Abstract
Can parental childcare be described as productive work? If so, is this work reducible to the specific physical activities designated in most time use surveys, or does it include more diffuse responsibilities for supervision, socialization, and management? These questions invite attention to debates over the meaning of work itself, which have been shaped not only by gender and academic discipline, but also by empirical results of diary-based time use surveys. Recent quantitative research strongly suggests that neither the temporal demands, nor the economic contributions of parental childcare are fully captured by conventional measurement of specific childcare activities. The numbers themselves urge us to look beyond the clock to carefully consider how time use categories are conceptualized.
Social science evolves via a complex dialectic between theoretical frames and empirical research that is often shaped by the identities and interests of its practitioners. Debates over the interpretation of childcare as “work” provide a rich illustration of this process. Women and men have often taken different sides on this issue, as have sociologists and economists. Yet much recent research on time use relies on standardized categories built into survey design without interrogating their meaning. Closer attention to the history of categorization offers some important lessons for the collection and interpretation of quantitative data.
The conceptualization and measurement of parental childcare bears on gender inequality, since mothers have traditionally taken far more responsibility for childcare than fathers. Paradigmatic assumptions vary by academic discipline. Neoclassical economic theory focuses on individual choices and subjective well-being, with little consideration of distributional conflict or the social value of family care. Sociology, where women are more generously represented, has been more attentive to the nexus between unpaid work and gender inequality. Still, the sociologists who pioneered modern time use research in the U.S. often defined work narrowly in terms of specific activities, a definition that is particularly problematic for analysis of parental childcare.
This essay makes a case for a broad definition of childcare that includes supervision, socialization, and management, as well as active face-to-face care. It begins with an historical account of early feminist claims that family care should be considered productive work, claims explicitly rejected by neoclassical economic theory, but implicitly supported by some economists and sociologists. Next, it examines the history of time use survey design in the U.S., noting an early tendency to consider housework, but not parental childcare, as work and, subsequently, to define only explicit childcare activities as work. Finally, it exploits an important innovation of the American Time Use Survey inaugurated in 2003 to provide evidence that purely activity-based measures seriously understate the temporal demands and economic contributions of parental childcare.
Defining and valuing work
One might assume that social scientists would agree on a definition of work, especially if they aim to measure it on the clock. Common usage of the word “housework” seems to imply that work is involved. Yet definitions of production have long been contested. Anna Howard Shaw put it this way in 1899: “A gentleman opposed to their enfranchisement once said to me, women have never produced anything of any value to the world. I told them the chief product of the women had been the men and left it to him to decide whether the product was of any value.” 1
In the late nineteenth-century, feminists wrote a letter to Congress complaining that the U.S. Census ignored the productive contributions of housewives and mothers, to little effect (Folbre, 1991). A founding father of neoclassical economic theory, Alfred Marshall of Cambridge University, confidently explained that women outside of paid employment should be categorized as “dependents.” He emphasized their moral obligation to care for family members and expressed concern that opportunities for work outside the home would corrupt their altruistic commitments (Folbre, 1991).
Many of the women who gained greater access to higher education and professional standing in the early 20th century were keenly interested in assessing unpaid work, including childcare, in economic terms (Folbre, 1991). In the 1920s, Hildegard Kneeland of the U.S. Department of Agriculture’s Bureau of Human Nutrition and Home Economics organized time use surveys of rural homemakers that later informed several historical analyses of time use in the U.S. (Kneeland, 1929; Vanek, 1980; Ramey, 2009). Kneeland acknowledged the difficulties of accurately measuring parental childcare even as she detailed its most obvious demands.
Kneeland reiterated earlier criticisms of the U.S. Census for treating only paid work as “gainful,” but also critically assessed feminist arguments in favor of wages for wives based on what it would cost to replace them with paid services. While endorsing the general principle of market valuation, Kneeland observed that time use surveys would be of limited assistance because of the complexity of household management and variation in work hours. She suggested, instead, that a housewife, as a partner in a productive enterprise, should be endowed with a legal right to one-half of her husband’s earnings (presumably in perpetuity) (Kneeland, 1929:40).
