin America.

In turn, at least in the United States, the combination of growing residential segregation and the growing effects of that segregation on schoolbased social networks means that the class gap in college completion, which was already substantial 30-40 years ago, has steadily expanded. This matters hugely, because completing college is increasingly important for socioeconomic success, physical and mental health, longevity, life satisfaction, and more. Figure 8 estimates the big picture over the past 40 years.⁵⁷ On the measure of post-secondary education that matters most – graduating from college – kids from affluent backgrounds are pulling further and further ahead, yet one more of the dispiriting scissors charts in this essay.

We North Americans like to think of ourselves as “rugged individualists” – in the image of the lone cowboy riding toward the setting sun, opening the frontier. But at least as accurate a symbol of the U.S. national story is the wagon train, with its mutual aid among a community of pioneers. Throughout our history, a pendulum has slowly swung between the poles of individualism and community, both in our public philosophy and in our daily lives.⁵⁸ In the past half century we have witnessed, for better or worse, a giant swing toward the individualist (or libertarian) pole in our culture, society, and politics. At the same time, researchers have steadily piled up evidence of how important social context, social institutions, and social networks – in short, our communities – remain for our well-being and our kids’ opportunities.

Community networks have powerful effects on health, happiness, educational success, economic success, public safety, and (especially) child welfare.⁵⁹ However, like financial capital and human capital, social capital is distributed unevenly, and differences in social connections contribute to the youth opportunity gap. Many studies have shown that better-educated Americans have wider and deeper social networks, both within their closest circle of family and friends and in the wider society.60 By contrast, less-educated Americans have sparser social networks, concentrated within their own family. Figure 9 shows that both race and class matter for the density of “close” friendship – the sort of “strong ties” that can provide socio-emotional and (in a pinch) material support.⁶¹ Contrary to romanticized images of close-knit communal life among the poor, lower-class Americans today, especially if they are non-white, tend to be socially isolated, even from their own neighbors.

Perhaps more important, more educated Americans also have many more “weak ties”, that is, connections to wider, more diverse networks. The reach and diversity of these social ties is especially valuable for social mobility and educational and economic advancement, because such ties allow educated, affluent parents and their children to tap a wealth of expertise and support that is simply inaccessible to parents and children who are less well-off.

As Figure 10 shows, college-educated parents are more likely to know all sorts of people. This weak-tie advantage is especially great when it comes to occupations that are most valuable for their kids’ advancement – professors, teachers, lawyers, medical personnel, business leaders – but it is visible even among more traditional working class connections, like police officers and neighbors. In short, the social networks of more affluent, educated families amplify their other assets in helping to assure that their kids have a richer set of opportunities.

Have these class differences in social networks changed in recent years? One recent study has shown a steady decline in such networks over the last 50 years, while another concludes that “Americans’ disengagement and their retreat to the relative social isolation of the homebody and communal patterns constitute a trend that, even if common to individuals of all classes, affects members of the lower classes disproportionately, ultimately reinforcing differences between social classes”.⁶² While hard evidence is still too limited for a final verdict, there is reason to believe that class differences in social ties – especially weak ties that are important for upward mobility – are not only great, but may be growing.

But what about the internet? Does it help to close the networking gap between rich kids and poor kids, or does it widen that gap? Research shows that compared to their poorer counterparts, young people from upper-class backgrounds (and their parents) are more likely to use the internet for jobs, education, political and social engagement, health, and news-gathering, and less for entertainment or recreation.⁶³ Even though lower-class kids by now have virtually equal physical access to the internet, they lack the digital savvy to exploit that access in ways that enhance their opportunities. The internet seems more likely to widen the opportunity gap than to close it.⁶⁴

Adults outside the family often play a critical role in helping a child develop his or her full potential via natural mentoring relationships that spring up with teachers, pastors, coaches, family friends, and so forth. Measurably, mentoring matters.⁶⁵ And here, too, there are substantial class differences. Figure 11 summarizes the pattern, showing that kids from affluent, educated homes benefit from a much wider and deeper pool of mentors. Kids from affluent families are two to three times more likely to have mentors from virtually all categories outside the family – teachers, family friends, religious and youth leaders, coaches. Poor kids, by contrast, are less likely to have any mentoring support at all, and if they do, it is mostly limited to their extended family, and thus not so likely to reach beyond the family’s own resources. Nearly two thirds of affluent kids (64 percent) have some mentoring beyond their extended family, while nearly two thirds of poor kids (62 percent) do not. This stunning disparity in turn widens the opportunity gap, because kids from more privileged backgrounds are generally savvier about how to climb the ladder of opportunity.

As we observed earlier, residential segregation by social class across America has been growing for decades, so fewer affluent kids live in poor neighborhoods, and fewer poor kids live in rich neighborhoods. Growing up in a poor family and going to school with poor kids both constrain opportunity, as we have seen. Here the question is whether growing up in a poor neighborhood imposes any additional, synergistic handicaps, and the answer is yes.

North America’s leading expert on neighborhoods, Robert J. Sampson, has shown that United States neighborhoods are deeply unequal and that that inequality has powerful effects on their residents. Pervasive neighborhood inequality, he writes, has consequences “across a wide range of how Americans experience life ... crime, poverty, child health, public protest, the density of elite networks, civic engagement, teen births, altruism, perceived disorder, collective efficacy, [and] immigration”. He concludes, “What is truly American is not so much individual [inequality] but neighborhood inequality”.⁶⁶

Affluent neighborhoods boost academic outcomes in part because youth-serving institutions, like quality childcare, libraries, parks, athletic leagues, and youth organizations, are more common there than in poor neighborhoods. Conversely, careful studies have documented that poor neighborhoods foster behavioral problems, poor mental and physical health, delinquency, crime, violence, and risky sexual behavior.⁶⁷ Most neighborhood studies have focused on cities, but recent research has shown depressingly similar effects in rural areas.⁶⁸

One consequence of these neighborhood differences is that trust in neighbors is higher in richer, more educated neighborhoods, and that trust in turn helps all the young people in the neighborhood, regardless of their family resources. The close association of neighborhood trust and neighborhood poverty is illustrated in Figure 12.⁶⁹ Regardless of your own characteristics, if you live in an affluent neighborhood, you are much more likely to know and trust your neighbors. As we have seen, more poor kids are living in poor neighborhoods, while more rich kids are living in rich neighborhoods, so the benefits of collective efficacy and trust are increasingly concentrated on rich kids. It does indeed take a village to raise a child, but poor kids in the United States (and many other countries) are increasingly concentrated in derelict villages.

