Who pays? the visible and invisible costs of child care

Marcia K. Meyers; Alesha Durfee

doi:10.1215/00161071-29-1-109

Who pays? the visible and invisible costs of child care

Marcia K. Meyers, Alesha Durfee

Research output: Contribution to journal › Review article › peer-review

1 Scopus citations

Abstract

Although the majority of young children now spend time in nonparental child care, we know relatively little about who provides this care and how its costs are distributed among parents, government, and other family members. In this article we use data from a survey of New York City families with children younger than six to estimate the contribution of parental expenditures, government assistance, and the market value of "donated" caregiving time by family, friends, and relatives. We conclude that uncompensated caregivers provide a substantial share of child care that is "invisible" in conventional economic measures.

Original language	English (US)
Pages (from-to)	109-128
Number of pages	20
Journal	Politics and Society
Volume	34
Issue number	1
DOIs	https://doi.org/10.1215/00161071-29-1-109
State	Published - Mar 2006

Keywords

Caregiving
Child care
Gender
Social policy
Subsidies

ASJC Scopus subject areas

Sociology and Political Science
Social Sciences (miscellaneous)
Political Science and International Relations

Access to Document

10.1215/00161071-29-1-109

Cite this

@article{b56bdfe37f17420485e61a14e50d82cb,

title = "Who pays? the visible and invisible costs of child care",

abstract = "Although the majority of young children now spend time in nonparental child care, we know relatively little about who provides this care and how its costs are distributed among parents, government, and other family members. In this article we use data from a survey of New York City families with children younger than six to estimate the contribution of parental expenditures, government assistance, and the market value of {"}donated{"} caregiving time by family, friends, and relatives. We conclude that uncompensated caregivers provide a substantial share of child care that is {"}invisible{"} in conventional economic measures.",