Kneeland’s discussion demonstrates that the concept of replacement cost—imputing the value of unpaid work by asking what it would cost to purchase a replacement for it—emerged before Margaret Reid (1934) provided a more formal definition of work as any activity that someone else, could, in principle, be paid to perform. The issues Kneeland raised—whether hours of unpaid work should provide a basis for their remuneration, and whether housewives should be treated like hourly employees or as partners in a larger social enterprise—remain relevant today.
Margaret Reid treated these issues more delicately in her classic and ultimately influential book, The Economics of Household Production (1934), citing Kneeland’s arguments without actually endorsing them. She aimed to persuade economists to recognize an important distinction between consumption (which provides purely personal benefits) and production (which generates benefits for others). As one of the few women of her generation to obtain a formal academic position in an economics department (the University of Chicago) she is sometimes characterized as a forerunner of Gary Becker and the neoclassical “New Home Economics.” Her theoretical perspective, however, was far more pragmatic, more akin to the institutionalist approaches of non-neoclassical economists like Richard Ely and John R. Commons.
Reid’s definition of work, sometimes referred to as the “Third Person Principle” in reference to the hypothetical substitute for the unpaid worker, requires no assumptions regarding preferences, motives, or subjective well-being. Whatever their motivation, Reid asserted, wives and mothers were performing services very similar to those that could be purchased from domestic servants, babysitters, and purveyors of other substitutes for home-produced services. If the supply of unpaid services were withdrawn, the demand for them would need to be met in some other way.
Reid’s definition was initially ignored by sociologists as well as economists. When Pitirim Sorokin and Clarence Berger published Time-Budgets of Human Behavior in 1939, they not only listed household activities separately from work, but also distinguished between those activities which were “personal,” such as caring for children, from those which were “physical,” such as arranging flowers or baking. The specific activity, “watching children” was placed under the rubric of “idling,” which also included “contemplating, daydreaming, and marveling at clouds” (Sorokin and Berger 1939:29).
Like Marshall, these sociologists made a sharp distinction between “activities done for the sake of others” and those with predominantly “utilitarian or hedonistic objectives.” The first included a variety of activities both intangible and tangible: “to be agreeable, to help someone, to inform someone, to greet co-workers, to care for someone, to please someone, to accommodate someone, to make child clean, to take child out, not to cause worry.” The second list, encompassing “economic” motives, included an odd variety of activities presumed remunerative: “research, earn a living, cheapest means, for business, to get pay, to save car fare, to plan budget, to wait for paycheck” (1939:91) To their credit, the authors emphasized that motivations were almost always mixed. Still, the notion that childcare was an activity done for the sake of others (i.e., for altruistic reasons) apparently disqualified it as work.
Researchers in the predominantly female discipline of home economics never considered moral obligations or altruistic commitments inconsistent with productive contributions. To the contrary, they interpreted them as necessary for provision of the highest-quality services, as Catherine Beecher explained in her 1841 Treatise on Domestic Economy. Practical concerns, such as how to run an efficient household, helped motivate research using the time-budget tools that Kneeland and others had earlier developed. In the 1960s, Kathryn Walker of Cornell University undertook an important survey of families in Syracuse, New York that treated family care as an important category of household production, noting that the decline in household production of goods had been counterbalanced by an increase in the production of household services (Walker, 1969).
Academic economists, by contrast, continued to ignore Reid’s definition of work. The version of neoclassical theory that developed in close physical proximity to her at the University of Chicago was conceptually remote. It built on Marshallian foundations to construct a more subtle defense of the notion that unpaid work is unproductive, emphasizing its subjective pleasures. Particularly in its textbook version, neoclassical economic theory defines work as an activity providing no intrinsic reward, performed only in order to derive satisfaction from the goods or services that can be obtained as a result.
If people are willing to perform a task simply because they enjoy it, there is no reason why the market should reward it—or if the market does reward it, it will do so at considerable discount. From a market-centric economic perspective, the willingness to pay reflected in demand curves determines whether another party benefits from an activity. Non-market transfers to child or adult needing assistance are therefore interpreted as the result of altruistic preferences. Altruistic preferences, in turn, imply that a person derives satisfaction from self-sacrifice, so they are compensated by psychic income. In other words, virtue is its own reward.
