A Demographic Perspective on Family Change

Abstract

Demographic analysis seeks to understand how individual microlevel decisions about child-bearing, marriage and partnering, geographic mobility, and behaviors that influence health and longevity aggregate to macrolevel population trends and differentials in fertility, mortality and migration. In this review, I first discuss theoretical perspectives—classic demographic transition theory, the perspective of the “second demographic transition,” the spread of developmental idealism—that inform demographers’ understanding of macrolevel population change. Then, I turn to a discussion of the role that demographically informed data collection has played in illuminating family change since the mid-20th century in the United States. Finally, I discuss ways in which demographic theory and data collection might inform future areas of family research, particularly in the area of intergenerational family relationships and new and emerging family forms.

The basic “project” of demography is to understand how individual (or couple) microlevel decisions about childbearing, marriage and partnering, geographic mobility, and behaviors that influence health and longevity aggregate up to macrolevel population trends and differentials in fertility, mortality, and migration. That is, how do microlevel behaviors ultimately influence the composition or character of a population or society at a given point in time and change in that population over time? Understanding intergenerational mechanisms of transmission, many involving the family, is critical to understanding population renewal and societal well-being and inequality (Goldscheider, 1995; Mare, 2011; Sweet & Bumpass, 1987).

In this review, I discuss how demographers think about family change at both the macrosocietal and microindividual, or household, level. First, demographers have developed theories about drivers of macrolevel population change, including economic development, ideational change, institutional shifts, and technological change, which also help us understand the changing context for decisions about family formation, dissolution, and intergenerational relationships. Second, because description is the bedrock of demographic analysis, family demographers have designed data collections that allow for the assessment of trends and the components of change in marriage, fertility, and other family behaviors. Family demographers focus on transitions—into marriage, into parenthood, out of marriage, out of the parental home, and so on—and offer a springboard for, or at least provide a backdrop to, broader studies of the timing, patterning, causes, and consequences of family behaviors across the life course.

Mortality and Fertility Change and the Theory of the Demographic Transition

The core of demographic analysis focuses on three processes that alter a population’s size and age and sex structure: fertility, mortality, and migration. To these one could add nuptiality, or marriage patterns, as a key process with a close connection to fertility change. To understand the demographic perspective, it is useful to begin with classic demographic transition theory, which argues that populations typically move through three stages. During the first stage, societies are characterized by high levels of mortality and fertility and have either a stable population size or low rates of population growth. If high levels of mortality are not counterbalanced by high levels of fertility, populations die out. In this first phase of the transition, economic organization is largely structured around the family, with the family’s survival essential to the long-term functioning of society (Coale, 1973). A recent study of the fertility response to high mortality under the Khmer Rouge in Cambodia is consistent with the notion that an upswing in mortality is perceived as a threat to population survival and is answered with a dramatic increase in fertility. Once the terror was over in Cambodia, marriages formed quickly (Heuveline & Poch, 2006), and a baby boom followed (Heuveline & Poch, 2007).

During the second stage of the demographic transition, improvements in technology and economic productivity raise living standards and mortality declines. When mortality declines, it typically begins with control of infectious diseases, which tends to benefit the young more than the old. There is a drop in infant and child mortality and an acceleration in population growth, and populations grow younger in the early stages of mortality decline because of increased child survival. Declines in fertility typically lag behind declines in mortality because social norms and values are in place that support high levels of fertility. Until norms and values recalibrate, there is a period of rapid population growth.

The third stage of the transition is marked by declines in fertility, as smaller families come to be defined as “a good thing” in the larger society and, at the individual or household level, as parents perceive an increasing economic advantage to limiting family size (Coale, 1973; Kirk, 1996; Lam 2011). Individual fertility decisions, repeated across households, aggregate up to a large-scale drop in fertility and an end to rapid population growth.