Around the world, social trust is almost always higher among haves than have-nots, and that pattern has long held true for American youth.⁷⁰ Trust has fallen among youth of all social backgrounds during the past half century.⁷¹ However, as Figure 13 shows, during the past several decades the long-standing class gap in social trust among American adolescents has significantly widened, producing yet another scissors chart. By the 21st century barely one out of every seven poor kids say that “most people can be trusted”. This deep personal isolation is one of the most troubling consequences of the broader societal changes that I have explored in this essay.

Participants in this conference may be forgiven for wondering why religion has played such a small role up to now in my analysis of family, parenting, and community. In fact, I end my essay now with a discussion of religion precisely because religious institutions are central and indeed archetypical as a backdrop to the long trend toward a two-tier opportunity structure.

I begin with the fact that religious institutions are central to youth wellbeing.⁷² Religious communities in the United States are important service-providers for young people and the poor. Weekly churchgoers are two to three times more likely to volunteer to help the poor and young people than are non-churchgoers, and are much more likely to contribute financially to those causes.This religious edge applies both to volunteering and giving through religious organizations, and (more surprising) to volunteering and giving through secular organizations.The crucial ingredient seems not to be theology but rather involvement in a religious congregation. As I wrote in an earlier book, loosely speaking, religious people seem to be nicer than non-religious people, other things being equal.⁷³

In addition to good works, religious involvement by youth themselves is associated with a wide range of positive outcomes, both academic and non-academic.⁷⁴ Compared to their unchurched peers, youth who are involved in a religious organization take tougher courses, get higher grades and test scores, and are less likely to drop out of high school. A child whose parents attend church regularly is 40 to 50 percent more likely to go on to college than a matched child of non-attenders.

Churchgoing kids have better relations with their parents and other adults, have more friendships with their peers, are more involved in extracurricular activities, and are less prone to substance abuse (drugs, alcohol, and smoking), risky behavior (like not wearing seat belts), and delinquency (shoplifting, misbehaving in school, and being suspended or expelled).

As with mentoring, religious involvement – when it happens – makes a bigger difference in the lives of poor kids than rich kids, in part because affluent youth are more exposed to other positive influences. In short, ceteris paribus, religious engagement is good for kids. All the generalizations in these paragraphs, I emphasize, are based on careful statistical controls.

Religious engagement has traditionally been less class-biased than virtually any other sort of community activity in the United States.⁷⁵ But very importantly, that is no longer true. Nowadays, poor families are generally less involved in religious communities than affluent families, and this class gap, too, is growing. Throughout the ups and downs of American religiosity during the past several decades, religious observance has tended to rise faster, or fall more slowly, among better-educated Americans than among their less well-off counterparts. Moreover, this growing class gap in church attendance appears among both blacks and whites. If you listen carefully today, hymns in American houses of worship are increasingly sung in upper-class accents.⁷⁶ Many religious leaders seem to be unaware of this essential truth about religion in 21st century America.

Not surprisingly, this same trend shows up among adolescents.Young people’s church attendance has fallen in all classes in recent decades, but has fallen twice as fast among kids from the lower third of the socioeconomic hierarchy as among kids from the upper third. The now-familiar scissors gap is shown in Figure 14. At least in the United States, but probably elsewhere, the Church has been unable to withstand the ever-widening gap between young people from “have” and “have not” family backgrounds. For this audience, that may perhaps be my most important conclusion.

Why? And So What?

If I had more time and space, I should have spent much time analyzing why this two-tiered society, with its fading of equal opportunity for all God’s children, has occurred now. I must here be brief.

Marriage has not lost its allure. An overwhelming majority of Americans from all classes want to marry, and most expect to marry. So why has the two-tier class divergence in actual behavior become so marked in the last three or four decades?

Economic inequality is clearly a primary villain. The greatly reduced economic prospects experienced by poorer, less-educated Americans over these decades (greater job instability and declining relative earnings) have made it far more difficult for them to attain and sustain the traditional pattern of marriage, while the upper tier of Americans have steadily gained more resources to maintain a family.

Culture is another important part of the story. Gender and sexual norms have changed, in particular, as have the roles of less-educated men and more-educated women.⁷⁷ For poor men, the disappearance of the stigma against premarital sex and non-marital birth, and the evaporation of the norm of “shotgun marriages”, broke the link between procreation and marriage. For educated women, the combination of birth control and greatly enhanced professional opportunity made delayed childbearing both more possible and more desirable.

Scholars debate the relative importance of “structural” (or economic) and “cultural” explanations for the emergence of the two-tier system. The most reasonable view is that both are important. Moreover, cause and effect are entangled here: poverty produces family instability, and family instability in turn produces poverty.

“Family values” conservatives have sometimes argued that liberalism and secularism cause family disintegration. But unwed births and single-parent families are widely distributed across America, and are concentrated neither in secular areas nor in “blue” states, which presumably have pursued more progressive policies. If anything, the opposite seems to be true: divorce and single-parent families are especially common in the southeastern, heavily Republican, socially conservative Bible belt.⁷⁸ Changing personal values are an important part of the story, but only in conjunction with adverse economic trends, and as an empirical matter, theology and ideology seem to have very little to do with it (I recognize that this empirical judgment will probably not win universal approval in this setting).

But in the end, I believe, the key explanation has been a moral decay – not as defined by our incessant culture wars, but in the even deeper sense of self-centeredness: America’s shriveled sense of “us” and of who count as “our kids”. Family breakup can often be better for the adults involved, but it is rarely good for the children. Moreover, caring for kids was once a responsibility not merely of a child’s biological parents, but a shared community responsibility, but that ethic has withered in recent decades. Half a century ago when adults spoke of “our kids”, they meant all the kids in town, even the kids “from the wrong side of the tracks”; but now “our kids” means “my own children”, not others’ children.That narrowing of the effective scope of “our kids” has had dramatically different effects on privileged and impoverished children, producing our tragically two-tiered societies.

This transformation of United States society into a two-tier society has very broad and deleterious consequences, not merely for our children, but for all of us.

– It is bad for our economy, because we simply cannot afford to discard talent, no matter where it might be found. Opportunity is not a zero-sum game; investing in poor kids will help everyone, including rich kids.

– It is bad for our democracy, because as they grow up, throw-away kids become ready tinder for demagogues.

– But above all, it is simply and deeply wrong. Ignoring kids simply because they are not our own children is morally indefensible by any standard.