keywords = "Caregiving, Child care, Gender, Social policy, Subsidies",

author = "Meyers, {Marcia K.} and Alesha Durfee",

note = "Funding Information: Meyers Marcia K. School of Social Work and Evans School of Public Affairs, University of Washington, 4101 15th Ave NE, Seattle, WA 98105; 206-616-4409; mkm36@u.washington.edu Durfee Alesha Women and Gender Studies at Arizona State University 03 2006 34 1 109 128 Although the majority of young children now spend time in nonparental child care, we know relatively little about who provides this care and how its costs are distributed among parents, government, and other family members. In this article we use data from a survey of New York City families with children younger than six to estimate the contribution of parental expenditures, government assistance, and the market value of “donated” caregiving time by family, friends, and relatives. We conclude that uncompensated caregivers provide a substantial share of child care that is “invisible” in conventional economic measures. gender child care caregiving social policy subsidies sagemeta-type Journal Article search-text 10.1177/0032329205284759POLITICS & SOCIETYMEYERS and DURFEE WHO PAYS? THE VISIBLE AND INVISIBLE COSTS OF CHILD CARE MARCIA K. MEYERS ALESHA DURFEE Although the majority of young children now spend time in nonparental child care, we know relatively little about who provides this care and how its costs are distrib- uted among parents, government, and other family members. In this article we use data from a survey of New York City families with children younger than six to esti- mate the contribution of parental expenditures, government assistance, and the market value of {"}donated{"} caregiving time by family, friends, and relatives. We con- clude that uncompensated caregivers provide a substantial share of child care that is {"}invisible{"} in conventional economic measures. Keywords: gender; child care; caregiving; social policy; subsidies Families'use of nonparental child care has risen steadily for several decades in the United States, reflecting both increasing rates of maternal employment and changing norms about children's early education and school readiness. Although some parents report no use of regular nonparental care, often arranging employ- ment schedules to share caregiving between parents, about two-thirds of all chil- dren younger than age six and nearly three-quarters of those with employed moth- ers are now in some form of nonparental care on a regular basis.. As many as 2.5 Support for this research was provided by the Russell Sage Foundation and the Katherine T. Mac- Arthur Foundation. We are grateful for the superb assistance of Sandra Garcia in preparing data and measuresfortheanalysis.Allerrorsofomissionandcommissionaretheresponsibilityoftheauthors. Please address correspondence to Marcia K. Meyers, Associate Professor, School of Social Work and Evans School of Public Affairs, University of Washington, 4101 15th Ave NE, Seattle, WA 98105; 206-616-4409; mkm36@u.washington.edu. POLITICS & SOCIETY, Vol. 34 No. 1, March 2006 109-128 DOI: 10.1177/0032329205284759 {\textcopyright} 2006 Sage Publications 109 million individuals are paid to provide this care and nearly as many may provide care on a regular basis without compensation.2 Given its growing social and economic importance, our knowledge about non- parental child care is surprisingly thin. There are limited data, and little consensus among analysts, about the sizeof the child care sector, the costs of producing child care services, or the distribution of these costs among parents, government, and familymembers and friends who {"}donate{"} their timeas uncompensated child care providers. Limitations in our knowledge reflect both theoretical challenges associated with defining and measuring the value of care work and empirical challenges relating to the lack of appropriate data and measures. With limited knowledge about who provides and who pays for child care, ithas been difficultto estimateits social and economic contributions, and nearly impossible to evaluate public and private contributions to its provision. In this article we address some of these gaps in our knowledge by estimating both the more- and less-visible costs of nonparental child care used by families with children younger than the age of six. Our analysis expands on prior work that has estimated the time children spend in nonparental care by monetizing the pub- lic and private costs of providing this care. Using detailed data collected as part of the New York Social Indicator Survey, in combination with city- and state-level administrative data, we estimate direct parental expenditures, the market value of services and subsidies provided by government, and alternative measures of the market value of uncompensated care by family, friends, and neighbors. In doing so, we provide new estimates of total direct expenditures by parents and govern- ment and argue that these underestimate the full economic value of nonparental child care. Unpaid family, friends, and relatives provide a significant share of all nonparental care that is largely invisible in conventional economic measures. If this labor is valued in terms of either market replacements for their services or compensation for their forgone wages, it would represent one-third to one-half of the total public and private contributions to the provision of nonparental child care. BACKGROUND Limitations in our knowledge about the costs of nonparental care are due, in large part, to the historic invisibility of caregiving work. As Nancy Folbre observes, Adam Smith's {"}invisible hand{"} of the market functions only as long as the {"}invisible heart{"} of caregiving relationships functions to produce and sustain each new generation of workers and citizens.3 Although caregiving work--what Folbre describes as {"}the production of people by means of people{"}--provides essential human resources for economic and social activities, it is largely invisible in conventional measures of productive economic activity. As long as this work remains invisible, many of its costs will remain invisible as well. 110 POLITICS & SOCIETY Measuring and Valuing Caregiving Work Questions about the public and private costs of providing nonparental child care relate to broader scholarly debates about the nature of care work and caring labor. In a recent review article, Paula England summarizes an emerging consen- sus among scholars about the characteristics of care work, including the shift in recent decades of an increasing amount of this work from the unpaid labor of women in the home to the paid labor of nonfamilial caregivers; the highly con- tested status of care work as a basis for state support; and the persistence of eco- nomic penalties, both for unpaid family caregivers and for those working in care work occupations.4 As she observes, reaching consensus about {"}the theoretical apparatus that explains the source of these empirical regularities{"} remains chal- lenging.5 The conceptual challenge of estimating the cost of providing care begins with the definition and measurement of this labor. Historically, most caregiving for children, elderly, disabled, and other dependent individuals was provided through the uncompensated labor of family and community members. This labor has been treated in economic theory as a form of nonmarket household production, as a welfare service, or, among labor economists, as a component of the {"}leisure{"} time that adults trade for waged work in the market. As a result, much of the adult labor devoted to caregiving has been excluded from conventional measures of eco- nomic activity. Nowhere is this more evident than in the official measure of eco- nomic productivity for the country--the Gross National Product (GNP)--which explicitly excludes unrenumerated household production. Given its exclusion from conventional measures of economic activity, it has been difficult to measure adults' caregiving labor. Researchers who have exam- ined this labor using time diaries conclude that the time devoted to unpaid domes- tic and caring work in the home is substantial--approximately equal to the time devoted to paid work in the market.6 Counting hours of labor devoted to unpaid carework still fails to capture its economic and social value, however. Some ana- lysts have estimated the value of uncompensated home production in terms of market replacements for these services--that is, as the cost of hiring a third party to perform them. These estimates suggest that the market value of this labor may equal as much as one-half or more of the GNP in severalindustrializedcountries.7 Calculating the market cost of replacing household labor may still underesti- mate its full value if this labor is undervalued in the market. Feminist scholars have argued for many years that women's caregiving labor provides social and economic benefits that are uncompensated when provided in the home and under- valued when provided through market exchanges. This undervaluation can be understood as a pervasive devaluation of women and labor that is associated with women's traditional roles as mothers and nurturers. It can also be understood in terms of a market failure related to the production of public goods. The market failure argument is particularly relevant to child care labor. In standard economic MEYERS and DURFEE 111 terms, children are public goods in the sense that the benefits of their future capa- bilities--as productive workers and citizens--are widely shared, but the costs of producing these capabilities are impossible to allocate to individual beneficia- ries.8 Economic theory predicts that markets will poorly compensate those who provide such services because it is impossible to prevent others from {"}free-rid- ing{"} by enjoying the benefits of healthy children without paying the costs of producing them. There is substantial empirical evidence that the gendered distribution of care- giving work, and its production of widely dispersed and nonexcludable benefits, do impose costs on caregivers that are not compensated through the market. Women who interrupt their employment to bear and raise children incur a sub- stantial motherhood or caregiving penalty in their wages, relative to both men and to women who do not have children.9 England has extended this line of research to document a wage penalty of 5 to 10 percent for individuals employed in an occupation involving caregiving, controlling for human capital, job skill, and