Across the demographic transition, the context of family life changes dramatically, with implications for intergenerational relationships and support, the duration of marriage, and the number of years spent in active parenting. At the beginning of the demographic transition, women have many pregnancies and spend their short adult life span having and caring for children—with many of those children dying in infancy and childhood. Because mortality is high, families are not large. By the end of the transition, women have fewer children, they live longer, and they almost always survive their childbearing years, and thus, they spend a much smaller proportion of their adult years having or caring for children. A focus on the fertility transition leads quite directly to an emphasis on the changing status and time allocation of adults, especially women.

In the West, this transition—from high mortality and fertility to low mortality and fertility—took place over two centuries. In the rest of the world, this transition is happening very quickly—often in just a generation or two, with a few decades, not centuries, for families to adjust to new lower mortality regimes. Arland Thornton (2005) claimed that the widespread adoption of family-planning policies also spread what he called developmental idealism, or the idea that low fertility was associated with progress, that having fewer children caused socioeconomic improvements, and that “modern” families were small families. His subsequent research has shown that these ideas are quite widespread across diverse countries (e.g., Argentina, Iran, Egypt, China, Nepal, the United States). The idea that progressive or modern behavior and socioeconomic advancement requires limiting family size is espoused by the majority of ordinary people in these countries (Thornton, Yount, Abbasi-Shavazi, Ghimire, & Xie, 2012).

In the mid-20th century, high fertility and rapid population growth were the overwhelming preoccupation of population theorists. In the 1960s, the overriding concern became the impending “population bomb” (Ehrlich, 1968). Indeed, as explained by David Lam (2011) in his presidential address to the Population Association of America, an unprecedented doubling of the world population (from 3 billion to 6 billion) occurred between 1960 and 1999.

Today, the major issue in the field of population studies is shifting toward population aging, especially under conditions of low fertility. By 2000, fertility was declining throughout the world, even in many of the highest-fertility countries of sub-Saharan Africa. Lam (2011) argued that there was a massive transformation in investment in children—away from having many children (quantity) under a regime of high (infant) mortality to heavily investing in fewer, surviving children (quality) in today’s world of much lower (infant) mortality.

A swift transition from high to low fertility often leaves the family support expectations of the older generation, raised under one regime, at odds with those of the younger generation, who may have new and different expectations about what they “owe” parents or what parents “owe” them. Currently, there is especially great interest in changing intergenerational support regimes in countries with strong institutions of filial piety and favoritism toward (oldest) sons, such as China (Xie & Zhu, 2009).

The relative size of dependent populations (the young and the old) and the population on which they depend (the working-age population) influences the “transfer balance” between the generations. Many developing countries are currently in “a window of opportunity”—often referred to as the period of the “demographic dividend” (Lee, 2007)—because they have large working-age populations (a result of earlier high fertility), relatively few children because of reduced fertility in the past decade or two, and a small (but growing) elderly population, thus making for a ratio of dependents to workers that is quite favorable for economic growth. However, the issue of rapid population aging looms on the horizon. It is already a major issue in Europe and Japan and will become a growing concern in developing countries as fertility declines worldwide.

The Second Demographic Transition

Some have argued that the demographic changes since the 1960s, particularly in Europe, warrant the label “second demographic transition” (Lesthaeghe, 1983, 1995; Lesthaeghe & Surkyn, 1988; Van de Kaa, 1987). In many European countries—and increasingly in Asian countries as well—fertility rates are well below replacement levels, which suggests a population decline in the future (Morgan & Taylor, 2006). The theory of the second demographic transition focuses on what competes with marriage and children and leads to less of both.

Like the first demographic transition, Lesthaeghe (1995) described stages in the second demographic transition. Between 1955 and 1970, divorce increased, fertility declined to low levels, and marriages were delayed. This was followed by a period in the late 1970s in Europe when premarital cohabitation and births to unmarried women increased. From the mid-1980s onward, divorce rates flattened out but cohabitation following divorce increasingly delayed or replaced remarriage. In some countries (most often in Northern Europe), delayed fertility at younger ages was recouped after age 30, resulting in near-replacement-level fertility (e.g., in Sweden). In other places (especially Southern and Eastern Europe), fertility declined to very low levels.

