4
The Role of the Family in the Production of Human Capital

The preceding chapter focused on household production of tangible goods and services, mainly items produced for immediate consumption (e.g., meals, clean clothing, basic care of children) but also a limited set of items that represent an investment for the future (e.g., the new roof a homeowner installs on the household dwelling). These tangible items are not, however, the only or even necessarily the most important outputs attributable to the activities of the household. This chapter is concerned with household investments that augment the human capital of household members, especially the investments of time that parents and other family members devote to the care and nurturing of children.

CONCEPTUAL FRAMEWORK

No other outputs are as quintessentially nonmarket as the production and care of children. While parents may purchase assistance with the care of their children, market care cannot fully substitute for their personal attention. Some observers appear to view family care as akin to sunlight, available without cost in effectively unlimited supply and thus of no economic interest. Given recent societal trends, however, it may not in fact be safe to assume the availability of family care. Since the 1960s, there has been increased participation of women in the labor force, higher divorce rates, and growing numbers of children who do not live in traditional nuclear families. To the extent that these trends may jeopardize the availability of the family care taken for granted in the past, they suggest the importance of better understanding how family care affects society’s production of human capital. Such understanding would help to inform public policy in a



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 79
Beyond the Market: Designing Nonmarket Accounts for the United States 4 The Role of the Family in the Production of Human Capital The preceding chapter focused on household production of tangible goods and services, mainly items produced for immediate consumption (e.g., meals, clean clothing, basic care of children) but also a limited set of items that represent an investment for the future (e.g., the new roof a homeowner installs on the household dwelling). These tangible items are not, however, the only or even necessarily the most important outputs attributable to the activities of the household. This chapter is concerned with household investments that augment the human capital of household members, especially the investments of time that parents and other family members devote to the care and nurturing of children. CONCEPTUAL FRAMEWORK No other outputs are as quintessentially nonmarket as the production and care of children. While parents may purchase assistance with the care of their children, market care cannot fully substitute for their personal attention. Some observers appear to view family care as akin to sunlight, available without cost in effectively unlimited supply and thus of no economic interest. Given recent societal trends, however, it may not in fact be safe to assume the availability of family care. Since the 1960s, there has been increased participation of women in the labor force, higher divorce rates, and growing numbers of children who do not live in traditional nuclear families. To the extent that these trends may jeopardize the availability of the family care taken for granted in the past, they suggest the importance of better understanding how family care affects society’s production of human capital. Such understanding would help to inform public policy in a

OCR for page 79
Beyond the Market: Designing Nonmarket Accounts for the United States variety of areas, such as the allocation of public spending to family programs. Concern about these trends has created particular anxiety about possible effects on young children and, in turn, a resurgence of interest in how family care, especially inputs of parental time, affects child outcomes (see Ginther and Pollak, 2004, on the effect of family structure on children’s educational outcomes). We do not mean to suggest, of course, that families are important only or even primarily because of the role they play in the development of children’s human capital—family life is far richer than any such characterization would imply. For many people, beyond whatever their upbringing may have done to prepare them for a productive life, relationships with family members are a source of deep and continuing joy. It is, however, the contribution families make to their children’s productive capacities that makes them of interest in the context of seeking to understand and account for society’s nonmarket production. It would be logical to treat the physical production of children—the 4 million infants who are born in the United States each year—as a component of the human capital produced in the home. If some are inclined to question whether these births represent real investment, they might consider the economic situation in year t+20 in the event there were no births in year t! Yet, children are people in their own right, not a commodity to be bought and sold, and many people are uncomfortable with viewing newborn infants as an output to be measured like other outputs. A similar discomfort is manifest, for example, in laws governing adoptions. Many childless couples would be willing to pay a large sum for the opportunity to adopt a child, but laws prohibit their making payments to birth parents in connection with an adoption (other than to cover birth-related expenses). In the same way that births could be viewed as an investment, it similarly would be logical to treat net immigration to the United States from abroad as an increment to the nation’s stock of human capital. Particularly with birth rates at modest levels in comparison with earlier historical periods, immigration accounts for a substantial share of observed growth in population, and changes in immigration rates can have a significant impact on the growth of a country’s potential labor supply. Indeed, some countries, such as Canada, have consciously structured their immigration policies to favor high-skilled immigrants who can be expected to make especially significant economic contributions. This is not, however, a theme that we pursue further in this report. Even taking as given the biological production of children and the raw material with which these children are born, parental time inputs are critical to the development of children’s intellectual and emotional readiness to learn. Parental time inputs that create a foundation for learning can be viewed as skill-enhancing investments in human capital. Likewise, family care augments the inputs of the medical care system in the production of health, another investment output that yields a flow of future benefits. It is common practice to treat the skills possessed by those entering the workforce as the product of the educational system, the health of the population to be the product of the medical care system, and so on.