occupational characteristics.10 The reasons for these penalties are multiple and interacting, reflecting women's career interruptions for caregiving, gendered occupational structures, institutionalized norms about caregiving work, and the selection of individuals, overwhelmingly women, into occupations that may pro- vide intrinsic rewards to compensate for the lack of economic remuneration.11 The result of these penalties is the imposition of private and largely invisible costs on caregivers in the form of forgone wages and earnings that are lower than what they might command, given their skills and experience, in other occupations. Prior Research Defining, measuring, and valuing caregiving work raises substantial concep- tual challenges. It also raises formidable empirical challenges that have limited our knowledge about the size of the child care sector and workforce, public and private expenditures for child care services, and the private costs that child care providers incur through their donation of uncompensated labor and through care penalties in their earnings. The first difficulty arises in the measurement of child care services and the size of the workforce. Because it is not provided through regular market exchanges, a substantial share of nonparental child care is invisible in conventional economic measures. Some nonparental care is provided as a market good in settings that are regulated by state and local government, including private child care centers, preschools, nursery schools, and licensed family child care homes. Other care is provided outside of the market through public education and welfare programs. Stillother, less formal forms of care are provided by home-based providers and by uncompensated family, friends, and neighbors. Data collected at the household level suggest that families use these more- and less-visible forms of child care almost equally. In the approximately two-thirds of 112 POLITICS & SOCIETY families with children younger than age six who use any care, children are about equally likely to be in child care centers or in relative care as their primary arrangement.12 Considering any use of relative care--as a primary arrangement or in combination with other forms of care--Snyder et al. estimate that more than a third of all children younger than age five are cared for by relatives on a regular basis while their parents are at work.13 Conventional economic measures are poorly designed to measure this diver- sity of arrangements. Warner describes recent efforts to study the child care sector as akin to observing an iceberg in which data on the {"}the formal licensed or regu- lated care that we see above the water line{"} capture only a small portion of the much larger, less-visible whole.14 The most visible market-based services are measured in Census data on establishments and workers covered by unemploy- ment insurance. When analysts compare these estimates to data from state regula- tory agencies, however, they typically find between 20 and 300 percent more employees at work in regulated child care settings. These estimates of the workforce more than double again if unregulated home-based providers are included through the use of self-employment tax information provided by the Internal Revenue Service. However, these expanded estimates still fail to capture the unknown number of individuals who are paid to provide care within the infor- mal economy, but do not report this self-employment income for tax purposes, or the family, friends, and neighbors who care for children on a regular basis without compensation. In a recent effort to address the limitations of these supply-side approaches, Burton et al. employ a demand-side approach to estimate the size of the child care workforce using household-level surveys of child care use.15 They estimate that the paid workforce caring for children younger than age six in the United States includes over 2.3 million individuals at a point in time--approximately 600,000 more than are identified in the Current Population Survey and more than twice the number tracked by the Bureau of Labor Statistics. They conclude that as a group, relative and nonrelative caregivers (other than family child care providers) make up nearly half of the paid child care workforce. Even more strikingly, they estimate that an additional 2.4 million individuals, 93 percent of whom are rela- tives, provide care for children younger than six during a given week without compensation. Given the invisibility of much nonparental child care, it is not surprising that knowledge about its costs is also limited and incomplete. Parents pay a portion of these costs through market exchanges; government pays another portion through social welfare and educational spending. Although it is possible to measure some of these expenditures using household-level survey data and aggregate data on government spending, it has been nearly impossible to combine these data to esti- mate the relative contributions of parents and government. If child care is measured as a market good, at least a portion of its costs can be estimated using data on parents'direct expenditures to purchase care or pay indi- MEYERS and DURFEE 113 vidual providers. Nearly all parents who use child care centers, family child care, or in-home caregivers pay for those services. An estimated 33 percent of those who use relative caregivers also pay for that care.16 How much do families pay for care? Giannarelli and Barsimantov, using the National Survey of America's Families (NSAF), estimate that about one-half of all families with employed mothers and children younger than age thirteen incur some child care expenses; among these families, expenditures average $286 per month, or about 9 percent of family earnings.17 This estimate is limited in several respects. It does not capture tax benefits that reduce parents'net expenses. It also fails to reflect expenditures by familiesin which mothers are not employed. Using the National Household Education Survey, Hofferth et al. estimate that although they are less likely to use nonparental care, more than a third of families in which mothers are looking for work or not in the labor force report some hours of nonparental care for preschool children.18 These families actually pay more per hour for this care than families in which mothers are employed, which may reflect their use of more expensive preschool programs or the higher costs of part-time care. These out-of-pocket parental expenditures capture only a portion of the com- pensated costs of nonparental child care. Federal, state, and local governments provide additional child care directly through public child care and early educa- tion services, for example, the federal Head Start and state pre-Kindergarten pro- grams. They provide other support through means-tested subsidies for parents' purchase of private services, primarily through the Temporary Assistance for Needy Families (TANF), Child Care and Development Fund (CCDF), and Social Services (SSBG) block grants. Less directly, governments make tax expenditures through policies that allow parents to deduct a portion of their private child care expenses from income taxes through the federal Child and Dependent Care Tax Credit and state-level tax credits in several states. Other government programs reduce the operating expenses of child care facilities through, for example, subsi- dized child care food programs and targeted tax credits. Data on these public expenditures are surprisingly limited and nearly impossi- ble to analyze at the household or facilitylevel. Child care programs are funded by several federal, state, and local agencies housed in various social service, welfare, tax, and early education agencies and no public agency tracks all of these outlays. Meyers et al. estimate that federal and state spending for preschool services and means-tested subsidies, and federal tax expenditures for individual child care tax benefits, exceeded $20 billion as of 2000.19 This estimate fails to capture signifi- cant public expenditures, however, including subsidized food programs, state supplementation of federal Head Start services, and federal and state tax benefits for providers. Aggregate government spending data tell us something about what govern- ment contributes to the production of child care. They tell us little about the share of total costs that are paid by government, because they are difficult to link to 114 POLITICS & SOCIETY either household-level child care consumption or to establishment- and provider- level production of these services. In one of the few recent estimates of house- hold-level receipt of government assistance, Giannarelli and her colleagues use the National Survey of America's Families (NSAF) to estimate that about 12 per- cent of employed families with a child younger than age 13 receive some type of nontax assistance from government or other organizations.20 Because parents are unlikely to know the value of government subsidies and programs, however, the NSAF does not collect data on the amount of this assistance that can be compared to parental expenditures. Parental and government expenditures are even more limited as a measure of the costs of nonparental child care by their exclusion of the {"}donation{"} of labor by the unpaid family, friends, and neighbors who make up an estimated one-half of the total workforce caring for children. We know something about the number of children who are cared for in these arrangements, but little or nothing about the amount of labor that is supplied by uncompensated caregivers or its value in market terms. Estimating the cost of replacing the donated labor of uncompensated family, friends, and neighbors with paid child care would provide a more complete accounting of the public and private costs of child care. It would still underesti- mate the full value of this care, and its costs from the providers'perspective, if the wages paid to child care providers are lower than the wages they could earn in alternative employment--that is, if they incur a {"}care penalty{"} for their employ- ment in a caregiving occupation. The exceptionally low wages of the child care workforce suggest that these workers do pay some penalty for their occupational location. Child care workers are among the lowest paid workers in the U.S. labor market. Considering only the share of the child care workforce captured in BLS surveys ({"}child care workers{"} and {"}preschool teachers{"} in institutional settings), analysts