 

 

¹ The primary sources for this essay include Robert D. Putnam, Our Kids: The American Dream in Crisis (New York: Simon & Schuster, 2015); Albert and Elizabeth Florez-Paredes, “Families in Latin America: Dimensions, Diverging Trends, and Paradoxes”, in Unequal Family Lives: Causes and Consequences in Europe and the Americas (Cambridge University Press, 2018), Naomi R. Cahn (Editor), June Carbone (Editor), Laurie Fields DeRose (Editor), W. Bradford Wilcox (Editor), 40-65; and “Marriage Trends in Latin America: A Fact Sheet”, National Healthy Marriage Resource Center (www.healthymarriageinfo.org, 2022). All those sources in turn include abundant and comprehensive notes that show the underlying scientific evidence.

² For deeper analysis of my concept of “social capital”, see my Bowling Alone: The Collapse and Revival of American Community, 20th anniversary edition (NewYork: Simon & Schuster, 2021).

³ Amoris Laetitia 32.

⁴ To simplify my discussion of social class, I generally compare the top third of the population in terms of socioeconomic status with the bottom two thirds. That distinction turns out to be empirically virtually identical with the one third of parents who have completed college versus the two thirds of parents with no more than a secondary education. Fortunately, we are able to use that same educational break in both the United States and Latin American analyses. Other measures, like family income or family wealth, are so closely correlated with education that that basic one-third/ two-thirds comparison is essentially the same, no matter what empirical measure of socioeconomic status we use. None of my generalizations turn on this specific operationalization of social class, and in fact, if I used more stringent measures, the evidence for my generalizations would be even stronger. 

⁵ “The Adventures of Ozzie and Harriet” was a very popular TV sit-com in the United States in the 1950s and 1960s, in which Ozzie was a stereotypical father who worked outside the home, leaving the children to Harriet, a stereotypical stay-at-home housekeeper. “Ozzie and Harriet” has since between a common term among family sociologists to describe this traditional division of labor.

⁶ Andrew J. Cherlin, “Demographic Trends in the United States: A Review of Research in the 2000s”, Journal of Marriage and Family 72 (June 2010): 406.

⁷ Representative critics of this traditional marriage, especially from a feminist point of view, include Judith Stacey, Unhitched: Love, Marriage, and Family Values from West Hollywood to Western China (New York: New York University Press, 2011); Stephanie Coontz, The Way We Never Were: American Families and the Nostalgia Trap (New York: Basic Books, 2000); Nancy Chodorow, The Reproduction of Mothering (Berkeley, CA: University of California Press, 1978); Arlie Hochschild, The Second Shift: Working Parents and the Revolution (New York: Avon Books, 1990); John R. Gillis, A World of Their Own Making: Myth, Ritual, and the Quest for FamilyValues (Cambridge, MA: Harvard University Press, 1996).

⁸ “Shotgun marriage” in American vernacular refers to a marriage that occurs after conception, but before birth, as the pregnant girl’s father forces the boy to marry the girl before the child appears. In the 1950s and 1960s, 52-60% of pre-marital pregnancies were resolved by a shotgun marriage, but by the early 1990s that had fallen to 23%, according to U.S. Census Bureau, “Trends in Premarital Childbearing, 1930 to 1994”, by Amara Bachu, Current Population Reports (Washington, DC, 1999), 23-197. For careful analysis of rates of pre-marital conception and shotgun marriage from (roughly speaking) the 1940s to the late 1970s, see Paula England, Emily Shafer and Lawrence Wu, “Premarital Conceptions, Postconception (‘Shotgun’) Marriages, and Premarital First Births: Educational Gradients in U.S. Cohorts of White and Black Women Born 1925-1959”. Demographic Research 27 (2012): 153-166. From roughly the late 1950s to the late 1970s, pre-marital conception among less-educated white women rose from about 20% to about 30%, while the rate among white college grads remained steady at about 10%. Among black women, the equivalent changes were from about 50% to about 70% for less educated black women and from about 25% to about 35% for black college graduates. Among women who conceived before marriage, the rate of shotgun marriages fell over this period from about 65% to about 45-50% for white women and from about 30% to about 5-10% among black women.

⁹ Statistics for these claims:

• Pre-marital sex: The fraction of Americans who believed that premarital sex was “not wrong” doubled from 24% to 47% in the four years between 1969 and 1973 and then drifted upward through the 1970s to 62% in 1982. Robert D. Putnam and David E. Campbell, American Grace (NewYork: Simon and Schuster, 2010), 92-93.

• Shotgun marriages: In the 1960s roughly half (52 percent) of all brides were pregnant, whereas 20 years later, only one quarter (27 percent) were. Patricia H. Shiono and Linda Sandham Quinn, “Epidemiology of Divorce”, Future of Children: Children and Divorce 4 (1994): 17.

• Divorce: The annual divorce rate for married women aged 15-44 more than doubled between 1965 and 1980. Shiono and Quinn, “Epidemiology of Divorce”, 17.

• Single-parent families: In the first half of the 20th century most single-parent families were such because of the death of a parent, but that fraction sharply declined from the 1930s to the 1970s. Leaving orphans aside, the fraction of 16-year-olds living with two biological parents declined from 85% in the 1960s to 59% in the 1990s. David T. Ellwood and Christopher Jencks, “The Spread of Single-Parent Families in the United States Since 1960”, in The Future of the Family, eds. Daniel Patrick Moynihan, Timothy M. Smeeding, and Lee Rainwater (NewYork: Russell Sage Foundation, 2004), 25-65. 

¹⁰ George A. Akerlof, Janet L.Yellen and Michael L. Katz, “An Analysis of Out-Of-Wedlock Births in the United States”, Quarterly Journal of Economics 11 (1996): 277-317.

11 Cherlin, The Marriage-Go-Round; David Popenoe, War Over the Family (New Brunswick, NJ: Transaction Publishers, 2005); Paul R. Amato, “Institutional, Companionate, and Individualistic Marriages: Change over Time and Implications for Marital Quality” in Marriage at the Crossroads: Law, Policy, and the Brave New World of Twenty-First-Century Families, eds. Marsha Garrison and Elizabeth S. Scott (Cambridge, UK: Cambridge University Press, 2012), 107-125; Robert N. Bellah, Richard Madsen, William M. Sullivan, Ann Swidler and Steven M. Tipton, Habits of the Heart: Individualism and Commitment in American Life (Berkeley, CA: University of California Press, 1985).