OCR for page 79
Beyond the Market: Designing Nonmarket Accounts for the United States In truth, these outputs are products jointly of the formal institutions in question and the household. Without the foundation and continuing support that is provided by the home environment, observed outcomes would be very different. Even in nonmarket satellite accounts that focus on the activities of the household, family care is treated not as an investment but rather as an activity that yields output for current consumption. Inputs of care time can be valued at replacement cost, adjusted where necessary and feasible for differences between productivity in the home and the market. Measures of output valuation can be derived from the cost of analogous market services, as is done, for instance, for child care in the home production satellite accounts developed by the Office for National Statistics of the United Kingdom. But this approach is quite distinct from that suggested for satellite accounts to value the outputs of the education and health sectors, in which the production of human capital is treated as an investment yielding a future flow of income or utility. In this chapter we consider the human capital production function, particularly the contributions that households make to skill capital, health capital, and other forms of human capital. There is a great deal that we do not know about the form of this production function, particularly as it relates to the development of children into well-functioning adults, and we make several recommendations for further research in this area. Given the current state of knowledge, we cannot at present recommend the development of a human capital investment account: not enough is known to proceed in a defensible manner. But it should be recognized that any effort to develop more focused human capital investment accounts—for example, the education accounts or the health accounts that we recommend in later chapters—may yield misleading conclusions if those using the accounts do not recognize the contributions made by prior and concurrent investment activities in the home. While any comprehensive accounting for investments in human capital seems beyond the range of what is feasible, at least for now, it is possible to say something about the magnitude and the value of home inputs to this investment. The chapter also presents evidence related to parents’ investment of time in their children and some thoughts about the valuation of that time. We conclude with a brief description of additional data that would be useful for making progress in this area. DEFINING HUMAN CAPITAL One question to be addressed at the outset is what we mean by “human capital.” Loosely, human capital investment consists of anything that enhances a person’s productive capacities, but this begs the question of which capacities should be considered productive and which should not. Most economic analyses of human capital have focused on market activities. In these analyses, in practice if not in principle, human capital consists of those capacities that raise a person’s