at the Center for the Child Care Workforce estimate that only eighteen occupations have lower mean wages than child care workers.21 The earnings of licensed family child care pro- viders and individual providers are even lower. The wages of child care workers are particularly low in comparison to those of other workers with similar educa- tional backgrounds and, despite the substantial increase in demand for child care services in the 1980s and 1990s, were an estimated one dollar per hour lower in 1997 in real dollars than they were ten years earlier.22 The possibility that child care workers incur a {"}care penalty{"} in their wages suggests that these same individuals might command a higher wage in other occu- pations. If so, our estimates of the cost of child care based on current caregiver wages will underestimate the full costs and value of these services. In one effort to discern how this {"}care penalty{"} may bias estimates of the cost of producing child care, Helburn and her colleagues estimate an adjustment to the wages of workers in child care centers to reflect their predicted wages in other occupations, given their age, gender, education, and other characteristics.23 By their estimate, nearly MEYERS and DURFEE 115 20 percent of the full cost of providing these services is contributed by workers through {"}forgone wages{"}--that is, the difference between actual wages and the wages that they could command in other occupations. England et al. estimate that among all care occupations, child care workers incur the largest wage penalty-- 41 percent--relative to employment in occupations not involving care work after adjusting for human capital, occupational skills demands, and other occupation and industry characteristics.24 RESEARCH QUESTIONS AND APPROACH Although a majority of families use nonparental child care on a regular basis, much of the labor involved in providing this care is invisible in standard economic measures, its full social value may be underestimated, and the distribution of its costs among parents, government, and uncompensated caregivers is largely unknown. In this article we address some of these gaps in our knowledge by using unusually detailed family-level survey data to consider the contributions of par- ents, government, and uncompensated caregivers to the provision of nonparental child care. We go beyond existing studies of the use of child care arrangements, or of time devoted to caring for children, to estimate the visible and less visible costs associated with providing this care. Combining survey data with administrative data from a variety of sources, we estimate (1) the share of families that incur out-of-pocket expenses for non- parental care, receive any form of government assistance, and use unpaid care by family, friends, and neighbors on a regular basis; (2) average family expenditures, the value of government assistance, and the market value of the donated labor of unpaid caregivers under different assumptions about their compensation; and (3) the resulting share of the total economic value of nonparental care that is contrib- uted by parents, government, and uncompensated family, friend, and neighbor caregivers. CONTEXT: CHILD CARE IN NEW YORK CITY Conducting these analyses with data from a single city has the advantage of allowing us to use administrative data for detailed imputations of the value of assistance. The primary disadvantage is that the data represent only one popula- tion and policy environment, potentially limiting generalizability to the United States as a whole. New York City provides a case study of public and private child care costs in the context of relatively generous government assistance and relatively con- strained family resources. New York Stateis one of only six statesthathave autho- rized pre-Kindergarten services for all four-year-olds, although the delivery of these services has lagged behind the promise. It is one of twenty-six states that provide a child care tax credit and one of eleven in which this credit is refundable. 116 POLITICS & SOCIETY Although spending for means-tested child care assistance is also high in absolute terms--New York City provided an estimated $230 million in means-tested sub- sidy assistance in fiscal year 19992000--the city is similar to other jurisdictions in being unable to serve all eligible families and rationing assistance. In terms of comparability to the United States as a whole, analyses by the Urban Institute indicate families in New York State are about as likely as those nationwide to have a child younger than five in nonparental care and, on average, use similar care arrangements.25 Relative to other families, however, families with total incomes under 200 percent of poverty in New York are less likely to be using center-based care and more likely to use relative care as a primary arrangement. This may reflect the {"}bricks and mortar{"} difficultiesof establishing new child care facilities given the City's density and real estate prices. The proportion of families with an employed mother and a child younger than thirteen who receive some form of nontax child care assistance from government or other organizations is higher in New York (16 percent) than nationwide (13 percent), which is consistent with the availability of both subsidy and preschool assistance.26 Family resources may also be more constrained in New York City than in the country as a whole. Child care costs are higher in New York than they are for the remainder of the United States; in 1999, families in New York State who paid for child care spent an average of $348 per month, more than the average of $303 spent by families nationwide.27 The proportion of children living in poverty in New York State is also higher than the rest of the country. Nearly one-quarter (22 percent) of New York children lived in poor families in 1998, compared to 18 percent of children nationwide.28 DATA AND MEASURES Data and Sample Data for this analysis are taken from waves II and III (1999 and 2002) of the New York Social Indicators Survey (NYSIS) and from a variety of city and state administrative sources. The NYSIS is a telephone survey administered to a ran- dom sample of New York City households by the Columbia University Social Indicators Research Center. The approximately thirty-minute survey covers a number of aspects of individual and family well-being, including adequacy of and satisfaction with institutional supports, family economics, living conditions, edu- cation, health, and employment. Each wave of the NYSIS collectsdatafrom a rep- resentative sample of approximately 1,500 randomly selected families in the City using computer-assisted telephone interview (CATI) technology and random- digit dialing (RDD). For our analyses, we use data on 646 families with at least one child between the ages of zero and five. {"}Family,{"} as defined by the NYSIS, includes a respondent, his or her resident spouse or partner, and any resident chil- dren between the ages of zero and seventeen. All estimatesare reported atthe fam- MEYERS and DURFEE 117 ily level and weighted, using Census data, to represent families with children in New York City.29 Measures Waves II and III of the NYSIS included a special module that collected unusu- ally detailed questions about all nonparental child care arrangements, child care expenditures, and receipt of governmental assistance (including tax credits, pub- lic preschool, and subsidies) in the prior twelve months. We make use of these data to estimate the amount of various types of nonparental child care used in the prior year along with the receipt of various forms of financial and nonfinancial assistance for this care. Because parents are unlikely to know the value of many forms of government assistance, we use administrative data for New York City or State, as appropriate, to impute the value of various benefits. To estimate the value of unpaid care provided by family, friends and neighbors, we estimate the total hours used and impute both the replacement cost (of purchasing equivalent mar- ket services) and the value of the caregivers' labor using alternative assumptions about caregiver compensation. Child Care Arrangements Nonparental care is defined as regular care for children by someone other than parents. Information was collected from parents with any children younger than thirteen on each type of nonparental care used for any child(ren) in the family and on the number of child(ren) in that form of care. Respondents were asked how many months the child(ren) were in each type of care and whether it was on a full- or part-time basis. The timethatindividuals devoted to providing care--for exam- ple, as uncompensated child care providers--was defined as the time that they cared for one or more children in the family. Parental Out-of-Pocket Expenditures Parental expenditures are estimated directly from survey responses to detailed questions about out-of-pocket payment for all forms of care, for all children, in the prior twelve months. We adjust these expenditures by subtracting the amount reimbursed through Federal or statechild care tax credits. Item-missing values are imputed using family demographics, including family structure, income level, education level, race, ethnicity, and immigration status. Governmental Assistance The receipt of specific forms of government assistance is estimated directly from survey questions about the use of any public preschool or other child care programs, the receipt of any subsidies (by parents or providers), and the use of federal and state child care tax credits in the previous twelve months. Questions about government assistance were asked for each type of care arrangement, a 118 POLITICS & SOCIETY more complete accounting of this assistance than has been available in other sur- veys. The value of government assistance is calculated by imputing and summing the value of tax credits, subsidies and/or vouchers, and Head Start/public pre- school for families reporting the receipt of governmental assistance. The value of tax credits is estimated using the TAXSIM model provided by the National Bureau of Economic Research. The TAXSIM program applies federal and state tax rules to estimate the value of both refundable and nonrefundable child care tax benefits using family-level data on income, family composition, child care expenditures, and other family-level characteristics. The value of child care subsidies is estimated by first calculating the annual hours of subsidized