¹² U.S. Department of Labor. Office of Policy Planning and Research, The Negro Family: The Case for National Action, by Daniel P. Moynihan (Washington, DC, 1965). 

¹³ Landmark scholarly recognition was McLanahan, “Diverging Destinies: How Children Are Faring under the Second Demographic Transition”.

¹⁴ Steven P. Martin, “Growing Evidence for a ‘Divorce Divide’? Education and Marital Dissolution Rates in the U.S. since the 1970s”, working paper (University of Maryland-College Park, 2005), accessed May 12, 2014; Steven P. Martin, “Trends in Marital Dissolution by Women’s Education in the United States”, Demographic Research 15 (2006): 552; Frank F. Furstenberg, “FiftyYears of Family Change: From Consensus to Complexity”, ANNALS of the American Academy of Political and Social Science 654 (July 2014): 12-30.

¹⁵ For a careful summary of these studies, see Sara McLanahan and Christine Percheski, “Family Structure and the Reproduction of Inequalities”, Annual Review of Sociology 34 (August 2008): 257-276.

¹⁶ An entire issue of the journal Future of Children is devoted to the issue of fragile families: “Fragile Families”, Future of Children 20 (Fall 2010): 3-230. Also see Sara McLanahan, “Family Instability and Complexity after a Nonmarital Birth: Outcomes for Children in Fragile Families”, in Social Class and Changing Families in an Unequal America, eds. Marcia J. Carlson and Paula England (Stanford, CA: Stanford University Press, 2011), 108-133; Sara McLanahan and Irwin Garfinkel, “Fragile Families: Debates, Facts, and Solutions” in Marriage at the Crossroads: Law, Policy, and the Brave New World of Twenty-First-Century Families, eds. Marsha Garrison and Elizabeth S. Scott (Cambridge, UK: Cambridge University Press, 2012), 142-169; McLanahan and Percheski, “Family Structure and the Reproduction of Inequalities”, 257-276; Marcia J. Carlson, Sara S. McLanahan, and Jeanne Brooks-Gunn, “Coparenting and Nonresident Fathers’ Involvement with Young Children After a Nonmarital Birth”, Demography 45 (May 2008): 461-488; and Sara McLanahan, Laura Tach and Daniel Schneider, “The Causal Effects of Father Absence”, Annual Review of Sociology 39 (July 2013): 399-427.

¹⁷ Cherlin, The Marriage-Go-Round.

¹⁸ Figure 1 is drawn from McLanahan and Jacobsen, “Diverging Destinies Revisited”. “High” education represents mothers in the top quartile of the education distribution; “low” education category represents mothers in the bottom quartile. Greg J. Duncan, Ariel Kalil, and Kathleen M. Ziol-Guest (unpub. ms., October 2014) have recently shown that the class gap in maternal age at any birth has grown even more rapidly than the class gap in maternal age at first birth, so that Figure 1 understates the aggregate growth of the class gap in maternal age for all children. Moreover, they find that this class gap in maternal age at birth now contributes roughly as much to the overall opportunity gap as the class gap in family structure. 

¹⁹ Karen Guzzo and Krista K. Payne, “Intentions and Planning Status of Births: 2000-2010”, National Center for Family & Marriage Research, FP-12-24 (Bowling Green State University, 2012). See also S. Philip Morgan, “Thinking about Demographic Family Difference: Fertility Differentials in an Unequal Society” in Social Class and Changing Families in an Unequal America, eds. Marcia J. Carlson and Paula England (Stanford, CA: Stanford University Press, 2011), 50-67. Recent data show large and increasing differences by education and income in unintended fertility: Heather Boonstra et al., Abortion in Women’s Lives (New York: Guttmacher Institute, 2006); Laurence B. Finer and Stanley K. Henshaw, “Disparities in Rates of Unintended Pregnancy in the United States, 1994 and 2001”, Perspectives on Sexual and Reproductive Health 38 (2006): 90-96. 

²⁰ Kelly Musick et al., “Education Differences in Intended and Unintended Fertility”, Social Forces 88 (2009): 543-572; Finer and Henshaw, “Disparities in Rates of Unintended Pregnancy in the United States, 1994 and 2001”, 90-96; Paula England, Elizabeth Aura McClintock and Emily Fitzgibbons Shafer, “Birth Control Use and Early, Unintended Births: Evidence for a Class Gradient”, in Social Class and Changing Families in an Unequal America, eds. Marcia J. Carlson and Paula England (Stanford, CA: Stanford University Press, 2011), 21-49; and Sara McLanahan, “Family Instability and Complexity after a Nonmarital Birth”, 108-133.

²¹ More detailed research shows that the racial gap within classes has narrowed, while the class gap within races has widened.

²² Martin, “Growing Evidence for a ‘Divorce Divide’?”

²³ Zhenchao Qian, “Divergent Paths of American Families”, in Diversity and Disparities: America Enters a New Century, ed. John Logan (New York: Russell Sage Foundation 2014).

²⁴ Cherlin, “Demographic Trends in the United States: A Review of Research in the 2000s”, 408.

²⁵ Wendy D. Manning, “Trends in Cohabitation: Twenty Years of Change, 1972-2008”, National Center for Family & Marriage Research FP 10-07 (2010), accessed April 18, 2014.

²⁶ Kathryn Edin and Timothy Nelson, Doing the Best I Can: Fathering in the Inner City (Berkeley, CA: University of California Press, 2013), 40.

²⁷ McLanahan, “Family Instability and Complexity after a Nonmarital Birth”; Edin and Nelson, Doing the Best I Can; Kathryn Edin, Timothy Nelson and Joanna Reed, “Daddy, Baby; Momma Maybe: Low-Income Urban Fathers and the ‘Package Deal’ of Family Life”, in Social Class and Changing Families in an Unequal America, eds. Marcia J. Carlson and Paula England (Stanford, CA: Stanford University Press, 2011), 85-107; Karen Benjamin Guzzo, “New Partner, More Kids: Multiple-Partner Fertility in the United States”, ANNALS of the American Academy of Political and Social Science 654 (July 2014): 66-86.

²⁸ Laura Tach, Kathryn Edin, Hope Harvey, and Brielle Bryan, “The Family-Go-Round: Family Complexity and Father Involvement from a Father’s Perspective”, ANNALS of the American Academy of Political and Social Science July 2014 654: 169-184.

²⁹ McLanahan and Percheski, “Family Structure and the Reproduction of Inequalities”, 258-59.