OCR for page 79
Beyond the Market: Designing Nonmarket Accounts for the United States wage rate, which is presumed to reflect their market productivity. One obvious complication is that not all people work for pay. This problem generally is addressed by imputing a potential market wage to those who are not employed. To the extent that a person’s capacity for home production is not taken into account, such a perspective may be misleading. A child who learns from a parent or other family member how to cook, repair a faucet, or create an attractive garden clearly has acquired valuable productive capacity that ought to be considered a part of their human capital. Whether within or outside the market, productivity may depend not only on knowing how to do certain things—“hard skills”—but also on other personal attributes necessary to accomplishment—what might be termed “soft skills” (see Carneiro and Heckman, 2003, for an interpretation of evidence on the contribution of family to development of these kinds of skills). Examples of diffuse, hard-to-measure soft skills include the ability to show up on time, to cooperate with others, to show initiative, to learn how to learn, and to cope with stress (Goleman, 1995, offers a popular discussion; see also Duncan and Dunifon, 1998a). Many of these soft skills are related to underlying personality traits that are partly inherited and partly shaped by early childhood experiences. Developmental psychologists have shown that a 3-year-old’s ability to exercise sufficient self-control to avoid eating a piece of candy placed on her tongue until given permission to do so is a good predictor of future success in school (see McCabe et al., 2004). Such capabilities are probably correlated to some extent with cognitive skills, but they are not easily inculcated or measured. The word “skill” may be a misnomer, insofar as much behavior that is economically productive is the result of preferences and habits rather than skills. Motivation and effort are important determinants of success (see Duncan et al., 1996). Individuals who like to work hard will tend to earn more than those who do not, if only because they will work longer hours. Individuals who enjoy contact with others and have so-called “pro-social” or “incentive-enhancing” preferences tend to earn more than those who do not (see Duncan and Dunifon, 1998b; Bowles et al., 2001; Bowles and Gintis, 2000). Not all economically advantageous preferences can be described in such positive terms. Also relevant are willingness to follow orders or to adhere to boring routines, or what behavioral economist William Simon describes as docility (Simon, 1990). Preferences and habits may have other sorts of effects. A child who learns early that regular exercise is an important part of daily life, for example, might be considered to have acquired a valuable attribute, insofar as that knowledge and the behavior it produces make a significant contribution to the child’s expected health and longevity. More broadly, society relies on citizens who take personal responsibility for their actions, respect the rights of others, and so on. Parents who take steps to impart these values to their children also might reasonably be considered to have added to their child’s human capital. Beyond these examples, there are skills that may contribute significantly to a

OCR for page 79
Beyond the Market: Designing Nonmarket Accounts for the United States person’s enjoyment of life but might not be considered “productive” in the usual sense. For example, how should one think about the acquisition of the skills required to play bridge, windsurf, or appreciate a fine operatic performance? There are arguments on both sides as to whether the development of these skills should be considered an investment in human capital. In this report, we generally have drawn a distinction between the production of outputs for consumption and the enjoyment associated with consumption itself, with the latter not generally considered an output we seek to measure. This perspective might argue against treating the acquisition of recreational skills as an investment. The fact that there is a market for instruction in all of these areas—that people are willing to pay for the acquisition of these skills—might suggest the opposite conclusion. THE HUMAN CAPITAL PRODUCTION FUNCTION From birth to age 5, children develop a foundation on which their subsequent development builds. There is ample evidence that the home environment has strong effects on children’s preparation for learning later in life. Families’ pre-school-age investments in what could be called childhood capital set the stage for the efficiency with which schools are able to educate students, for the amount of on-the-job learning people do after joining the work force, and for other kinds of learning that takes place later in life. Like the basic research and development that is a necessary precondition for having technology to embed in new machines, familial investments in young children—or some substitute for those investments when the family cannot or does not make them—are a precondition for having children capable of learning and eager to learn as they approach school (see National Research Council, 2000). The same argument applies to the complementary production of habits, tastes, and attitudes that affect people’s willingness and capacity to produce good health, which in turn influences the effectiveness of the health care sector. Many parents combine some paid care with caring for their children themselves. The effects of having young children in day care is a subject of ongoing debate. Based on research to date, there is an emerging consensus that long hours of nonmaternal child care for children under the age of 3 can have adverse effects on children’s development, but much depends on the quality of that care relative to maternal care, which itself is highly variable across individuals (see National Institute of Child Health and Human Development Early Child Care Research Network, 1997; Love et al., 2003). As any parent or schoolteacher can attest, there are important complementarities between the efforts of schools and the efforts of families in encouraging and assisting school-age children to learn. Quite distinct from what parents do before schooling begins, their ongoing encouragement also affects the ability and willingness of children to learn in school and the capacity of schools to teach children. Unlike the raw materials purchased by the manufacturing firm for its production