Activities undertaken for their own sake—those considered intrinsically rewarding—are taken to resemble leisure whether they provide benefits to others or not. Work for pay, presumably motivated by pay alone, is considered productive because someone is willing to pay for it. Market exchange, by definition, leads to productive outcomes, and the prices generated by the forces of supply and demand become the arbiter of value (with some exceptions labeled externalities or spillovers). The distinction between market work and non-market work places the latter in a less economically significant category.
Gary Becker and other practitioners of the “New Home Economics” called attention to the importance of household production as an input into consumption as well as a direct source of utility, but seldom acknowledged its implications for the measurement of family living standards or poverty lines. Benefits to others cannot be measured, because they too take the form of subjective well-being, merged as a result of altruistic preferences with the subjective satisfaction of the provider. The contribution of non-market work to living standards, as well as the skill and effort put into it, becomes irrelevant, and value can only be assessed by opportunity cost—comparison with individual wages foregone.
Valuation of unpaid work based on opportunity cost is often described as a simple alternative to replacement cost, but it has profoundly different implications. Consider the following examples: If person A earns only 80% of what person B earns, their non-market work is also only 80% as valuable; since men generally earn more than women, their household production time is valued at a higher rate. If mother A earns only half as much as mother B, an hour she spends breastfeeding her infant would be valued half as high as an hour mother B spends doing the same thing. Since leisure, like non-market work, is valued at opportunity cost, Person A, devoting 4 hours a day to active childcare after getting home from paid employment, makes no greater economic contribution than Person B, who earns exactly the same hourly wage, but devotes 4 hours in the evening alone, watching a football game on TV.
Unlike Marshall, Becker never argued that women have a greater moral obligation than men to provide childcare; he simply assumed that, if they voluntarily chose to do so, they must derive more satisfaction from it. As with the more prosaic textbook version of neoclassical reasoning, intrinsic motivation disqualifies non-market work as productive work with implications for the economy as a whole. Notice the consonance with Sorokin’s and Berger’s assertion in 1939 that “watching children” was a form of idling like daydreaming or watching clouds.
Fortunately, the sociologists who pioneered large-scale time use surveys in the 1960s were largely unencumbered by neoclassical economic reasoning. They embraced the third-person definition of work, which could be easily operationalized, standardized, and coded as specific activities. Since this did not necessarily imply changes in the definition of income or output, mainstream economists accepted the nomenclature, and some even agreed that unpaid work should be valued at what it would cost to replace it with comparable market substitutes (Kendrick, 1979; Eisner, 1989). As more time use survey data gradually became available, empirical research began to undermine the assumption that unpaid work should be understood only as a source of subjective satisfaction. Resistance to viewing childcare as a form of productive work, however, proved harder to overcome.
Survey designs
When Alexander Szalai and his colleagues launched a major multinational time use survey of 12 countries in the late 1960s, they originally planned to exclude housewives (as well as other members of the non-employed population) but ultimately decided that such exclusion would bias comparisons between Western Europe and socialist countries where women’s employment was higher (Szalai and Scheuch 1972:19). This was good judgment on their part; their tallies of unpaid work generated what may well have been the first international evidence of employed women’s “double day” (Robinson et al., 1972: 131). The interdisciplinary researchers tallied childcare separately from housework but also distinguished it from leisure. However, they focused on specific childcare activities, distinguishing these from childcare as a “responsibility” or “childcare in the generic sense” (Stone, 1972:249, 252). None of the specific activities included supervision, being “on call,” time with children, or managerial responsibilities.
One of the contributors to the Szalai volume, sociologist/psychologist John Robinson, went on to conduct several time use surveys in the U.S. and to establish the University of Maryland as a leading center of time use research. Robinson took a wide-ranging, intellectually curious, yet also rigorous approach to the analysis of data collected in 1965, 1975, and 1985. While best known for his contribution to debates over historical trends in leisure time, he was attentive to issues of childcare measurement, and explicitly noted the importance of supplementing measures of specific childcare activities by attention to secondary childcare (conducted simultaneously with other reported activities) and time spent in the presence of children. He emphasized that data on “who else was present” while engaged in household activities fails to register significant amounts of parental worrying and planning and other times when the child is “out of sight but not out of mind” (1977:70). On the other hand, he never explicitly acknowledged a basic supervisory constraint that becomes especially binding when mothers seek employment outside the home: young children cannot safely be left alone without a much older child or an adult in proximity and on call.