³⁰ Contrary to much popular commentary in America, these recent trends have little or nothing to do with an increase in teen pregnancy, which, in fact, has been steadily and sharply declining among all races for more than twenty years, with very little effect on rates of non-marital births or child poverty or social mobility. Of all unwed births in America nowadays, more than three quarters are to post-teen adults, and that share is growing. “Children having children” may be a significant problem, but it is not the central challenge facing the working-class family in America. Finer and Henshaw, “Disparities in Rates of Unintended Pregnancy in the United States, 1994 and 2001”; Federal Interagency Forum on Child and Family Statistics, America’s Children: Key National Indicators of Well-Being, 2013, “Births to Unmarried Women”, accessed April 23, 2014, “Trends in Teen Pregnancy and Childbearing”, Office of Adolescent Health, U.S. Department of Health & Human Services, November 21, 2014, as consulted December 1, 2014, citing H Hamilton, B.E., Martin, J.A., Osterman, M.J.K., & Curtin, S.C. (2014). Births: Preliminary Data for 2013. Hyattsville, MD: National Center for Health Statistics. Retrieved November 14, 2014. Pamela J. Smock and Fiona Rose Greenland, “Diversity in Pathways to Parenthood: Patterns, Implications, and Emerging Research Directions”, Journal of Marriage and Family 72 (June 2010): 579; Furstenberg, “Fifty Years of Family Change”.

³¹ “Marriage Trends in Latin America: A Fact Sheet”, 5.

³² “Women with high levels of education are not only choosing cohabiting unions over marital unions but they are also postponing union formation and childbearing, whereas the least-educated women are choosing cohabitation more, but without postponement”. Esteve and Florez-Paredes, 48-49.

³³ “Marriage Trends in Latin America: A Fact Sheet”, 4.

³⁴ Esteve and Florez-Paredes, 60, emphasis added. Of course, this general judgment does not deny that many Latin American men have been and are devoted husbands and fathers.

³⁵ Mexico, Colombia, and Brazil.

³⁶ https://www.huffingtonpost.co.uk/2016/03/09/pregnancy-around-the-world-age-of-new-mums_n_9416064.html

³⁷ Esteve and Florez-Paredes, 47.

³⁸ Esteve and Florez-Paredes, 51-52.

³⁹ Institute of Medicine, From Neurons to Neighborhoods: The Science of Early Child Development, ed. Jack P. Shonkoff and Deborah A. Phillips (Washington, D.C.: National Academies Press, 2000). This section relies heavily on the excellent selection of working papers and issue briefs compiled at the Center on Developing Child at Harvard University, I am grateful to the Center’s founding director, Professor Jack P. Shonkoff, M.D., for guidance and encouragement, though I remain solely responsible for this summary of the field. Other key citations include Paul Tough, How Children Succeed: Grit, Curiosity, and the Hidden Power of Character (New York: Houghton Mifflin Harcourt, 2012); Gary W. Evans and Michelle A. Schamberg, “Childhood poverty, chronic stress, and adult working memory”, The Proceedings of the National Academy of Sciences 106 (April 21, 2009): 6545-6549; James J. Heckman, “Skill Formation and the Economics of Investing in Disadvantaged Children”, Science 30 (June 2006): 1900-1902; James J. Heckman, “An Effective Strategy for Promoting Social Mobility”, Boston Review (September/October 2012); Eric I. Knudsen, James J. Heckman, Judy L. Cameron and Jack P. Shonkoff, “Economic, Neurobiological, and Behavioral Perspectives on Building America’s Future Workforce”, The Proceedings of the National Academy of Sciences 103 (July 5, 2006): 10155-10162; Jack P. Shonkoff, Andrew S. Garner, The Committee on Psychosocial Aspects of Child and Family Health, Committee on Early Childhood, Adoption, and Dependent Care, and Section on Developmental and Behavioral Pediatrics, “The lifelong effects of early childhood adversity and toxic stress”, Pediatrics 129 (January 1, 2012): e232-246.

⁴⁰ National Scientific Council on the Developing Child, “Young Children Develop in an Environment of Relationships”, Center on the Developing Child Working Paper No. 1 (2004).

⁴¹ Tough, How Children Succeed; Walter Mischel, Yuichi Shoda and Monica Larrea Rodriguez, “Delay of Gratification in Children”, Science 244 (May 26, 1989): 933-938; Angela L. Duckworth and Martin E.P. Seligman, “Self-discipline Outdoes IQ in Predicting Academic Performance of Adolescents”, Psychological Science 16 (December 2005): 939-944; James J. Heckman, Jora Stixrud and Sergio Urzua, “The Effects of Cognitive and Noncognitive Abilities on Labor Market Outcomes and Social Behavior”, Journal of Labor Economics 24 (July 2006): 411-482; Flavio Cunha and James Heckman, “The Technology of Skill Formation”, American Economic Review 97 (May 2007): 31-47.

⁴² Center on the Developing Child, “Science of Neglect”, InBrief Series, Harvard University, 1, accessed May 7, 2014.

⁴³ Heckman, “An Effective Strategy for Promoting Social Mobility”.

⁴⁴ Poor kids (<200% FPL): 4% parent death; 11% parent imprisoned; 10% saw parental physical abuse; 12% saw neighborhood violence; 10% mentally ill family member; 13% alcohol/drug problem family member. Not-poor kids (>400% FPL): 2%; 2%; 3%; 4%; 6%; 6%. Data from “National Survey of Children’s Health”, Data Resource Center for Child and Adolescent Health, Child and Adolescent Health Measurement Initiative (2011/12).

⁴⁵ Gary W. Evans, “The Environment of Childhood Poverty”, American Psychologist 59 (February-March 2004): 77-92 and works cited there; Jamie L. Hanson et al., “Family Poverty Affects the Rate of Human Infant Brain Growth”, PLOS ONE 8 (December 2013) report that directly increasing the income of poor parents has measurable positive effects on children’s cognitive performance and social behavior, strongly suggesting that the link between social class and child development is causal, not spurious.

⁴⁶ Jane Waldfogel and Elizabeth Washbrook, “Income-Related Gaps in School Readiness in the United States and the United Kingdom” in Persistence, Privilege, and Parenting: The Comparative Study of Intergenerational Mobility, eds. Timothy M. Smeeding, Robert Erikson and Markus Jantti (NewYork: Russell Sage Foundation, 2011). Extracurricular involvement is discussed in Chapter 4.