OCR for page 79
Beyond the Market: Designing Nonmarket Accounts for the United States of final goods, in the educational arena there is an ongoing relationship between the efforts of the firm—the school—and the supplier of that raw material—the family. One indication of the widespread acceptance that environment matters is the support that programs like Head Start have enjoyed. The recognition that certain 6-year-olds may need remedial attention to become “school ready” implies that there are many 6-year-olds who do not need such remedial attention because they received the support needed to become school ready from their families. Spending on the Head Start Program is considered a part of national economic output. At least in concept, so too should be the familial efforts that prepare so many young children for their school experience without the need for compensatory preparation. The discussion thus far might suggest that parents’ influences on their children are the result of deliberate decisions. In fact, many of the effects that parents—and later, schools—have on children’s preferences and habits are better described as spillovers than as explicit decisions (England and Folbre, 2000). Children growing up with highly educated, economically successful parents enjoy environmental advantages that are independent of the level of parental effort and attention they receive. Parents who want to set a good example for their children may be limited by their own shortcomings. Similarly, public schools can control their curriculum, but not the characteristics of their student body that affect motivation and readiness to learn. Many studies of “ethnic capital” and “social capital” describe environmental impacts over which neither parents nor children have much control (see Brooks-Gunn et al., 1993; Wilson, 1995). Small initial differences in capabilities can have feedback effects that lead to much larger differences, augmenting inequalities (Lundberg and Starz, 2000). These effects have implications for the measurement of relationships among the home environment, education, earnings, and well-being. Genetic differences, too, may be important in explaining observed variation in students’ interest and ability to learn. There is a vast literature on this topic, but still an enormous amount that is not known. The educational psychology literature, for example, has not produced scale measures that quantify a child’s learning receptivity. There are agreed-upon measures of “IQ” and of achievements in certain subjects, though even they are subject to criticism. Less developed are measures of a child’s social or emotional capacity for learning—the eagerness, the curiosity, the habit or capability to concentrate, and so on, that are critical to success in the educational system. Creation of better measures of both the cognitive and the noncognitive dimensions of children’s development is an important area for further research. Human beings are complex entities, and human development is a multidimensional process. Wise (2004) suggests using “sentinel capabilities”—measurable genetic, environmental, and health variables that affect an individual’s capacity to transform potential into functioning—as key indicators of a broader pattern of development. Whether or not this particular approach proves to be a

OCR for page 79
Beyond the Market: Designing Nonmarket Accounts for the United States productive strategy, research in this area would be an important step toward modeling and analyzing families’ contributions to preparing their children for schooling and later life. Using presently available measures, the child development literature has explored the relationship between the family environment and child outcomes. But this literature does not tell how—through what mechanism—parents’ interactions with their children affect those children’s learning receptivity. Research on “parenting” offers interesting distinctions among styles and suggests that certain styles are more or less successful in producing certain attributes of the children. Still, that literature provides no basis for specifying a production function relating child outcomes to parental inputs, if indeed the process is so deterministic. Further research devoted to identifying the causal effects of different parenting behaviors on children’s cognitive and noncognitive development is warranted. One strategy that has yielded useful information on related questions has been to exploit variations in outcomes between twins or siblings who are separated at birth and raised in different home environments. Differences in outcomes between siblings who share all or a portion of their genetic make-up but have experienced different upbringings may support stronger inferences about the effects of environmental differences than differences in outcomes between individuals whose genetic make-up differs in unknown ways. One practical problem is the difficulty of identifying and matching up pairs of twins separated at birth. Much of the available knowledge comes from studies that include twin pairs separated up to 2 years after birth, which, by way of the common early environment, introduces some “contamination” to the results.1 A further area for investigation is the relationship between the cognitive, emotional, and social development of children at the time they enter school and later educational and labor-market performance. A literature exists that attempts to relate developmental attributes of children at the time they enter school to later educational attainment, but there remains much that is not known. Conclusion 4.1: For all of the reasons discussed above, we do not believe that the state of knowledge at the present time is such that efforts to develop a comprehensive human capital satellite account would be fruitful. There is too much that is not known about the relationship between inputs of household time and other resources, about the interaction between household inputs 1   In part because of the limitations of twins studies, researchers also have used the natural experiment created by adoption to examine the relationships between family environment and outcomes. The growing literature in molecular genetics (e.g., Plomin, et al., 2002) provides additional evidence about the determinants of children’s cognitive and noncognitive development.  