The U.S component of the larger Szalai-led multinational project, a 1965 survey, was centered at the University of Michigan, where Robinson was a graduate student. This University also sponsored surveys conducted in 1975/76 and 1981/82 (for a more detailed chronology of these surveys, including methodological details, see Ver Ploeg et al., 2000:26–28). Two economists, Russell Hill and Frank Stafford, took a particular interest in the childcare data (Hill and Stafford 1980). Heavily influenced by Beckerian reasoning, and following the example of Arleen Leibowitz (1974), they viewed parental childcare as an altruistic contribution to the educational attainment and potential earnings of the next generation.
The later Michigan surveys followed the Szalai project’s precedent, categorizing childcare as time devoted to physically caring for the baby or child, to teaching the child, to reading/talking to the child, indoor and outdoor playing, and time devoted to medical care (Hill and Stafford 1980). When Robinson and colleagues mounted a survey from the University of Maryland in 1985, they too coded specific face-to-face childcare activities. However, Robinson explicitly included both childcare and housework in a list of “productive” activities, and also cautioned that “while parents spend less than an hour a day with their children (and fathers less than 20 min) as a primary activity, earlier diary studies established that total time with children was closer to 25 h a week, which more accurately reflects the time commitment that having a child brings with it” (Robinson, 1997:xxi) (emphasis mine).
In their 1998 book, Time for Life, Robinson and his co-author Geoffrey Godbey summarized trends in parental childcare between 1965 and 1985, reporting that mothers spent, on average, about the same amount of time in care activities per week (9.9 h in 1965, 8.3 h in 1975, and 8.9 h in 1985) (Robinson and Godbey, 1997:105). They also reported that secondary childcare had not declined, but that they lacked comparable data for “contact time” with children in 1985 (1998:106). Their lack of attention to this measure implied that parental childcare activity (whether primary or secondary) was work, but that “contact time” was not. They did not address what seems, in retrospect, a major puzzle in their results: their 1985 survey showed a difference of less than an hour a day in the childcare time of employed and non-employed mothers, hardly indicative of significant conflict between paid employment and family care. This puzzle would later be addressed in more detail by Robinson’s colleagues at the University of Maryland, and subsequent time-diary surveys conducted there in 1995, 1998–99, and 2000–2001 (see later discussion).
In the meantime, analysis of university-sponsored surveys revealed other findings even more directly relevant to the conceptualization of work. The diary-based approach made it possible to ask respondents to rate their level of subjective well-being or enjoyment (Juster et al., 1981). However, Robinson and Godbey reported that in 1985, respondents to their survey rated paid work 7 out of 10 on “enjoyment,” while cooking at home yielded 6.6 and childcare 6.4 (1997:245). In other words, childcare was less, rather than more, enjoyable than both paid employment and cooking. While results are mixed, a later survey also reported that childcare is not distinctively enjoyable (Fisher, 2010).
Of course, much depends on how subjective rewards are defined and described. Parents are more likely to describe their experience as meaningful rather than as fun (Senior, 2014). Yet many paid employees also describe their jobs as meaningful (Cassar and Meier, 2018). A variety of contextual variables affect subjective experience; the same activity can be fun and meaningful at one point in time, and tedious and boring at another. The more voluntary an activity is, the more we might expect it to have positive valence. Some research indicates that fathers in the U.S. enjoy childcare more, on average, than mothers, perhaps because they do much less of it (Connelly and Kimmel, 2015; see also Musick et al., 2016).