⁴⁷ Jane Waldfogel, What Children Need (Cambridge: Harvard University Press, 2006), 161

⁴⁸ This chart is based on the annual DDB Needham Life Style surveys described in Robert D. Putnam, Bowling Alone: The Collapse and Revival of American Community (New York: Simon and Schuster, 2000), 420-424. The question was simply agree or disagree: “Our whole family usually eats dinner together”. Questions on family dinners have occasionally been asked in other surveys, such as the 2003 and 2007 National Surveys of Children’s Health, but only in a few years and only since 2000, so they are much less useful in detecting long-term trends. Figure 5 is limited to parents with children under 18 at home and weighted to account for differences in single-parent and two-parent families.

⁴⁹ Sabino Kornrich and Frank Furstenberg, “Investing in Children: Changes in Parental Spending on Children, 1972-2007”, Demography 50 (February 2013): 1-23; Neeraj Kaushal, Katherine Magnuson and Jane Waldfogel, “How is Family Income Related to Investments in Children’s Learning?” in Whither Opportunity? Rising Inequality, Schools and Children’s Life Chances, eds. Greg J. Duncan and Richard J. Murnane (NewYork: Russell Sage Foundation, 2011), 187-206.

⁵⁰ Rand D. Conger, Katherine J. Conger and Monica J. Martin, “Socioeconomic Status, Family Processes, and Individual Development”, Journal of Marriage and Family 72 (June 2010): 685-704, esp. 695.

⁵¹ Evrim Altintas, “Inequality in Mothers’ Time Investments in Children in the U.S. (1965-2013)”, (Nuffield College, 2014) is the source of Figure 7. Unlike prior work on this topic, the data in Figure 7 have been adjusted to account for the very low time investment in childcare by non-residential fathers; since a large and growing fraction of kids in lower-education households are being raised by single mothers, this adjustment has a substantial effect on the size and growth of the class gap. For earlier work on this topic, see Garey Ramey andValerieA. Ramey, “The Rug Rat Race”, Brookings Papers on Economic Activity Spring (Economic Studies Program, The Brookings Institution, 2010): 129-99; Meredith Phillips, “Parenting, Time Use, and Disparities in Academic Outcomes”, in Whither Opportunity? Rising Inequality, Schools and Children’s Life Chances, eds. Greg J. Duncan and Richard J. Murnane (New York: Russell Sage Foundation, 2011), 207-228; and Ariel Kalil, Rebecca Ryan and Michael Corey, “Diverging Destinies: Maternal Education and the Developmental Gradient in Time with Children”, Demography 49 (November 2012): 1361-1383. The latter show that the education gap is largest in childcare activities that are specifically important for child’s development at a particular age (play and basic care between age 0-2, teaching/talking/reading between age 3-5 and management/organizational activities between age 6-13). “Specifically, we found (as have others before us) an ‘education gradient’ in mothers’ use of time: in almost all cases, highly educated mothers spend more time than less-educated mothers in the broad categories of child time investments that promote development. However, we also identified a ‘developmental gradient’, such that highly educated mothers shift the composition of their time in ways that specifically promote children’s development at different developmental stages. Specifically, the education gradient in basic care and play is greatest when youngest children are infants and toddlers (0 to 2), which is precisely when children most require parents’ time on such basic activities as bathing and feeding and also precisely the age when parent–child play is at its most developmentally appropriate. The education gradient for teaching is greatest when youngest children are preschool aged (3 to 5), which is precisely when time spent in learning activities (such as reading and problem solving) best prepare children for school entry. Conversely, the education gradient in management is greatest when youngest children are between the ages of 6 and 13 – precisely the ages when parental management is a key, developmentally appropriate input”.

⁵² Crnic and Low, “Everyday Stresses and Parenting”, 243-268; Deater-Deckard, Parenting Stress, and sources cited there.

⁵³ This observation was made during a private meeting between the author, President and Mrs. Bush, and the President’s senior advisors in March 2007.

⁵⁴ Jaap Dronkers and Rolf van der Velden, “Positive but also Negative Effects of Ethnic Diversity in Schools on Educational Performance? An Empirical Test Using PISA data”, in Integration and Inequality in Educational Institutions, ed. Michael Windzio (Dordrecht: Springer, 2013): 71-98 and the works cited there.

⁵⁵ Useful entryways to the massive literature on this topic include James S. Coleman et al., Equality of Educational Opportunity (Washington, DC: U.S. Department of Health, Education &Welfare, Office of Education, OE-38001 and supplement, 1966), 325; Gary Orfield and Susan E. Eaton, Dismantling Desegregation (New York: The New Press, 1996); Claude S. Fischer et al., Inequality by Design: Cracking the Bell Curve Myth (Princeton, NJ: Princeton University Press, 1996); Richard D. Kahlenberg, “Economic School Integration”, in The End of Desegregation, eds. Stephen J. Caldas and Carl L. Bankston III (Hauppauge, NY: Nova Science, 2003), esp. 153-155; Russell W. Rumberger and Gregory J. Palardy, “Does Segregation Still Matter? The Impact of Student Composition on Academic Achievement in High School”, The Teachers College Record 107 (September 2005): 1999-2045; John R. Logan, Elisabeta Minca and Sinem Adar, “The Geography of Inequality: Why Separate Means Unequal in American Public Schools”, Sociology of Education 85 (July 2012): 287-301; and for a comprehensive recent overview, Gregory J. Palardy, “High School Socioeconomic Segregation and Student Attainment”, American Educational Research Journal 50 (August 2013): 714-754. Reyn van Ewijk and Peter Sleegers, “The Effect of Peer Socioeconomic Status on Student Achievement: A Meta-Analysis”, Educational Research Review 5 (June 2010): 134-150 found that the effect of the socioeconomic composition of a child’s classroom on his or her test scores is twice as large as the effect of the socioeconomic composition of his or her school.This entire line of research was stimulated in the 1960s by concerns about the effects of racial segregation, and in that era class segregation heavily overlapped with racial segregation. During the past half century, however, class segregation has grown, while racial segregation has diminished, and it is now possible to compare the adverse effects of racial and class segregation. While racial segregation continues to be a major national problem, virtually all relevant studies have concluded that class segregation is at least as pernicious in its effects on student achievement. See Richard D. Kahlenberg, “Socioeconomic School Integration”, North Carolina Law Review 85 (June 2007): 1545-1594.