OCR for page 79
Beyond the Market: Designing Nonmarket Accounts for the United States and the inputs of the educational system, the health care system, and so on, and about the measurement of the relevant human capital outputs. Thus, the basic research foundation needs to be strengthened before a human capital account can be contemplated. Specifically, researchers need to continue working (1) to develop better measures of children’s cognitive and, especially, noncognitive development; (2) to identify the causal effects of different aspects of parenting behavior on child development; and (3) to better understand the relationship between children’s cognitive and noncognitive development at the time they enter school and these children’s later educational and labor-market outcomes. While the deficiencies in knowledge are real, in answering questions about schools’ effectiveness in promoting skill capital, it matters very much what parents and other family caregivers are contributing to the development of pre-school-aged children. In seeking “responsibility” within public schools—and looking at the financial expenditures, the curricula, and the quality of teaching staffs, and then using test scores or other output measures to assess the effectiveness of those “inputs”—it is important to remember that important non-school “inputs” from the family or the home affect the measured outputs. This nonmarket investment in children’s human capital both precedes the influence of the public school during their first 5 years of life, and coexists with the school’s influence during every year thereafter. Put somewhat differently, the product of the formal educational sector is fundamentally a collaborative product requiring the cooperation of students, which may be reinforced or undermined by families. When the effectiveness of a school is assessed by looking at student achievement, treated as a function of school inputs without adequately accounting for the considerable efforts of the students or the compensatory and complementary efforts of the family, the results are apt to be misleading (Michael, 1999). Just as the schools often take too much credit for the remarkable achievements of their students, they are also too often accused of performing poorly when the difficulties lie in the home, not with the school. Similar points can be made with regard to the production of health capital. At young ages, children’s diet and exercise, exposures to harmful and beneficial environments, and so on, are determined by their parents; recent literature also has emphasized the importance of the uterine environment. At older ages, children continue to be influenced by their parents’ health attitudes and habits. The maintenance of health capital is an interactive process, involving individuals’ own behavior to promote their health (or to put it at risk) and their use of market-produced medical care to prevent or remediate health reductions. But too often, health is discussed and treated as though it were purely the product of the health sector, rather than a joint product of market-provided goods and services and home activities.

OCR for page 79
Beyond the Market: Designing Nonmarket Accounts for the United States Perhaps one of the most undervalued, and surely one of the least well-measured, aspects of nonmarket-produced human capital involves the personal habits of citizenship—acceptance of personal responsibility, respect for the rights and interests of others, a willingness to take on social responsibilities, and so on. Values of ethical conduct, principled guidance from religious teachings or social commitment, empathy and civility, all are somehow and to varying degrees encouraged in youths. Through exposure to examples, admonitions and guidance from parents and caregivers, and observation and experience among their families, their ethnic, religious, and neighborhood communities, and their peers, children learn, they make judgments, and they form habits that influence behavior later on. These attributes have considerable influence on subsequent skill- and health-capital formation and on individuals’ personal well-being and social contributions. In principle, the efforts and expenditures of resources by parents and families, and by nonmarket institutions like social clubs and religious or community organizations could be accounted for in some objective fashion. There is little dispute that these inputs and their outputs have value, but this is something we do not know how to measure. Among the benefits of good child-rearing is that, if and when a child becomes a parent, he or she is likely to perform more effectively in that role. There is evidence, for example, that children’s cognitive development, subsequent educational attainment, and health status all are positively related to their parents’ education (Haveman and Wolfe, 1984, 1995; Wolfe and Haveman, 2001, 2002). Similarly, the home environment has an important effect on children’s emotional and social development (National Research Council, 2000). In valuing the investment in a child’s human capital, surely one would want to account for the enhancement of that child’s capacity for raising productive, healthy children in his or her own turn, though again it is most unclear how one might do this. A further point to be made (see Grossman, 1999) is that different types of human capital influence the production of other types within the nonmarket sector, particularly for education and health. It is probably the case that healthier people learn more easily and surely the case that more skilled or more educated people are more efficient in producing good health. The interrelationship between skill and health emphasizes yet again that the market sector production of these two products is not separable from the nonmarket production. One could argue that simplifying assumptions are a necessary feature of any national income accounting framework. But one implication of this discussion is that many empirical estimates of the contribution of formal education to national income may be misspecified in two countervailing directions. In one direction, failure to account fully for the costs or benefits of parental inputs of time and money into the production of children’s capabilities may overstate the rate of return to formal education. In the other direction, the persistent tendency to value education entirely in terms of its contribution to market earnings may understate its total contribution, since the increased education of parents is surely a major