Second-demographic-transition theory gives considerable weight to ideational change: Whereas the motivation for the first demographic transition was linked to changes in the value of children, in the second demographic transition the goals change to self-actualization and individual adult fulfillment as institutions that regulate family behavior are weakened. An increase in both individualism and a focus on the quality of adult relationships places high demands on the institution of marriage—which is entered far more cautiously and exited quickly when it does not meet the individual needs of the partners. Greater opportunities for women and growing gender equality also encourage those trends. Institutions that once had authority over and regulated family behaviors (e.g., the extended family, religious institutions) no longer have the power to influence individuals’ family decisions. Low or below-replacement fertility is also viewed as a function of increasing individual autonomy and self-fulfillment and as the result of the move toward gender symmetry in adult work and family roles. There is no reason to expect that fertility decline will end at replacement levels, as assumed in the (first) demographic-transition theory. It may fall even lower—indeed, it has—in many parts of the world.

Macrolevel theories of demographic transition form the backdrop for thinking about large-scale shifts in family composition and behavior. Much of the theorizing in family demography assumes, at least implicitly, that decisions—about when and whom to marry, whether to have children and how many, whether to remain in a marriage and so forth—are the outcome of rational calculation of the costs and benefits of alternative courses of action. Family demographic analysis of decision making has been heavily influenced by neoclassical economic theory of fertility (Becker, 1960), marriage formation (Becker 1974, 1981), marital dissolution (Becker, Landes, & Michael 1977), and marriage markets (Becker 1974, 1981; Oppenheimer, 1988).

The Demographic Perspective and Family Change in the United States, 1950–Present

Theoretical perspectives in demography are developed to explain “demographic facts” about family change. One of the roles that family demographers have played in the field is to design and collect the data used to describe some of the key changes in American family life. As in Europe, the United States has experienced a postponing of first marriage; increases in divorce (at least until 1980); more nonmarital childbearing and cohabitation; a decline in fertility; the aging of the population; and increases in women’s labor force participation, resulting in a shift in the household division of labor. The United States appears to have unique features too: more family instability and a strong social-class gradient in family behaviors, especially with respect to marriage and childbearing (Cherlin, 2009, 2010). Family demographers have provided much of this picture.

Developing theory is a process, often begun when we observe something we do not fully understand. Bengtson, Acock, Allen, Dilworth-Anderson, and Klein (2005) suggested, “Theory is the attempt to move beyond the what of our observations … to the questions of why and how what we have observed or examined has occurred” (p. 5). Demographers often engage in documenting the “what”—the rates of marriage formation, the divorce rate, fertility rates, and so on (Casper & Bianchi, 2002). This rich description can be thought of as the first and essential step in developing theory.

One could argue that the pioneering family demographer for understanding and interpreting US family change, at least change in the middle of the 20th century, was Paul Glick of the U.S. Census Bureau. With the advent of the household-based monthly Current Population Survey (CPS) in the late 1940s and subsequent introduction of yearly supplements to the CPS, such as the annual March supplement on income and living arrangements and the June supplement on fertility, it became possible to track U.S. marriage, fertility, and household formation and dissolution trends in a way that had not been feasible in the earlier half of the 20th century (when many of these changes could be assessed only on a 10-year cycle with the Census of Population).

In his descriptions of American households and families, Glick (1947) was guided by his theory of the family life cycle. In the middle of the 20th century, the vast majority of individuals in the United States married by their early 20s, bore children soon after, raised those children to adulthood, and then entered an “empty nest” phase until one partner died and the other was widowed. The concept of the family life cycle had a relatively good fit with the empirical evidence in the 1950s and 1960s, and the new data from the CPS allowed the tracking of baby-boom households as they expanded. However, with the large-scale social changes in the later 1960s and 1970s—as family formation began to be delayed and family dissolution increased—new description of the rise in divorce and the increase in cohabitation was needed. The theory of the family life cycle described a smaller and smaller share of adult life courses, and the data structures became less able to capture emerging family trends.