⁵⁶ Sean F. Reardon and Kendra Bischoff, “Growth in the Residential Segregation of Families by Income, 1970-2009”, Technical report for US2010 Project (Russell Sage Foundation and American Communities Project of Brown University, 2011); https://urban.jrc.ec.europa.eu/thefutureofcities/social-segregation#the-chapter; https://www.lincolninst.edu/publications/articles/urban-spatial-segregation

⁵⁷ Estimates in this chart are drawn from “Family Income and Unequal Educational Opportunity, 1970 to 2011”, Postsecondary Education Opportunity 245 (November 2012). The basic trends shown in Figure 8 are broadly consistent with the results in Bailey and Dynarski, “Gains and Gaps: Changing Inequality in U.S. College Entry and Completion”, which are methodologically more reliable, but limited to two points in time (roughly 1982 and 2003). The estimates in Figure 8 probably overstate the level of college graduation among the kids from the richest quartile by about 10 percentage points, I use this chart because it gives a more continuous picture of the trends over time. (The chart also shows the equivalent Bailey/Dynarski data points as “open” dots). See also Patrick Wightman and Sheldon Danziger, “Poverty, Intergenerational Mobility, and Young Adult Educational Attainment”, in Investing in Children: Work, Education, and Social Policy in Two Rich Countries, eds. Ariel Kalil, Ron Haskins and Jenny Chesters (Washington, DC: Brookings Institution Press, 2012), 208-236.

⁵⁸ For recent accounts of these pendulum swings, see Robert D. Putnam, Bowling Alone: The Collapse and Revival of American Community (New York: Simon & Schuster, 2000); E.J. Dionne, Jr., Our Divided Political Heart: The Battle for the American Idea in an Age of Discontent (New York: Bloomsbury USA, 2012); and Robert D. Putnam, with Shaylyn Romney Garrett, The Upswing: How America Came Together a Century Ago and How We Can Do It Again (Simon & Schuster, 2021).

⁵⁹ For an introductory overview of this massive literature, see Putnam, Bowling Alone, 287-363.

⁶⁰ Peter V. Marsden, “Core Discussion Networks of Americans”, American Sociological Review 52 (February 1987): 122-131; Claude S. Fischer, To Dwell among Friends: Personal Networks inTown and City (Chicago: University of Chicago Press, 1982); Karen E. Campbell, PeterV. Marsden and Jeanne S. Hurlbert, “Social Resources and Socioeconomic Status”, Social Networks 8 (March 1986): 97-117; Marjolein I. Broese Van Groenou and Theo Van Tilburg, “Network Size and Support in Old Age: Differentials by Socio-Economic Status in Childhood and Adulthood”, Ageing and Society 23 (September 2003): 625-645; Ivaylo D. Petev, “The Association of Social Class and Lifestyles: Persistence in American Sociability, 1974 to 2010”, American Sociological Review 78 (August 2013): 633, 651.

⁶¹ The specific question in the Benchmark survey was “About how many close friends do you have these days? These are people you feel at ease with, can talk to about private matters, or call on for help”. This national survey included 30,000 respondents in 2000; for more details and access to the raw data, see http://www.ropercenter.uconn.edu/data_access/data/datasets/social_capital_community_survey.html See also Campbell, Marsden and Hurlbert, “Social Resources and Socioeconomic Status”, 97-117.

⁶² Putnam and Garrett, The Upswing, chap. 4; Petev, “The Association of Social Class and Lifestyles”, 633, 651.

⁶³ Eszter Hargittai and Amanda Hinnant, ”Digital Inequality: Differences in Young Adults’ Use of the Internet”, Communication Research 35 (October 2008): 602-621; Fred Rothbaum, Nancy Martland, Joanne Beswick Jannsen, “Parents’ Reliance on the Web to Find Information about Children and Families: Socio-Economic Differences in Use, Skills and Satisfaction”, Journal of Applied Developmental Psychology 29 (March-April 2008): 118-128; Eszter Hargittai andYuli Patrick Hsieh, “Digital Inequality”, in The Oxford Handbook of Internet Studies, ed. William H. Dutton (Oxford, UK: Oxford University Press, 2013), 129-150.

⁶⁴ Eszter Hargittai, “The Digital Reproduction of Inequality”, in Social Stratification, ed. David Grusky (Boulder, CO: Westview Press, forthcoming), 936-944.

⁶⁵ Evidence of the effects of mentoring can be found in Jean Baldwin Grossman and Joseph P. Tierney, “Does Mentoring Work?: An Impact Study of the Big Brothers Big Sisters Program”, Evaluation Review 22 (June 1998): 403-426; David L. DuBois , Bruce E. Holloway, Jeffrey C. Valentine and Harris Cooper, “Effectiveness of Mentoring Programs for Youth: A Meta-Analytic Review”, American Journal of Community Psychology 30 (April 2002): 157-197; David L. DuBois et al., “How Effective Are Mentoring Programs for Youth? A Systematic Assessment of the Evidence”, Psychological Science in the Public Interest 12 (August 2011): 57-91; Lance D. Erickson, Steve McDonald and Glen H. Elder, Jr., “Informal Mentors and Education: Complementary or Compensatory Resources?” Sociology of Education 82 (October 2009): 344-367. David L. DuBois and Naida Silverthorn, “Characteristics of Natural Mentoring Relationships and Adoles-cent Adjustment: Evidence from a National Study”, Journal of Primary Prevention 26 (2005): 69-92 report that informal mentoring led to improvements in a broad array of positive and negative adolescent behavior: completion of high school, college attendance, working 10+ hours a week, binge drinking, using drugs, smoking, gang memberships, fighting, risk-taking, self-esteem, life satisfaction, depression, suicidal thoughts, general health, general physical activity, having an STD, using birth control, and using condoms.

⁶⁶ Robert J. Sampson, Great American City: Chicago and the Enduring Neighborhood Effect (Chicago: University of Chicago Press, 2012), 356, emphasis in original. The study of neighborhood effects has been tormented by complicated methodological concerns, especially what is termed “selection bias”. Since people generally choose where to live, if people in a given neighborhood have distinctive characteristics, it is possible that they brought those traits with them to the neighborhood, rather than those traits being “caused” by the neighborhood context. The best contemporary studies, however, have been attuned to that risk, and our discussion here is based on findings that seem robust in the face of that methodological issue. In fact, cross-sectional studies may actually underestimate true neighborhood effects, by ignoring the impact of long-term effects. On these methodological issues, see Sampson, Great American City, especially chapters 12 and 15; Robert J. Sampson and Patrick Sharkey, “Neighborhood Selection and the Social Reproduction of Concentrated Racial Inequality”, Demography 45 (February 2008): 1-29; and Tama Leventhal, Véronique Dupéré and Elizabeth Shuey, “Children in Neighborhoods”, in Handbook of Child Psychology and Developmental Science, 7th edition, volume 4, eds. Richard M. Lerner, Marc H. Bornstein and Tama Leventhal (Hoboken, NJ: Wiley, forthcoming 2015). At the center of these debates is the so-called “Moving to Opportunity” experiment of the 1990s that followed a randomly selected group of poor families who were enabled to move to low poverty neighborhoods and then carefully compared to a control group of similar families who did not so move. For an overview of the complex and mixed results, see Jens Ludwig et al., “Neighborhood Effects on the Long-Term Well-Being of Low-Income Adults”, Science 337 (2012): 1505-1510, and Lisa Sanbonmatsu et al., Moving to Opportunity for Fair Housing Demonstration Program – Final Impacts Evaluation (Washington, DC: U.S. Department of Housing and Urban Development, 2011).