OCR for page 79
Beyond the Market: Designing Nonmarket Accounts for the United States factor both in their own direct nonmarket production and in the subsequent development of their children’s capabilities (Leibowitz, 1974). FAMILY INPUTS TO THE DEVELOPMENT OF CHILDREN’S HUMAN CAPITAL We concluded above that our knowledge base is not yet sufficient to warrant development of a comprehensive human capital satellite account. Having so concluded, however, we believe there is merit in considering what can be said about the magnitude of household investments in children, if for no other reason than to understand what implicitly is neglected when researchers focus on the production of human capital in the educational sector, the production of health capital in the health care sector, and so on. Production of a child possessing any level of human capital typically requires both out-of-pocket expenditures on behalf of the child and inputs of parental time. As is true with other sorts of home production, out-of-pocket expenditures on children already are reflected in the national income and product accounts. What is not reflected is the time that parents and other family members devote to children. Time-use studies from a number of countries indicate that married or cohabiting mothers of children under age 5 average between 2 and 3 hours a day in primary child care, depending on their hours of paid employment, while fathers average about 1 hour a day (Gauthier et al., 2001). In these studies, primary care is defined as time during which caring for a child is the primary activity in which a parent or other adult is engaged. If the reported figures seem small, it is because primary care represents only a small share of total time devoted to (or constrained by) children. Family members often care for children while simultaneously engaging in other activities, such as preparing meals or watching television. Survey results from Australia, a country that collects detailed information on secondary activities as part of its ongoing time-use data program, show that counting secondary activities more than doubles the hours devoted to child care (Ironmonger, 2003). The need to be available or “on call” also imposes significant constraints on parental schedules. Like firemen who spend relatively little of their time fighting fires, parents stand ready to provide attention, even when children are sleeping. In recent cognitive pretesting for the American Time Use Survey (ATUS), the Bureau of Labor Statistics found that participants “strongly suggested that the concept of secondary child care is not intuitively meaningful, because most parents would consider those activities, ‘just part of being a parent’” (Schwartz, 2002, p. 35). That task, “being a parent,” extends well beyond tallies of activity hours. Statistics Canada opted for a national time-use survey that omits consideration of secondary activities but includes stylized questions regarding care time. The Bureau of Labor Statistics adopted a similar strategy for the ATUS, asking