care received by all children in the family in the prior twelve months and then imputing the maximum subsidy reimbursement availablein New York City on the basis of the most recent New York Market Rate Survey. The sur- vey is conducted by New York State to estimate the market costs of various forms of care in each region of the state, and is used to set reimbursement rates for assis- tance through the TANF, CCDF, and other subsidy programs. These imputations adjust for the regional price of care by type of setting, the age(s) and number of child(ren), whether that care was provided on a full- or part-time basis, and the number of months each child was in the subsidized setting. To estimate the value of Head Start and public preschool services, we use survey data to determine how many children were enrolled in either type of program during the prior twelve months, the number of months enrolled, and whether for a full- or part-day pro- gram. We impute the value using governmental cost methods--that is, distribut- ing total public expenditures (for direct services) among all recipient children to determine costs on a per-child basis. Value of Care Donated by Family, Friends, and Neighbors A substantial share of the families in our sample report using {"}free{"} care pro- vided by family, friends, and neighbors on a regular basis for one or more of their children. To estimate the cost of this care we employ two approaches: estimating the replacement cost of purchasing equivalent market services and estimating the opportunity costs or forgone wages of caregivers in terms of alternative alloca- tions of their under different assumptions about hourly wages. We first estimate the total number of hours of uncompensated care that fami- lies received in the prior year, using survey questions about the number of chil- dren cared for and the length of the care arrangement (average weekly hours and the number of weeks used) in the prior twelve months. Our first approach to esti- mating the market value of this care assumes that the replacement cost would be similar to the cost of using family child care. As with child care subsidies, we use the New York Market Rate Survey to impute the value of family child care, adjust- ing for the amount of care provided. We use the rates for preschoolers (threefive years old) because the NYSIS does not contain specific information as to which child was cared for in these arrangements. Based on the number of hours reported, MEYERS and DURFEE 119 we apply the rate for either full-time ($103 per week) or part-time ($67 per week) care for each child in care. Our second approach assumes that uncompensated caregivers face opportu- nity costs when allocating their time to unpaid care work rather than market work. The NYSIS does not provide sufficiently detailed data on caregivers to estimate the hourly wage that they would be likely to command in other occupations. Instead, we first impute their potential earnings at the New York State minimum wage of $5.15 per hour. We assume that this is the minimum that they would earn were they to allocate their time to alternative employment. By using the minimum wage we assume that, if paid for providing care, indi- viduals would incur the {"}care penalty{"} associated with working in child care. To estimate the private costs associated with this penalty we also impute a higher hourly wage for their time. In the absence of data with which to predict an individ- ual wage, we apply the {"}living-wage{"} level that New York City has recently adopted as a minimum for all city employees and contractors.30 When enacted in 2002, the living-wage rate was $8.10 per hour for employees with health benefits, with an additional allowance of $1.50 per hour for employees who do not have health benefits. Wage-based imputations are adjusted for the number of weeks that parents reported using the care and set the providers'hours at either full-time (forty hours per week) or part-time (twenty hours per week). Unlike imputations based on the cost of purchasing family child care, the wage-based estimates do not adjust for the number of children in care. We calculate the share of average costs that are paid by parents and govern- ment and the estimated value of donated caregiving time as the share of the weighted average, at the family level, contributed by each. RESULTS Sample Characteristics In the NYSIS sample, weighted to represent the population of families with any children younger than six in New York City, about one-half of families have children younger than three (Table 1). Nearly 40 percent of families are headed by Hispanic parents and about one-quarter are headed by both white non-Hispanic and African American parents. Just over one-quarter of the families have income at or below the federal poverty line. In about one-half of allfamilies,mothers were currently employed, in school, or in job preparation activities. Sample character- istics are similar for those families who did and did not use some form of nonparental care, with the exception of higher rates of maternal employment (59 versus 48 percent) among those reporting some child care use. 120 POLITICS & SOCIETY Use and Payment for Nonparental Child Care When we consider all forms of care used on a regular basis for children youn- ger than six, the large majority of New York City families with young children use some form of nonparental care during the year (Table 2). Among all families with young children, more than 80 percent used some form of care. About one-third of all families, and nearly one-half of those using any care, use formal care in a child care center, family child care home, or early education or preschool programs. Substantially more families--63 percent of all and 84 percent of child care users--use care that is provided by family, friends, or neighbors. The majority of parents who use child care pay something for this care out-of- pocket (59 percent); 20 percent pay the full amount. Fewer of these families receive some form of assistance from government (55 percent). Federal and state tax credits are the most common form of assistance (35 percent), followed by enrollment of children in public Head Start or pre-K programs (24 percent). Con- siderably fewer familieswho use care receivegovernment subsidies (12 percent). As described above, the large majority of all families who use any nonparental child care use some care provided by family, friends, and neighbors. Although parents pay for a portion of this care, and a small share is subsidized by govern- ment, the majority is donated without charge to either parents or government. MEYERS and DURFEE 121 Table 1 Sample Characteristics (Means and Standard Deviations, Weighted), New York City Families with any Child Aged Zero to Five All Families (N = 646) Families Using Nonparental Child Care (N = 526) Standard Standard Mean Deviation Mean Deviation Any infants in family (age 0 to 2) 0.55 0.50 0.54 0.50 Youngest child older than 3 0.45 0.50 0.46 0.50 Number of children age 0 to 5 in family 1.10 0.37 1.10 0.37 Respondent is Hispanic 0.38 0.49 0.35 0.48 Respondent is black, non-Hispanic 0.26 0.44 0.31 0.46 Respondent is white, non-Hispanic 0.22 0.41 Respondent's education < high school degree 0.21 0.40 0.14 0.34 Respondent's education = high school degree 0.35 0.48 0.37 0.48 Respondent's education = some college or more 0.43 0.50 0.49 0.50 Family earnings less than 100% Federal Poverty Level 0.34 0.47 0.37 0.48 Family earnings 101200% Federal Poverty Level 0.23 0.42 0.21 0.41 Family earnings 201400% Federal Poverty Level 0.28 0.42 0.25 0.44 Family earnings more than 400% Federal Poverty Level 0.20 0.40 0.16 0.37 Mother is employed, in school, or in job training 0.49 0.50 0.58 0.49 Respondent is in Wave 2 0.51 0.50 0.49 0.50 Source: New York City Social Indicators Survey, Waves II and III. Note: Means calculated using weighted data. One-third of all families, and nearly one-half of child care users, use some care that is provided without charge by family, friends, or neighbors. Value and Distribution of Child Care Services Detailed data on the hours and types of care used for each child in the family in the prior year allow us to estimate the total economic value of care under various assumptions (Table 3). If we estimate the value of nonparental child care on the basis of parental expenditures alone, New York Cityfamiliespay a littlemore than $2,000 on average per year. This is lower than the annualized estimate that Giannarelli et al. report for New York State,31 which may reflect the restriction of their sample to families with employed mothers (who may use and pay for more care) and their inclusion of care for children older than six (who may be in care for fewer hours per day). When we include our estimates of the value of government assistance, the average value of child care per family is closer to $4,000 per year and expendi- tures for this care are split nearly evenly between parents and government. Our estimates of the share of all child care using families who receive assistance are higher than those reported in prior studies, which is likely due to our ability to do a more complete accounting for public child care programs, subsidies, and tax ben- efits. Our estimate of the share of direct expenditures paid by government pro- 122 POLITICS & SOCIETY Table 2 Frequencies of Child Care Usage, Costs, and Receipt of Assistance (Means and Standard Deviations, Weighted), New York City Families with any Child Aged Zero to Five All Families (N = 646) Families Using Nonparental Child Care (N = 526) Standard Standard Mean Deviation Mean Deviation Uses formal child care 0.35 0.48 0.47 0.50 Uses Head Start or public pre-K 0.18 0.38 0.24 0.43 Uses care provided by family, friends, or neighbors 0.63 0.48 0.84 0.36 Uses care donated by family, friends, or neighbors 0.36 0.48 0.49 0.50 Pays some amount out-of-pocket for nonparental child care 0.44 0.50 0.59 0.49 Pays all costs out-of-pocket for nonparental child care 0.15 0.36 0.20 0.40 Receives child care subsidy 0.09 0.28 0.12 0.32 Receives child care tax credit 0.25 0.42 0.35 0.46 Receives any governmental assistance with nonparental child care 0.41 0.49 0.55 0.50 Receives governmental assistance with formal child care (other than Head Start or public pre-K) 0.07 0.26 0.10 0.30 123 Table3 EstimatedValuesofParental,Government,and{"}Donated{"}ContributionstoChildCare(CalculatedattheFamilyLevel,Weighted),NewYorkCityFamilieswithanyChild AgedZerotoFiveUsingNonparentalChildCare(N=526) TotalValue TotalValueofChildCare