Paul Glick and Arthur Norton attempted to keep up with emerging trends in cohabitation by developing the idea and measurement of the POSSLQ—persons of the opposite sex sharing living quarters (Glick & Norton, 1977). But it was demographers at the University of Wisconsin, Larry Bumpass and James Sweet, who led the next wave of data collection that allowed for innovation in family demographic research. The National Survey of Families and Households, first fielded in 1987–1988, with a second wave in 1992–1994, allowed the study of cohabitation to blossom and ultimately fed back in important ways to official government statistical data collections. By the early 1990s, few could deny the importance of changing family behaviors in the area of marriage and family formation, and data collections such as the census, the CPS, the National Survey of Family Growth, and others were propelled into capturing cohabitation along with marriage.

Andrew Cherlin (2010), in his decade review of demographic trends for Journal of Marriage and Family, suggested that we may again be at a crossroads where new data collection will be necessary to advance both description of family change and theory about the family. A host of recent changes in the family, he argued, increasingly disconnect family membership from household living arrangements. As most of our ongoing, large-scale surveys for tracking family demographic change are household based—that is, they rely on sampling frames for which the housing unit or household is the basic unit for data collection—it is becoming more and more difficult to capture the family relationships of greatest theoretical interest. This is not the only issue Cherlin identified; he also was critical of our ability to capture the timing of certain types of family events, given changes in the way partnerships are formed. A number of existing longitudinal data collections shed light on these family processes and the challenges they pose for theory and existing data collection efforts.

Marriage and Cohabitation

Traditionally, the percentage of the population who married was a key indicator of risk for childbearing. Today a decoupling of marriage and fertility is under way, not only in Europe but also elsewhere. Changes in intimate unions include a dramatic delay in entering into (legal) marriage, the rise in unmarried (heterosexual and same-sex) cohabitation, and the decline in marriage in many countries and among some population subgroups in the United States. The median age at first marriage rose to 28 years for men and 26 years for women in the United States (U.S. Census Bureau, 2010). The experience of living as an unmarried partner before marrying has rapidly become the modal experience for younger cohorts in the United States. For marriages formed between 1997 and 2001, 62% were preceded by cohabitation (Kennedy & Bumpass, 2008). Cohabitation is also increasingly an alternative to remarriage after separation and divorce, and in several European countries, it is often a long-term substitute to marriage. Cohabitation, like many family behaviors, also has a strong social-class gradient in the United States. According to analysis of women ages 19 to 44 in the 2002 National Survey of Family Growth (Kennedy & Bumpass, 2008), two thirds of those with a high school degree or less have ever cohabited, compared to a considerably lower 45% of those with a college degree.

Two other features of cohabitation in the United States have become apparent in the work of demographers who study this phenomenon. Children are increasingly present in cohabiting unions. Around 40% of births are classified as nonmarital (Hamilton, Martin, & Ventura, 2009), but about half of the nonmarital births are to mothers who are cohabiting with the father of the child (Kennedy & Bumpass, 2008). It seems misleading to think of these as births to single mothers. Yet we have also learned from the descriptive work of demographers that in the United States, unlike in European countries, such as France or the Scandinavian countries, long cohabitations are still uncommon. Children born to cohabiting partners in the United States have a very high probability of experiencing the breakup of the cohabiting union and ending up in a single-parent family within a few years of birth. There is a much greater likelihood that children potentially lose paternal investments (of time and money) when partnerships are never formalized by marriage. In the Fragile Families Study, which tracks urban parents and children from birth, among the sample of parents who were married at the time of the child’s birth, 77% remained married by the time the child was 5 years old. Among mothers who were not married to the child’s father at the time of birth, only 36% lived with the biological father 5 years later (Carlson & McLanahan, 2010).