OCR for page 79
Beyond the Market: Designing Nonmarket Accounts for the United States each respondent to estimate how many hours during the survey day children were “in their care” (Schwartz, 2002). Yet even this approach restricts respondents to time periods in which the adult and at least one child were awake. Similarly, the Child Development Supplement of the Panel Study of Income Dynamics (PSID), another source of data on time use, in this case from the child’s perspective, collects information on time children spend with adults, but not directly engaged in activity with them. In this survey, too, children’s sleep and personal care time are excluded from consideration. Given the limitations of available data, most empirical studies of time devoted to parenting focus solely on primary activities. Still, the results are interesting and reveal some common patterns internationally. The University of Essex Multinational Time Use Survey data set includes data on time-budget surveys from more than 20 industrialized countries. These data show that increases in maternal employment cause only a small reduction in maternal time in activities with children. Moving from out of the labor force into a 35-hours-per-week job is associated with a decrement of only 7 hours a week in mothers’ primary child care time (Gauthier et al., 2001). Studies of the more limited time-use data sets available for the United States confirm that maternal employment has surprisingly small effects. They also show that fathers’ time with children remains low regardless of such factors as their own or their spouses’ hours of work (Bianchi, 2000; Hofferth, 2000; Hofferth and Sandberg, 2001; Sandberg and Hofferth, 2001). What has happened in the United States to parental time in activities with children as mothers have increased their employment? The data available to answer such a question are far from perfect (Budig and Folbre, 2004), but a small survey administered in 1998-1999 provides useful data for comparison with the results of an earlier 1965 survey. One recent study making use of these data (Bianchi, 2000) reports that mothers were engaged in primary activities with their children aged 18 and under for about 12 hours per week in 1998, compared with about 10 hours per week in 1965, a statistically significant increase. Time devoted to secondary activities with children also increased significantly, from just over 15 hours to just over 19 hours per week. The broadest measure, maternal time spent with children, increased only slightly, from about 37 hours per week to just over 38 hours (Bianchi, 2000, p. 405; Sayer et al., 2003). Somewhat surprisingly, none of the measures show a significant decline in the time mothers devote to their children. One explanation for these findings might be that time with children is a sufficiently high priority that, as mothers’ hours of paid work have grown, other activities have been sacrificed. Another, perhaps complementary, explanation might be that changes in household technology have made it easier than in the past to combine child care time with other activities. The availability of prepared foods that need only to be heated before consumption, for example, might make it easier to combine child care and meal preparation than would have been the case when meals had to be prepared from scratch; more widespread use of home computers might make it easier to com-

OCR for page 79
Beyond the Market: Designing Nonmarket Accounts for the United States bine child care with paying bills or carrying out job-related tasks; cable television and DVD players might make it easier to combine child care with entertainment activities; and so on. It is clear, however, that the choice of the appropriate definition of “child care” affects interpretation of trends. A related issue is the quality or density of care time. In addition to parents sometimes combining other activities with supervision of their children, an hour of a parent’s time can be devoted to the care of one or several children, and it may or may not overlap with the presence of other adult caregivers. Interestingly, because fertility rates have fallen over the course of the century, even if maternal time devoted to child care remains roughly constant, the amount of parental time spent per child may actually have increased (Bianchi, 2000). These issues bear directly on how the monetary value of parental time inputs should be estimated. Analysis of PSID time-use data for 1997, based on time diaries for about 2,800 children under the age of 13, shows that children spent about 47 percent of their time sleeping or engaging in personal care. Of the time they spent participating in activity with another person—that is, of the 53 percent of their time they were not sleeping or engaged in personal care—about 45 percent was spent with a parent or a relative who probably was not paid. Much of this time involved overlaps: of the time they spent in activities with at least one parent, at least 32 percent involved the participation of another parent or family member and 22 percent involved the participation of a sibling. The PSID child development supplement diaries permit a clear matching of which child was getting parental input, but still have the problem of a parent spending time with other siblings at the same time. The average child aged 12 or under spent about 29 hours per week in activities with a parent (Folbre et al., 2004). PSID data also show that parents invest in their children in a reactive fashion. Those who receive negative feedback on their child’s performance or functioning have been shown to invest more in those children—one example being extensive investments of parental time to obtain some developmental threshold for a child with Down’s syndrome. In the PSID child development supplement sample, even parents of children with less severe disabilities have been shown to devote more parental time to children reported to have physical limitations (Stafford, 1996). Who invests and why matters for the interpretation of data used to study parental time investment and early human capital development. As already noted, the new ATUS will provide a wealth of information on how Americans spend their time, including information on the time that parents spend with their children. As noted in Chapter 2, however, the ATUS covers only the population aged 15 and older; it does not provide ongoing data on children’s time use, of the sort collected as part of the PSID. Such data are likely to be important for a full understanding of how families affect children’s development, and thought should be given to how they might best be collected.