For the same reason, men may be less interested in childcare, on average, than women are. Even at the University of Michigan, a major epicenter of time use research, attention to parental childcare was, until recently, rather limited. From its outset in 1968, the University-sponsored Panel Survey of Income Dynamics (PSID) asked stylized questions about the hours respondents spent in housework in the previous week. However, it did not inquire after hours spent in childcare until 2017 (Insolera et al., 2021). A Child Development Supplement to the PSID, first instituted in 1997 (CDS-PSID), provided data important for analysis of child outcomes, but not for analysis of maternal or paternal time allocation. Likewise, the National Survey of Families and Households (NSFH) in the U.S. administered in 1987–88, 1992–94, and 2001–2003 asked respondents to recount their hours in nine household tasks, but not in childcare. 2 The National Survey of the Changing Workforce (NSCW) began collecting family childcare reports in 1992, but only for employed parents. 3
The absence of data on time devoted to childcare had significant consequences for empirical research on women’s bargaining power in the home. For instance, an early analysis of the household division of labor between spouses examined only housework, asserting that “the ‘work’ component (e.g., bathing and feeding the child) cannot be satisfactorily separated from the ‘leisure’ component (e.g., playing with the child)” (Blair and Lichter, 1991:96). Once this precedent was set, many subsequent empirical studies followed suit. For instance, the title of a later article equated the gender division of labor in housework (not including childcare) with the gender division of household labor (which presumably includes childcare) (Bianchi et al., 2000).
I, myself, fell prey to this omission. For reasons of data availability and desire for consistency with earlier studies, I contributed to an empirical analysis of the gender division of labor that also focused on housework alone (Bittman et al., 2003). The omission of childcare in analysis of the National Survey of Families and Households goes without comment in Gupta, (2006). A more recent study using the pre-2017 Panel Study of Income Dynamics rationalizes its limitations by noting that childcare time had been ignored in many previous studies and citing economic research claiming that childcare resembles leisure as much as work (Killewald and Gough, 2010:18).
Obviously, the issue of who does more unpaid work than who hinges on how work is defined and valued. The resurgence of global feminism in the 1970s gradually ramped up political pressure for reconceptualization as well as expanded data collection. Momentum built after l985, when the Third U.N. World Conference on Women in Nairobi in 1985 passed a resolution protesting women’s long hours of unpaid work (Luxton, 1997). In 1988, the New Zealand activist and policymaker Marilyn Waring mobilized public attention with her widely read book, If Women Counted (Waring, 1988; Bjornhold and McKay, 2014).
When grassroots women’s groups rallied around the issue in the U.S. they seldom (if ever) made a distinction between housework and childcare. In 1991, U.S. Democratic Representative Barbara-Rose Collins of Michigan (the first black woman from her state to be elected to Congress) introduced the Unremunerated Work Act, calling for the Bureau of Labor Statistics to “conduct time use surveys of unremunerated work performed in the United States and to calculate the monetary value of such work.” 4 The bill never made it out of committee, but it got the attention of the U.S. Bureau of Labor Statistics (BLS) (Horrigan and Herz, 2004).
In 1997, the BLS held a planning conference, co-sponsored by the MacArthur Foundation Research Network on the Family and the Economy, to discuss plans for a nationally representative time use survey, drawing on the experience of national statistical offices in Australia and Canada that had implemented earlier surveys and inviting input from a large group of time use researchers (Ver Ploeg et al., 2000; Horrigan and Herz, 2004). As a participant in this conference, I expressed concern that the core questions of the survey, focused, like previous surveys, on activities, would not capture more diffuse responsibilities for children such as supervision (Ver Ploeg et al., 2000:55). This concern was consistent with previous research showing that “secondary childcare” and “time with” children far exceeded hours of active care, helping explain why differences in active childcare by employed and non-employed mothers were relatively small (Zick and Bryant, 1996).
The BLS took this issue seriously, and, from its outset in 2003 the ATUS added a distinctive feature, a question posed to respondents in households with a child under the age of 13 after they had filled out the activity diary, inquiring whether any child under the age of 13 was “in your care” during a period in at least one child was awake? Tabulations excluded time that household adults reported they themselves were asleep. While the tabulated summaries described responses as “secondary childcare,” the question clearly went beyond an activity-based classification, measuring something more akin to supervision and responsibility.