Another feature of cohabitation in the United States, discovered by demographers who had assessed trends in cohabitation but wanted to “get behind the numbers” with qualitative, in-depth interviews, is that cohabitors often drift into this living arrangement (Manning & Smock, 2005). Particularly among young adults, couples rarely have a “grand plan” whereby they will live together for a time and then proceed to formalize their relationship with marriage, at least not when they first start living together (Sassler, 2004). Although individuals have perhaps always had some difficulty in pinpointing the start of their relationship, or determining exactly when they became “committed” to each other, the date of a marriage is fixed and can probably be reported with a fair amount of accuracy. Moving in together is sometimes more ambiguous, more a process, without a legally defined starting point. This is part of the reason Andrew Cherlin is skeptical that we can track the exact timing of union formation today as easily as in the past, because cohabitation accounts for a much larger share of coresidential, intimate relationships.

European family demographers and sociologists have begun exploring a type of relationship that is even more ambiguous, hard-to-define, and difficult to empirically study—what has come to be called living-apart-together (LAT) relationships. At least one study in the United States has attempted to measure this phenomenon, using nationally representative data from the General Social Survey collected in the late 1990s. Respondents were asked whether they had a steady romantic partner with a follow-up question on whether they lived together: Of women and men older than age 22, 7% and 6%, respectively, said they did have such a partner with whom they did not share a residence (Strohm, Seltzer, Cochran, & Mays, 2009).

Why have relationships that involve marriage decreased so much? Researchers have offered several explanations, including increased uncertainty about the stability of marriage, the erosion of norms against cohabitation and sexual relations outside of marriage, the wider availability of reliable birth control, economic changes, and increased individualism and secularization. Influenced by second-demographic-transition theory, one account focuses on changes in values and attitudes. Young men and women may have become more individualistic and less willing to sacrifice for the good of others, even their children, and therefore be less likely to marry or remain in an unsatisfying union than in the past.

A second account focuses on women’s increased opportunities and the declining returns to marriage for women. The neoclassical economic perspective argues that women’s increased education and rising employment increase the likelihood of divorce and reduce fertility because the value of women’s time spent in the labor market increases as does the “opportunity cost” (e.g., foregone wages) of spending time raising children. Women’s economic independence, it is argued, contributes to divorce because it erodes specialization in marriage and reduces the “gains from trade” in marriage (Becker, 1991; Becker et al., 1997). When a women’s “comparative advantage” in the home is high, the couple may allocate more of her time to the home and his to the market and both have much to gain in marriage and much to lose upon divorce. This division of labor makes less sense today as women’s labor-market opportunities and investments in their own human capital increase and become more similar to men’s opportunities and investments.

A third account focuses on the deteriorating economic circumstances of young adults, particularly men with only a high school degree or less education. Recent changes in the economy have adversely affected the economic opportunities available to young men. Difficulty finding a good “career” job increases the time it takes for young men in recent cohorts to get established in the labor market. Without stable jobs, men may be reluctant to marry, and women may perceive themselves as better off not married to such men (Oppenheimer, 1997, 2003). This increased uncertainty as to how good an economic provider men will eventually be extends the period between reaching adulthood and entering a first marriage (Oppenheimer, 1988).

(Low) Fertility

A modern fertility regime—the regime that many countries have either achieved or seem to be trending toward—includes a number of commonalities, according to Morgan (1996). Women tend to have only one or two children, sometimes three, and have those children sometime between their late teens and late 30s (with the average age at first birth shifting upward). There are relatively large proportions of births to unmarried mothers (although this has been less characteristic of Asia than Europe and, hence, is not universal). There tends to be substantial racial, ethnic, or socioeconomic variation in childbearing patterns. Period conditions affect fertility behavior, especially first-birth timing, with poor economic conditions favoring postponement of marriage and childbearing.

Today, the overriding fertility concern is quickly becoming very low fertility (labeled “lowest low fertility”), total fertility rates below an average of 1.3 children per woman in countries of Southern and Eastern Europe and parts of Asia (Billari & Kohler, 2004). Indications, at least in Europe, are that fertility levels may have reached their lowest level and be rebounding somewhat. In all countries in Europe (except Moldova), levels are now greater than 1.3 (Goldstein, Sobotka, & Jasilioniene, 2009), although sometimes not much above that level. Most of the countries with such low fertility levels also have very low mortality levels and are experiencing rapid population aging, imminent population decline, and a high percentage of the population in older age. They will have difficulty supporting their generous packages of old-age assistance in the future without an increase in fertility, substantial immigration, or both. This turn of events makes mortality and population aging and (international) migration processes central to current demographic analysis of family change.