OCR for page 79
Beyond the Market: Designing Nonmarket Accounts for the United States VALUING THE TIME PARENTS DEVOTE TO CHILDREN The “cost” of a child is the subject of a moderately large literature. The U.S. Department of Agriculture regularly provides estimates of spending on children based on analysis of the Consumer Expenditure Survey. These estimates are used to help define reimbursement rates for foster care, as well as the child support responsibilities of noncustodial parents. In 2000, parents in middle-income, two-parent families with two children spent an average of $8,915 a year per child under age 12 (Lino, 2001). Significantly, these figures omit the value of parental time inputs, but available time-use data help to quantify that time. The PSID, for example, provides estimates of the average amount of time parents in two parent families with two children devoted to children under age 12 in 1997. Much depends on how child care time is defined. Using a relatively narrow measure that includes only activities in which both child and adult are involved, children received 31 hours of nonoverlapping parental care per week. By the broadest measure—counting all parental and other family time in activities with children plus supervisory time (defined as all time that children were not in the care of another person and not sleeping)—children received 79 hours of family care per week. In Chapter 3 we argue that for most forms of household production the appropriate wage rate to use in valuing time inputs is the quality-adjusted replacement wage. If parents’ productivity in child care equaled that of paid child care workers, it would make sense to use an estimate of child care workers’ wages to value parental time input. In 2000, the average wage for a child care worker was $7.43 per hour. Multiplying 31 hours of weekly child care (the narrower measure derived from the PSID) by this low wage rate provides a lower-bound estimate of the value of parents’ nonmarket time of $11,977 per year, per child (Folbre, 2004). This number suggests that the cost of a child in two-parent, two-child, middle-income families, including both cash expenditures and the value of time, amounts to $20,892, more than twice as much as cash expenditures alone. While foster parents and divorced custodial parents may devote less time, on average, to young children than married-couple parents, their time expenditures surely also are substantial. The dollar cost estimate just cited hinges on parental time being no more (and no less) productive than that of hired child care providers. Of all the activities of household production, however, time devoted to the care of family members is probably the least completely substitutable with replacement market inputs. Many families benefit from the assistance of paid caregivers for part of the day, but a child who is raised entirely by paid caregivers may be less likely than one who also enjoys consistent parental attention to develop into a productive, healthy adult. If so, the appropriate replacement cost to use in valuing parental time inputs into the care of their children, at least at the margin, could be significantly higher than the wage paid to child care workers. Given the present state of

OCR for page 79
Beyond the Market: Designing Nonmarket Accounts for the United States knowledge about the production function for human capital, broadly defined, however, it would be difficult to say what the right wage rate might be. Another candidate wage rate for use in valuing parental time inputs would be the opportunity cost of that time, perhaps proxied by the parent’s market wage rate. This is an alternative that we have rejected in the case of other sorts of home production, reflecting in part our belief that there is likely to be only a weak relationship between productivity in most home production tasks and market wage rates. In the case of parents’ interactions with their children, however, there is evidence that parents with more human capital are more productive not only in the market, but also in fostering the development of their children’s human capital. What we do not know much about, however, is the strength of the relationship between the productivity of parents’ market time and the productivity of the time they devote to child-rearing activities. As an aside, it may also be worth noting that the opportunity costs of parental time devoted to child care—or more broadly, taking responsibility for their children—may go beyond any potential earnings forgone as a consequence of spending a marginal hour in child care rather than market work. The commitment to bear and raise a child often has life-style implications—e.g., deciding to stop smoking or cease alcohol consumption during pre-pregnancy and pregnancy intervals, changing or postponing career commitments, or choice of location and attributes of the family home. It is difficult to know, however, how one would incorporate this insight into a family care satellite account. One concern with using the opportunity cost of parents’ time to value their contributions to their children’s human capital is the difficulty, if not the impossibility, of distinguishing between the consumption and the investment components of this care; and we do not suggest that time-use surveys or satellite accounts should attempt to do so. Nonetheless, the relationship between family care and human capital has important implications for the interpretation of nonmarket satellite accounts for other sectors, most notably education and health.