TotalValueofChildCareDonatedby ofChildCareProvidedFamily,Friends,ValueofEstimated PaidOut-ThroughandNeighbors,ParentGovernmentCostof TotalValueof-PocketGovernmentIfValuedatExpendituresAssistanceDonatedCare ofChildCarebyParentsAssistanceMarketRatesasShareofasShareofasShareof (PerFamily)(PerFamily)(PerFamily)(PerFamily)WeightedTotalWeightedTotalWeightedTotal MeanMeanMeanMeanMeanMean Parentalout-of-pocketexpendituresonly$2,041$2,041$1,949------------------------ Parentalexpendituresplusvalueof governmentassistance$3,990$2,041$1,949------51percent49percent------ Parentalexpenditures,governmentassistance, plusdonatedcare... valuedatthecostoffamilychildcare$5,788$2,041$1,949$1,78335percent34percent31percent ...valuedatthecostofpayingtheprovider atminimumwage$6,792$2,041$1,949$2,77030percent29percent41percent ...valuedatthecostofpayingtheprovider theNewYorkCity {"}LivingWage{"}$9,684$2,041$1,949$5,66221percent20percent59percent Source:NewYorkCitySocialIndicatorsSurvey,WavesIIandIII. vides a high estimate of the distribution of these costs between parents and gov- ernment, given the relatively extensive provision of public preschool in New York City and the availability of child care tax credits in the state. When we add our estimates of the economic value of the {"}invisible{"} care donated by family, friends, and neighbors, we reach substantially different con- clusions about public and private contributions to child care. If the child care cur- rently donated by family, friends, and neighbors were replaced with family child care purchased at the average market rate, the total economic value of the child care used would increase by about one-third per family. This estimate reflects the very low wages earned by child care workers, however, and family child care pro- viders in particular. If the family, friends, and neighbors who provide care to New York City families earned even the New York State minimum wage of $5.15 per hour for the labor they currently provide for free, the total value of nonparental care would be more than $6,700 per family per year, with more than 40 percent of this total provided by these caregivers. If these providers were paid what New York City deems a living wage for city employees and contractors, the total value of child care would be nearly 2.5 times higher than the expenses currently covered by parental and government expenditures. Government assistance accounts for only 20 percent of this total. Nearly 80 percent of the economic value of non- parental child care is provided privately, by parents (21 percent) and by the esti- mated wages of family, friends, and neighbors who provide care (59 percent). DISCUSSION Much of the adult time and labor that are devoted to the care and nurturing of young children is rendered invisible by conventional measures of economic activ- ity and output. Most of this labor is provided by parents and its value is entirely uncounted in measures, such as the GDP, that define economic productivity in terms of market exchanges. To an increasing extent, families are substituting nonparental child care for a portion of their own caregiving labor. Like parental caregiving, much of the labor provided by these nonparental caregivers remains invisible because they are not compensated through conventional market ex- changes and because, when paid, they incur a caregiving penalty in their wages. As a result, although some of the costs of nonparental child care are captured in data on parental and government expenditures, these data substantially underesti- mate the total labor devoted to child care, its economic value, and, arguably, its social value. We estimate that about one-third of New York City families with young chil- dren use some form of {"}formal{"} care (in public or private centers, preschools, and family child care homes) during the year and nearly two-thirds use care provided by family, friends, or neighbors on a regular basis. Some of this care by relatives and friends is compensated by parents and/or through government subsidies. But one-third of all families with young children, and about one-half of those using 124 POLITICS & SOCIETY child care, receive some regular care that is {"}donated{"} by these family, friends, and neighbors without direct charge. Assessing the market value of this invisible {"}donated{"} labor substantially alters our estimates of the total economic value of nonparental child care services, as well as the share of the value provided by public and private parties. When we consider only parental out-of-pocket payments and government expenditures for services, subsidies, and tax credits, the value of child care is almost equally dis- tributed between parents and government. If the family, friends, and neighbors who care for children without direct charge were compensated at the rate of fam- ily child care providers, however, we estimate that they would provide about one- third of the total economic value of care. Because family child care workers are among the lowest paid workers in the U.S. workforce, valuing the donated labor of family, friend, and neighbor caregivers at market rates for family child care may underestimate both the market value of their labor, if allocated to other employment, and the social value of their work. If these caregivers could earn even the minimum wage for their hours of work in other employment, they are {"}donating{"} more than 40 percent of the total value of child care. At this wage, many of these individuals would still earn poverty-level incomes, even with full- time employment. New York City has established a {"}living wage{"} standard for city employees and contractors that is $19,968 for full-time, full-year employ- ment. If currently uncompensated caregivers were able to earn this wage for their {"}donated{"} hours of caregiving, their donated labor would account for nearly 60 percent of the total value of care. Under this assumption, only about 20 percent of the value of all child care is contributed by government. Our analysis of the visible and invisible {"}costs{"} of child care is unusual in that it captures the full range of contributions from parents, government, and unpaid caregivers. It is still limited in some important respects. Although we have detailed data on family-level receipt of government assistance, our estimates do not include other forms of assistance--including food subsidies and tax benefits--that are received directly by some child care centers and family child care providers. We adjust our estimates of the market value of donated caregiving time in order to demonstrate the extent to which market prices may undervalue paid caregiving work. We do not adjust for the possible undervaluing of the child care labor that is purchased through parental payments and government assistance. Doing so would raise both the total estimated value of child care and the relative contribu- tions of parents and government. These estimates are also limited in using data from a single city that may not be representative of other parts of the country. In particular, given the relatively gen- erous preschool, tax credit, and subsidy assistance available in New York City, the share of the value of child care provided through government assistance may be higher in our sample than they would be in other regions. Parents in New York also pay more for child care, on average, than parents in other regions, rendering our estimates of relative share more representative. MEYERS and DURFEE 125 Although limited in some respects, this analysis provides one of the first esti- mates of the visible and invisible {"}costs{"} of nonparental child care. It suggests if we fail to recognize and value the {"}invisible{"} labor of uncompensated caregivers, we will seriously underestimate the total economic value of nonparental child care and the private contributions--primarily by family members--to its provi- sion. In doing so, we further devalue this work and reify the conceptualization of it as private and costless. NOTES 1. F. L. Sonenstein, G. Gates, S. R. Schmidt, and N. Bolshun, Primary child care arrangements of employed parents: Findings from the 1999 National Survey of America's Families (Washington, D.C.: The Urban Institute, 2002). 2. A. Burton, M. Whitebook, M. Young, D. Bellm, C. Wayne, R. Brandon, and E. Maher, Estimating the Size and Components of the U.S. Child Care Workforce and Care- giving Population (Washington, D.C.: Center for the Child Care Workforce, 2002). 3. N. Folbre, The Invisible Heart: Economics and Family Values (New York: The New Press, 2001). 4. P. England, {"}Emerging Theories of Care Work,{"} Annual Review of Sociology 31 (2005): 38199. 5. Ibid., 395. 6. L. Goldschmidt-Clermont and E. Pagnossin-Aligisakis, Measures of Unrecorded Economic Activities in Fourteen Countries (New York City: United Nations Development Programme, Human Development Report Office, 2001); J. Gershuny and J. P. Robinson, {"}HistoricalChangesintheHouseholdDivisionofLabor,{"}Demography25(1998):53752. 7. K. Cloud and N. Garrett, {"}A Modest Proposal for Inclusion of Women's Household Human Capital Production Analysis of Structural Transformation,{"} Feminist Economics 2 (1996): 93119; Folbre, The Invisible Heart: Economics and Family Values. 8. N. Folbre, {"}Children as Public Goods,{"} The American Economic Review 84 (1994): 8690; P. England and N. Folbre, {"}The Cost of Caring,{"} The Annals of the American Acad- emy of Political and Social Science 561 (January 1999): 3951; J. C. Gornick and M. K. Meyers, Families That Work: Policies for Reconciling Parenthood and Employment (New York: Russell Sage Foundation, 2003). 9. J. C. Gornick, M. K. Meyers, and K. Ross, Public Policies and the Employment of Mothers: A Cross-National Study. Working Paper No. 140 (Syracuse, N.Y.: Maxwell School of Citizenship and Public Affairs, Syracuse University, 1996); A. Crittenden, The Price of Motherhood: Why the Most Important Job in the World Is Still the Least Valued (New York: Metropolitan Books, 2001); J. Waldfogel, {"}Understanding the 'Family Gap'in Pay for Women with Children,{"} Journal of Economic Perspective 12 (1998): 13756; M. J. Budig and P. England, {"}The Wage Penalty for Motherhood,{"} American Sociological Re- view 66 (2001): 20425. 10. P. England, M. J. Budig, and N. Folbre, {"}Wages of Virtue: The Relative Pay of Care Work,{"} Social Problems 49 (2002): 45573. 11. England, {"}Emerging Theories of Care Work.