⁶⁷ A recent, comprehensive overview of neighborhood effects on children is Leventhal, Dupéré and Shuey, “Children in Neighborhoods”.

⁶⁸ Cynthia M. Duncan, Worlds Apart: Poverty and Politics in Rural America, 2nd ed. (New Haven:Yale University Press, 2014).

⁶⁹ Figure 12 depicts the simple correlation between trust and poverty, but the correlation remains robust and substantial with controls for personal finances, education, citizenship, ethnicity, crime rates, income inequality, ethnic diversity, language, commuting time, residential mobility, homeownership, gender, region, and age. See Robert D. Putnam, “E Pluribus Unum: Diversity and Community in the 21st Century: The 2006 Johan Skytte Prize Lecture”, Scandinavian Political Studies 30 (June 2007): 137-174, especially Table 3.The same pattern also applies to how often neighbors speak with one another.

⁷⁰ See Putnam, Bowling Alone, 138; and Orlando Patterson, “Liberty Against the Democratic State: On the Historical and Contemporary Sources of American Distrust”, in Democracy and Trust, ed. Mark E. Warren (Cambridge, UK: Cambridge University Press, 1999), 187-191.

⁷¹ Putnam, Bowling Alone; Wendy M. Rahn and John E. Transue, “Social Trust and Value Change:The Decline of Social Capital inAmericanYouth, 1976-1995”, Political Psychology 19 (September 1998): 545-565; April K. Clark, Michael Clark and Daniel Monzin, “Explaining Changing Trust Trends in America”, International Research Journal of Social Sciences 2 (January 2013): 7-13; Jean M. Twenge, W. Keith Campbell, and Nathan T. Carter, “Declines in Trust in Others and Confidence in Institutions Among American Adults and Late Adolescents, 1972-2012”, Psychological Science (October 2014) vol. 25 no. 10 1914-1923.

⁷² For present purposes I do not distinguish among different denominations since the generalizations at the center of my argument apply more or less equally to Catholics, Protestants, Jews, and other faith traditions, as we showed in Robert D. Putnam and David E. Campbell, American Grace: How Religion Divides and Unites Us (New York: Simon & Schuster, 2010).

⁷³ Putnam and Campbell, American Grace (2010), especially Chapter 13. Statistics in this paragraph come from the 2006 Faith Matters national survey, described in that book. All these correlations persist when controlling for self-selection, as well as other demographic factors.

⁷⁴ John M. Wallace and Tyrone A. Forman, “Religion’s Role in Promoting Health and Reducing Risk among American Youth”, Health Education and Behavior 25 (December 1998): 721-741; Mark D. Regnerus and Glen H. Elder, Jr., “Staying on Track in School: Religious Influences in High and Low-Risk Settings” (paper presented at the annual meeting of the American Sociological Association, Anaheim, CA, August 2001); Chandra Muller and Christopher G. Ellison, “Religious Involvement, Social Capital, and Adolescents’ Academic Progress: Evidence from the National Education Longitudinal Study of 1988”, Sociological Focus 34 (May 2001): 155-183; Christian Smith and Robert Faris, “Religion and American Adolescent Delinquency, Risk Behaviors, and Constructive Social Activities”, research report of the National Study of Youth and Religion (Chapel Hill, NC, 2002), accessed August 21, 2014; Jonathan K. Zaff, Kristin A. Moore, Angela Romano Papillo and Stephanie Williams, “Implications of Extracurricular Activity Participation during Adolescence on Positive Outcomes”, Journal of Adolescent Research 18 (November 2003): 614; Jennifer L. Glanville, David Sikkink and Edwin I. Hernandez, “Religious Involvement and Educational Outcomes: The Role of Social Capital and Extracurricular Participation” Sociological Quarterly 49 (Winter 2008): 105-137. These studies control for many other factors that might make the correlations spurious. The best studies of selection bias in the case of religious engagement conclude that, if anything, this bias tends to obscure, not exaggerate, the effects of religion: Mark D. Regnerus and Christian Smith, “Selection Effects in Studies of Religious Influence”, Review of Religious Research 47 (September 2005): 23-50; Jonathan H. Gruber, “Religious Market Structure, Religious Participation, and Outcomes: Is Religion Good for You?” Advances in Economic Analysis & Policy 5 (December 2005).

⁷⁵ Eric Dearing et al., “Do Neighborhood and Home Contexts Help Explain Why Low-Income Children Miss Opportunities to Participate in Activities Outside of School?”, Developmental Psychology 45 (November 2009):1545-1562. Author’s analysis of Social Capital Community Benchmark Survey (2000); out of seventeen different types of organizations, only self-help, veterans, and seniors groups are less class biased in their membership than religious groups.

⁷⁶ Putnam and Campbell, American Grace, 252-253. The same generational trend of increasing class bias in church attendance appears in the General Social Survey, the National Educational Studies, and in the Roper Political and Social Trends archive, with either education (relative or absolute) or income as a measure of socioeconomic status, though more clearly with education. Attendance measures differ from archive to archive, but the trends by education are similar. The growth of the class gap is sharper for men than for women, and if anything, sharper among blacks than among whites and among evangelical Protestants than among other traditions. If all races are analyzed together, this trend is masked, because nonwhites are poorer, less educated, and more religious, but the growing class gap appears in each race, considered separately.

⁷⁷ England, McClintock and Shafer, “Birth Control Use and Early, Unintended Births: Evidence for a Class Gradient”.

⁷⁸ Jennifer Glass and Philip Levchak, “Red States, Blue States, and Divorce: Understanding the Impact of Conservative Protestantism on Regional Variation in Divorce Rates”, American Journal of Sociology 119 (January 2014): 1002-1046.