Largely because in-your-care time was a new concept that was not directly comparable with results from previous U.S. surveys or those in other countries, relatively few researchers took advantage of this variable. The precedent set by emphasis on physical activities deflected attention from more diffuse supervisory and managerial responsibilities as well as social time in which parents and children were both present. Over the last 20 years, however, analysis of ATUS data has revealed the quantitative significance of in-your-care and social time, calling conventional valuations of parental childcare time into question
Childcare as activity and responsibility
In her Economics of Household Production, Margaret Reid emphasized the pervasive character of mothers’ responsibilities: Even though she may not be on active duty, evidence of her labor is about her. She is continually on call. Much so-called leisure has a “string attached.” In a study of the time spent in the care of 24 infants, one year and under, it was found that the mother averaged 41 minutes per day away from the baby and thus entirely freed from the responsibility of his care (Reid, 1934:319).
This observation, based on a very small sample and made almost 90 years ago, has proved prescient. Recent sociological research calls attention to a “parenting package” that extends well beyond active care to include time spent with children, which likely contributes to their socialization and well-being (Fomby and Musick, 2018). Milkie, (2022) argues that “connection time” measured by co-presence becomes more important as children get older and provides much in the way of the everyday relationship building between children and adults.
The ATUS measure of in-your-care time is also part of the parenting package. Its inclusion is consistent with John Robinson’s early observations about the limits of active childcare measures and with analysis of Australian time use surveys that explicitly probed for “passive” childcare (Bittman and Wajcman, 2000; Ironmonger, 2004; Bittman et al., 2004). It is also consistent with analysis of many other time use surveys, including the Child Development Survey of the PSID (Folbre and Yoon, 2007; Folbre et al., 2005; UNWomen, 2021).
One important catalyst for a broad definition of parental childcare has been growing interest among sociologists in the relationship between time allocation, increases in female labor force participation, and gender inequality. In collaboration with John Robinson, Melissa Milkie and others at the University of Maryland, Suzanne Bianchi played a leading role in analysis of trends in maternal time use in the U.S. offering evidence that employment tended to reduce housework and leisure rather than either primary or secondary childcare activities (Bianchi, 2000; Bianchi et al., 2006). However important this finding, it left open the possibility that mothers spent less time supervising and socializing with their children, activities often combined with housework and leisure.
Like Robinson before her, Bianchi was attentive to this issue. She co-authored an early assessment of the “in your care” data in the ATUS emphasizing that it was conceptually broader than earlier measures of secondary childcare activity, and not comparable with the results of earlier U.S. time use surveys (Bianchi et al., 2005). This assessment warned that, while earlier measures might be too narrow, the ATUS measure might be too broad, and “better suited for estimating constraints on parents’ time than it is for use as a measure of what children receive from that time” (2005:24).
Constraints on parental time did not seem binding between 1947 and 1997, when women’s employment steadily increased in the U.S. By the early 2000s, it was becoming apparent that the leveling off was not a short-term phenomenon (England, 2010). Improvements in the gender wage gap also faltered. Comparative analysis of these trends in women’s employment and earnings suggested that the U.S was lagging behind Northwestern Europe partly because of its relatively stingy public support for employed parents (Blau and Kahn, 2017). In recent years, empirical research on the economic costs of motherhood in the U.S. has proliferated (Correll et al., 2007; Budig et al., 2012; Glauber, 2018).
Average duration, overlaps, and ratios of childcare and co-presence of parents and children (mothers and fathers in households with a child under 13, in hours per day, 2004–2019).
Source: Suh and Folbre (2022).
These reports overlap in time, and precedence is given to the more active category. In Table 1, time devoted to active care takes precedence in the “without overlaps” category, even if another type of care was being provided to another child at the same time. Similarly, time when children were reported in-your-care takes precedence over time that children were reported present in the same room, because it directly implies responsibility. Co-presence often overlapped active and in-your-care time; about half an hour a day was non-overlapping, showing that not all time with children was reported as in-your-care. On the other hand, well more than an hour a day on average was reported as non-overlapping in-your-care time, presumably with children in another room or outside the house.