{"} 12. S. Hofferth, K. Shauman, and R. Henke, Characteristics of Children's Early Care and Education Programs: Data from the 1995 National Household Education Survey (Washington, D.C.: U.S. Department of Education, 1998); Sonenstein et al., Primary child care arrangements of employed parents: Findings from the 1999 National Survey of Amer- ica's Families. 126 POLITICS & SOCIETY 13. K. Snyder and S. Adelman, The Use of Relative Care While Parents Work: Findings from the 1999 National Survey of America's Families (Washington, D.C.: Urban Institute, 2004). 14. M. Warner, {"}The Threefold Importance of Child Care: The Need for a New Eco- nomic Framework,{"} 7. Background Paper for the May 2005 Cornell Workshop, Articulat- ing the Economic Significance of the Childcare Sector. 15. Burton et al., Estimating the Size and Components of the U.S. Child Care Workforce and Caregiving Population. 16. Hofferth et al., Characteristics of Children's Early Care and Education Programs: Data from the 1995 National Household Education Survey. 17. L. Giannarelli and J. Barsimantov, Child Care Expenses of America's Families. Occasional Paper No. 40, Assessing the New Federalism Project (Washington, D.C.: The Urban Institute, 2000). 18. Hofferth et al., Characteristics of Children's Early Care and Education Programs: Data from the 1995 National Household Education Survey. 19. M. K. Meyers, D. Rosenbaum, C. Ruhm, and J. Waldfogel, {"}Inequality in Early Childhood Education and Care: What Do We Know?,{"} In K. Neckerman, ed., Social Inequality. (New York: Russell Sage Foundation, 2004). 20. L. Giannarelli, S. Adelman, and S. Schmidt, {"}Getting Help with Child Care Expenses,{"} Occasional Paper No.62.,Assessingthe New Federalism Project (Washington, D.C.: The Urban Institute, 2003). 21.Center forthe Child Care Workforce. Current DataonChild Care Salaries andBen- efits in the United States. (Washington, D.C.: Center for the Child Care Workforce, 2002); M. Whitebook, C. Howes, and D. Phillips, Worthy Work, Unliveable Wages: The National Child Care Staffing Survey 19881997 (Washington, D.C.: Center for the Child Care Workforce, 1998). 22. D. Blau, The Child Care Problem: An Economic Analysis. (New York: Russell Sage Foundation, 2001). 23. S. Helburn, M. L. Culkin, C. Howes, D. Bryant, R. Clifford, D. Cryer, D. et al., eds., Cost, Quality, and Child Outcomes in Child Care Centers (Denver: Department of Eco- nomics, Center for Research in Economic and Social Policy, University of Colorado at Denver, 1995). 24. P. England, M. J. Budig, and N. Folbre, Wages of Virtue: The Relative Pay of Care Work. 25. Sonenstein, Gates, Schmidt, and Bolshun, Primary child care arrangements of employed parents: Findings from the 1999 National Survey of America's Families. 26. Giannarelli, Adelman, and Schmidt, Getting Help with Child Care Expenses. 27. Ibid. 28. L. Fender, C. T. O'Brien, T. Thompson, K. Snyder, and R. Bess, Recent Changes in New York Welfare and Work, Child Care, and Child Welfare Systems. (Washington D.C.: The Urban Institute, 2002). 29. For information on NYSIS sample, survey, and weighting procedures see http:// www.columbia.edu/cu/ssw/projects/surcent/. 30. New York City Administrative Code, Section 6-109. 31. Giannarelli, Adelman, and Schmidt, Getting Help with Child Care Expenses. Marcia K. Meyers is Professor of Social Work and Public Affairs at the University of Washington. She earned a Master of Public Administration at Harvard University and a Master of Social Work and a doctorate from the University of California Berkeley. Dr. Meyers's research focuses on public policies and programs for vulner- MEYERS and DURFEE 127 able populations, with a particular focus on issues of poverty, inequality, and policy implementation. Current research projects examine the impact of U.S. state policy regimes on the labor force participation of mothers, on inequality in access to early childhood education and care, and on disposable family income. In collaboration with researchers at the Rockefeller Institute of Government, she is studying the front-line delivery of welfare reforms in eleven sites around the country. She is coau- thor of Families That Work: Policies for Reconciling Parenthood and Employment, and her papers have appeared recently in the Journal of Policy Analysis and Man- agement, Journal of European Social Policy, Social Service Review, Journal of Pub- lic Administration Research and Theory, Social Science Quarterly, and Demography. Alesha Durfee is an assistant professor in Women and Gender Studies at Arizona State University. Her research explores the ways that structural inequalities both constrain and are reinforced by individual behavior, and assesses the impacts of social policies implemented in an attempt to ameliorate these inequalities. Her sub- stantive interests include child care utilization and policy,domestic violence and the justice system, and welfare reform. 128 POLITICS & SOCIETY 1. F. L. Sonenstein, G. Gates, S. R. Schmidt, and N. Bolshun, Primary child care arrangements of employed parents: Findings from the 1999 National Survey of America's Families (Washington, D.C.: The Urban Institute, 2002). 2. A. Burton, M. Whitebook, M. Young, D. Bellm, C. Wayne, R. Brandon, and E. Maher, Estimating the Size and Components of the U.S. Child Care Workforce and Caregiving Population (Washington, D.C.: Center for the Child Care Workforce, 2002). 3. N. Folbre, The Invisible Heart: Economics and Family Values (New York: The New Press, 2001). 4. P. England, “Emerging Theories of Care Work,” Annual Review of Sociology 31 (2005): 381–99. 5. Ibid., 395. 6. L. Goldschmidt-Clermont and E. Pagnossin-Aligisakis, Measures of Unrecorded Economic Activities in Fourteen Countries (New York City: United Nations Development Programme, Human Development Report Office, 2001); J. Gershuny and J. P. Robinson, “Historical Changes in the Household Division of Labor,” Demography 25 (1998): 537–52. 7. K. Cloud and N. Garrett, “A Modest Proposal for Inclusion of Women's Household Human Capital Production Analysis of Structural Transformation,” Feminist Economics 2 (1996): 93–119; Folbre, The Invisible Heart: Economics and Family Values . 8. N. Folbre, “Children as Public Goods,” The American Economic Review 84 (1994): 86–90; P. England and N. Folbre, “The Cost of Caring,” The Annals of the American Academy of Political and Social Science 561 (January 1999): 39–51; J. C. Gornick and M. K. Meyers, Families That Work: Policies for Reconciling Parenthood and Employment (New York: Russell Sage Foundation, 2003). 9. J. C. Gornick, M. K. Meyers, and K. Ross, Public Policies and the Employment of Mothers: A Cross-National Study . Working Paper No. 140 (Syracuse, N.Y.: Maxwell School of Citizenship and Public Affairs, Syracuse University, 1996); A. Crittenden, The Price of Motherhood: Why the Most Important Job in the World Is Still the Least Valued (New York: Metropolitan Books, 2001); J. Waldfogel, “Understanding the {\textquoteleft}Family Gap'in Pay for Women with Children,” Journal of Economic Perspective 12 (1998): 137–56; M. J. Budig and P. England, “The Wage Penalty for Motherhood,” American Sociological Review 66 (2001): 204–25. 10. P. England, M. J. Budig, and N. Folbre, “Wages of Virtue: The Relative Pay of Care Work,” Social Problems 49 (2002): 455–73. 11. England, “Emerging Theories of Care Work.” 12. S. Hofferth, K. Shauman, and R. Henke, Characteristics of Children's Early Care and Education Programs: Data from the 1995 National Household Education Survey (Washington, D.C.: U.S. Department of Education, 1998); Sonenstein et al., Primary child care arrangements of employed parents: Findings from the 1999 National Survey of America's Families . 13. K. Snyder and S. Adelman, The Use of Relative Care While Parents Work: Findings from the 1999 National Survey of America's Families (Washington, D.C.: Urban Institute, 2004). 14. M. Warner, “The Threefold Importance of Child Care: The Need for a New Economic Framework,” 7. Background Paper for the May 2005 Cornell Workshop, Articulating the Economic Significance of the Childcare Sector. 15. Burton et al., Estimating the Size and Components of the U.S. Child Care Workforce and Caregiving Population . 16. Hofferth et al., Characteristics of Children's Early Care and Education Programs: Data from the 1995 National Household Education Survey . 17. L. Giannarelli and J. Barsimantov, Child Care Expenses of America's Families . Occasional Paper No. 40, Assessing the New Federalism Project (Washington, D.C.: The Urban Institute, 2000). 18. Hofferth et al., Characteristics of Children's Early Care and Education Programs: Data from the 1995 National Household Education Survey . 19. M. K. Meyers, D. Rosenbaum, C. Ruhm, and J. Waldfogel, “Inequality in Early Childhood Education and Care: What Do We Know?,” In K. Neckerman, ed., Social Inequality . (New York: Russell Sage Foundation, 2004). 20. L. Giannarelli, S. Adelman, and S. Schmidt, “Getting Help with Child Care Expenses,” Occasional Paper No. 62., Assessing the New Federalism Project (Washington, D.C.: The Urban Institute, 2003). 21. Center for the Child Care Workforce. Current Data on Child Care Salaries and Benefits in the United States . (Washington, D.C.: Center for the Child Care Workforce, 2002); M. Whitebook, C. Howes, and D. Phillips, Worthy Work, Unliveable Wages: The National Child Care Staffing Survey 1988–1997 (Washington, D.C.: Center for the Child Care Workforce, 1998). 22. D. Blau, The Child Care Problem: An Economic Analysis . (New York: Russell Sage Foundation, 2001). 23. S. Helburn, M. L. Culkin, C. Howes, D. Bryant, R. Clifford, D. Cryer, D. et al., eds., Cost, Quality, and Child Outcomes in Child Care Centers (Denver: Department of Economics, Center for Research in Economic and Social Policy, University of Colorado at Denver, 1995). 24. P. England, M. J. Budig, and N. Folbre, Wages of Virtue: The Relative Pay of Care Work . 25. Sonenstein, Gates, Schmidt, and Bolshun, Primary child care arrangements of employed parents: Findings from the 1999 National Survey of America's Families . 26. Giannarelli, Adelman, and Schmidt, Getting Help with Child Care Expenses . 27. Ibid. 28. L. Fender, C. T. O'Brien, T. Thompson, K. Snyder, and R. Bess, Recent Changes in New York Welfare and Work, Child Care, and Child Welfare Systems . (Washington D.C.: The Urban Institute, 2002). 29. For information on NYSIS sample, survey, and weighting procedures see http:// www.columbia.edu/cu/ssw/projects/surcent/. 30. New York City Administrative Code, Section 6-109. 31. Giannarelli, Adelman, and Schmidt, Getting Help with Child Care Expenses . ",