What has driven fertility to such low levels? Current rates may be artificially low because the timing of fertility is changing in Europe and elsewhere. When the timing of childbearing is shifting to older ages, there is a period of very low fertility when older cohorts have completed their childbearing and younger cohorts are delaying the start of their childbearing. Fertility rises as these younger cohorts eventually have their children but at older ages. Fertility rates are very low in Southern Europe but have rebounded in the North, in Scandinavia and France. One hypothesis is that gender roles have changed to be more egalitarian in Northern Europe and that this is fertility enhancing. In Southern Europe, the family system in which women shoulder virtually all domestic tasks is incompatible with the increase in women’s labor-force participation. Hence, women are postponing or foregoing motherhood (McDonald, 2000).

Social-Class Divergence in U.S. Fertility Trends

Because fertility in the United States is around replacement level—and has been at that level since the late 1970s—the concerns about low levels of fertility are not as great an issue in the United States as they are in Europe. In the United States, there is far more concern about social-class differences in fertility, in particular the timing of childbearing relative to marriage. On the one hand, childbearing is increasingly delayed among the well educated, and childlessness is on the increase (Biddlecom & Martin, 2006). Currently in the United States, among women age 40 to 44, 20% have never had a child, double the percentage 30 years ago (Dye, 2008). Among the well educated, couples tend to marry before having children, and thus most children of college-educated parents are born to two married parents. On the other hand, among the less educated, childbearing often occurs relatively earlier in life, is frequently unplanned, and takes place outside marriage. Marriage is delayed or forgone but children are not, which results in a large proportion of births to unmarried women.

Ellwood and Jencks (2004) showed that there has been very little increase in nonmarital child-bearing among highly educated U.S. women over the past 40 years, whereas nonmarital child-bearing has increased quite significantly among those who have a high school education or less. In addition, nonmarital childbearing varies by race and ethnicity. About 28% of White children are born to unmarried parents, compared to 72% of African American children and 51% of Hispanic children (Hamilton et al., 2009).

The high percentage of births outside marriage draws attention to a phenomenon called multipartner fertility, or the likelihood that a woman has children by more than one partner (and vice versa, the likelihood that a man fathers children with more than one woman). Although not a “new” phenomenon—divorced fathers have frequently had children by a new partner—the current concern in the United States stems from the fragility of unmarried cohabitation, the high rate of nonmarital fertility, and the much greater likelihood that children potentially lose paternal investment (of time and money) when partnerships are never formalized by marriage.

Compared with their European counterparts, American women spend more time as lone mothers rearing children without a father present (Cherlin, 2009). Current estimates suggest that single parents account for about one quarter of U.S. households with children younger than age 18 (Kreider & Elliott, 2009), with 85% headed by a single mother and 15% by a single father. Single mothers in the United States generally face much greater economic strains than other family types. Because women earn less than men, and single mothers are typically younger and less educated than other mothers, single mothers are particularly vulnerable economically. Children of never-married single mothers have less than one fourth the family income than children in two-parent families, and the average incomes of families headed by divorced mothers is also less than half that of two parents (Casper & Bianchi, 2002). Mothers who never marry the father of their children are less likely to receive child support (with about 20% reporting receiving it regularly) than are divorced or separated mothers (with about 60% reporting receiving some support) (Casper & Bianchi, 2002).

Conclusion

This brief overview of family demographic trends, review of classic demographic transition theory, and discussion of the theory of the second demographic transition helps to understand the very low levels of fertility and the movement away from marriage, particularly in Europe but also in the United States. Are descriptions of family demographic change and the ideas embodied in these theoretical perspectives important to family researchers more generally? I would argue that they are for the following reasons. First, the emphasis on the interplay of mortality and fertility decline in classic demographic transition theory is important for understanding changing intergenerational relationships in Western countries. The changing age structure and attendant changes in ties of obligation between the generations, currently a central focus in population studies, will likely become a more dominant focus in family studies in the coming decades. These issues were highlighted by Bengtson (2001), when he argued that the multigenerational family was changing shape and developing a beanpole structure as multiple generations cosurvived. He anticipated that reliance on assistance through intergenerational ties might well surpass reliance on marriage ties for an increasing segment of the population and at particular points in the life course—among young adults slow to marry, middle-aged single parents without adequate support from a partner, older adults needing assistance in late life from adult children.