These estimates of the non-overlapping parenting package sum to 6.02 h per day per father and 8.75 h per mother in households with any child under 13 (Row 8, Table 1)—far greater than their respective averages of active care—1.23 h and 2.35 h, respectively (Row 1, Table 1). Differences by age of children (in the 0–4 age category, with none older, versus the 5–12 age category, none younger) suggest that the younger group absorbs more of all three types of childcares on average than the older group. (For more detailed disaggregation by age and multivariate analysis, see Suh and Folbre, in progress).
As Reid observed, the string tying a mother to a child under the age of 1 is especially binding. It seems obvious that young children constrain parental (and, especially, maternal) time, since leaving them home alone is both legally and culturally culpable. The more important question is whether childcare that is not “active” should be considered a productive contribution—that is, work. At risk of repeating earlier arguments, it should not matter whether parents enjoy this time or not—they almost certainly enjoy childcare (as well as paid employment) at least some of time.
Supervisory care clearly fits the category of work that someone else could be paid to provide—this is exactly what babysitters do and represents an important component of typical nanny’s job description. Some paid occupations entail responsibilities for remaining physically present without necessarily engaging in activities (such as security guards) and some managers and professionals are remunerated for being on call and available even if not physically present.
Even in the absence of supervisory constraint, time parents spend in the presence of children may be perceived as a productive activity, since it contributes to their socialization and well-being (Milkie, 2022). The extent to which parents prioritize active care versus time with children in the U.S. varies considerably by race and ethnicity, and failure to acknowledge the significance of co-presence understates the relative contributions of Black and Hispanic parents, especially fathers (Milkie, 2022). Housework activities may be more salient than time spent with children but are more likely to take place when there is a child at home that needs supervision or attention. If someone has to stay home anyway, they may well try to make themselves useful by putting a pot of beans on the stove or popping clothes in the washing machine. Which, then, is the more “primary” use of time—the housework or the care?
The conceptual stakes are high for economists, in particular. Many countries are now developing satellite national accounts that integrate estimates of the imputed value of non-market work into measures of extended Gross Domestic Product (GDP) (Van de Ven et al., 2018). Initial efforts to construct such estimates in the U.S. did not make use of the ATUS data on in-your-care time (Landefeld et al., 2009; Bridgman et al., 2012, 2022). A more recent estimate that includes consideration of this category of childcare, assigning it a relatively low replacement cost, reports very large quantitative effects, increasing estimated GDP by 43% rather than previous adjustments of about 26% (Suh and Folbre, 2017). Inclusion of in-your-care time also influences comparisons of the living standards of families with and without children (Folbre et al., 2018).
The complexity of childcare
Quantitative findings from time use surveys have helped transform the meaning of work and enrich explanations of gender inequality. Ease of quantification, however, has often trumped conceptual inquiry, with particularly unfortunate consequences for the measurement and valuation of childcare in the U.S. The historical account provided here reveals the persistent influence of embedded assumptions that childcare is not really work. It also shows how a design innovation introduced by the American Time Use Survey provides a way of quantifying the temporal demands of responsibility for children.
Attention to the time parents spend with a child “in their care” is an important step toward acknowledgment of the complexity of their contributions. Still, it raises as many questions as it answers. Time-diaries that ask respondents what they did on the previous day may provide a reliable snapshot of national averages, but they do not capture the lived experience of parental responsibility—coping with unexpected disruptions, emergencies, and pandemics (Doucet, 2022). Young children complicate overall household management—a demanding task that is often conducted in short intervals over the course of a day and probably under-reported (Winkler and Ireland, 2009).
Even when the time that parents spend with their children is unencumbered by specific responsibilities, it has important consequences for social and emotional outcomes. Attention to explicit developmental activities such as reading aloud to children should not distract from the bigger picture, and racial and ethnic differences in parenting priorities warrant careful consideration (Milkie, 2022). Efforts to combine qualitative with quantitative methods could yield a richer picture of the meaning of “care” as well as the meaning of “work.”
In affluent countries like the U.S. the clock is always on the wall, on the wrist, or in the phone. It will, almost by definition, remain an important tool of time use research. Yet as innovations in survey design demonstrate, it is important to look beyond the clock, the better to look back to better ways of using it.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