Second, demographic theory about the linkages between fertility and mortality change also suggests that we cannot divorce or compartmentalize the care needs of the elderly and children. Families provide both, and there is growing interest in whether individuals in modern low-fertility and low-mortality settings face increased risks of being “sandwiched” between the needs of (slow-to-launch) adult children and elderly parents who live longer than in the past. Declining fertility and the beanpole structure of families mean fewer adult children to share care of an elderly parent. At the same time, older adults cosurvive and become a source of potential support for grandchildren to a far greater extent than in earlier times. Swartz (2009) noted in her review of intergenerational relationships that it is more common for a child’s grandmother to be alive today than it was for the child’s mother to be alive a century ago.

Third, increased marital disruption and nonmarriage—trends that are the focus of second-demographic-transition theory about ideational change and increased individualism—lead to hypotheses about the strength of kinship ties across the life course. If second-demographic-transition theory claims are correct, not only are marriages more fragile but also extended kin in many families may well feel less obligation to one another than in the past. The “ties that bind” weaken under a regime of increased family disruption. Changes in marriage, divorce, and childbearing complicate the intergenerational picture, as financial and care obligations no longer necessarily depend on biological or marital ties. In groups where marriage is increasingly fragile, intergenerational ties may become more important than nuclear family ties to the rearing and well-being of children (Bengtson, 2001). But how strong are those ties? For example, a substantial increase in the percentage of children who grow up in a stepfamily increases the number of persons with whom the typical individual has some familial connection but, at the same time, may weaken norms of obligation (Coleman & Ganong, 2008). How these changes affect familial support over the life course will likely become a burgeoning area of family research. In addition, the rapid pace of fertility change in Southeast Asia, China, and other parts of the developing world offers the possibility of comparative study that may help us better understand U.S. families. Family norms of obligation in the Asian context are quite different from those in the West: Obligations to paternal kin often trump obligations to maternal kin, unlike in the United States, where kin ties tend to have a maternal bias. The swift demographic transition in the developing world offers leverage for deepening our understanding of family outcomes in diverse contexts.

Fourth, the description of family change that has been undertaken by demographers, originating from theoretical constructs like the family life cycle that may no longer fit, push all family scholars to rethink our constructs and invest in new, enhanced, and more dynamic data collection. Of necessity, we will need to push designs that are less tied to households if we are to understand family ties that typically span households. The study of new family forms would benefit from collaboration between family scholars who are expert in relationship quality and family demographers who track tends in family structures and living arrangements. Demographic description of the “what” of newer family forms opens the door for new interdisciplinary work on new questions that are arising. For example, do types of “loose” romantic connections (e.g., cohabitation, LATs) fit some stages in the life course better than others? What are the legal implications when family ties and obligations are less formalized? Are gender inequalities exacerbated? Are children left unprotected? Do men pay a price later in life for their more tenuous connections to their children earlier in life?

In conclusion, what do I think that a better understanding of the family demographic perspective might contribute to family research and theory more broadly? I believe that the evidence that family demographers have provided about family change—the postponement of marriage, increase in cohabitation, decline in fertility, decoupling of marriage and fertility, and increase in the complexity of intergenerational ties—forms a bedrock upon which family scholars can build. I also hope that an understanding of the theory of demographic transition might illuminate hypotheses to be tested about the changing nature of family and kinship ties across generations and about the consequences of newer family forms for individual health and well-being. A worthy goal is to open new areas of exploration and collaboration among family scholars and to provide one more valuable lens with which to view and understand the 21st-century demographic and social changes that are transforming family life.