year = "2006",

month = mar,

doi = "10.1215/00161071-29-1-109",

language = "English (US)",

volume = "34",

pages = "109--128",

journal = "Politics and Society",

issn = "0032-3292",

publisher = "SAGE Publications Inc.",

number = "1",

}

TY - JOUR

T1 - Who pays? the visible and invisible costs of child care

AU - Meyers, Marcia K.

AU - Durfee, Alesha

N1 - Funding Information: Meyers Marcia K. School of Social Work and Evans School of Public Affairs, University of Washington, 4101 15th Ave NE, Seattle, WA 98105; 206-616-4409; mkm36@u.washington.edu Durfee Alesha Women and Gender Studies at Arizona State University 03 2006 34 1 109 128 Although the majority of young children now spend time in nonparental child care, we know relatively little about who provides this care and how its costs are distributed among parents, government, and other family members. In this article we use data from a survey of New York City families with children younger than six to estimate the contribution of parental expenditures, government assistance, and the market value of “donated” caregiving time by family, friends, and relatives. We conclude that uncompensated caregivers provide a substantial share of child care that is “invisible” in conventional economic measures. gender child care caregiving social policy subsidies sagemeta-type Journal Article search-text 10.1177/0032329205284759POLITICS & SOCIETYMEYERS and DURFEE WHO PAYS? THE VISIBLE AND INVISIBLE COSTS OF CHILD CARE MARCIA K. MEYERS ALESHA DURFEE Although the majority of young children now spend time in nonparental child care, we know relatively little about who provides this care and how its costs are distrib- uted among parents, government, and other family members. In this article we use data from a survey of New York City families with children younger than six to esti- mate the contribution of parental expenditures, government assistance, and the market value of "donated" caregiving time by family, friends, and relatives. We con- clude that uncompensated caregivers provide a substantial share of child care that is "invisible" in conventional economic measures. Keywords: gender; child care; caregiving; social policy; subsidies Families'use of nonparental child care has risen steadily for several decades in the United States, reflecting both increasing rates of maternal employment and changing norms about children's early education and school readiness. Although some parents report no use of regular nonparental care, often arranging employ- ment schedules to share caregiving between parents, about two-thirds of all chil- dren younger than age six and nearly three-quarters of those with employed moth- ers are now in some form of nonparental care on a regular basis.. As many as 2.5 Support for this research was provided by the Russell Sage Foundation and the Katherine T. Mac- Arthur Foundation. We are grateful for the superb assistance of Sandra Garcia in preparing data and measuresfortheanalysis.Allerrorsofomissionandcommissionaretheresponsibilityoftheauthors. Please address correspondence to Marcia K. Meyers, Associate Professor, School of Social Work and Evans School of Public Affairs, University of Washington, 4101 15th Ave NE, Seattle, WA 98105; 206-616-4409; mkm36@u.washington.edu. POLITICS & SOCIETY, Vol. 34 No. 1, March 2006 109-128 DOI: 10.1177/0032329205284759 © 2006 Sage Publications 109 million individuals are paid to provide this care and nearly as many may provide care on a regular basis without compensation.2 Given its growing social and economic importance, our knowledge about non- parental child care is surprisingly thin. There are limited data, and little consensus among analysts, about the sizeof the child care sector, the costs of producing child care services, or the distribution of these costs among parents, government, and familymembers and friends who "donate" their timeas uncompensated child care providers. Limitations in our knowledge reflect both theoretical challenges associated with defining and measuring the value of care work and empirical challenges relating to the lack of appropriate data and measures. With limited knowledge about who provides and who pays for child care, ithas been difficultto estimateits social and economic contributions, and nearly impossible to evaluate public and private contributions to its provision. In this article we address some of these gaps in our knowledge by estimating both the more- and less-visible costs of nonparental child care used by families with children younger than the age of six. Our analysis expands on prior work that has estimated the time children spend in nonparental care by monetizing the pub- lic and private costs of providing this care. Using detailed data collected as part of the New York Social Indicator Survey, in combination with city- and state-level administrative data, we estimate direct parental expenditures, the market value of services and subsidies provided by government, and alternative measures of the market value of uncompensated care by family, friends, and neighbors. In doing so, we provide new estimates of total direct expenditures by parents and govern- ment and argue that these underestimate the full economic value of nonparental child care. Unpaid family, friends, and relatives provide a significant share of all nonparental care that is largely invisible in conventional economic measures. If this labor is valued in terms of either market replacements for their services or compensation for their forgone wages, it would represent one-third to one-half of the total public and private contributions to the provision of nonparental child care. BACKGROUND Limitations in our knowledge about the costs of nonparental care are due, in large part, to the historic invisibility of caregiving work. As Nancy Folbre observes, Adam Smith's "invisible hand" of the market functions only as long as the "invisible heart" of caregiving relationships functions to produce and sustain each new generation of workers and citizens.3 Although caregiving work--what Folbre describes as "the production of people by means of people"--provides essential human resources for economic and social activities, it is largely invisible in conventional measures of productive economic activity. As long as this work remains invisible, many of its costs will remain invisible as well. 110 POLITICS & SOCIETY Measuring and Valuing Caregiving Work Questions about the public and private costs of providing nonparental child care relate to broader scholarly debates about the nature of care work and caring labor. In a recent review article, Paula England summarizes an emerging consen- sus among scholars about the characteristics of care work, including the shift in recent decades of an increasing amount of this work from the unpaid labor of women in the home to the paid labor of nonfamilial caregivers; the highly con- tested status of care work as a basis for state support; and the persistence of eco- nomic penalties, both for unpaid family caregivers and for those working in care work occupations.4 As she observes, reaching consensus about "the theoretical apparatus that explains the source of these empirical regularities" remains chal- lenging.5 The conceptual challenge of estimating the cost of providing care begins with the definition and measurement of this labor. Historically, most caregiving for children, elderly, disabled, and other dependent individuals was provided through the uncompensated labor of family and community members. This labor has been treated in economic theory as a form of nonmarket household production, as a welfare service, or, among labor economists, as a component of the "leisure" time that adults trade for waged work in the market. As a result, much of the adult labor devoted to caregiving has been excluded from conventional measures of eco- nomic activity. Nowhere is this more evident than in the official measure of eco- nomic productivity for the country--the Gross National Product (GNP)--which explicitly excludes unrenumerated household production. Given its exclusion from conventional measures of economic activity, it has been difficult to measure adults' caregiving labor. Researchers who have exam- ined this labor using time diaries conclude that the time devoted to unpaid domes- tic and caring work in the home is substantial--approximately equal to the time devoted to paid work in the market.6 Counting hours of labor devoted to unpaid carework still fails to capture its economic and social value, however. Some ana- lysts have estimated the value of uncompensated home production in terms of market replacements for these services--that is, as the cost of hiring a third party to perform them. These estimates suggest that the market value of this labor may equal as much as one-half or more of the GNP in severalindustrializedcountries.7 Calculating the market cost of replacing household labor may still underesti- mate its full value if this labor is undervalued in the market. Feminist scholars have argued for many years that women's caregiving labor provides social and economic benefits that are uncompensated when provided in the home and under- valued when provided through market exchanges. This undervaluation can be understood as a pervasive devaluation of women and labor that is associated with women's traditional roles as mothers and nurturers. It can also be understood in terms of a market failure related to the production of public goods. The market failure argument is particularly relevant to child care labor. In standard economic MEYERS and DURFEE 111 terms, children are public goods in the sense that the benefits of their future capa- bilities--as productive workers and citizens--are widely shared, but the costs of producing these capabilities are impossible to allocate to individual beneficia- ries.8 Economic theory predicts that markets will poorly compensate those who provide such services because it is impossible to prevent others from "free-rid- ing" by enjoying the benefits of healthy children without paying the costs of producing them. There is substantial empirical evidence that the gendered distribution of care- giving work, and its production of widely dispersed and nonexcludable benefits, do impose costs on caregivers that are not compensated through the market. Women who interrupt their employment to bear and raise children incur a sub- stantial motherhood or caregiving penalty in their wages, relative to both men and to women who do not have children.9 England has extended this line of research to document a wage penalty of 5 to 10 percent for individuals employed in an occupation involving caregiving, controlling for human capital, job skill, and occupational characteristics.10 The reasons for these penalties are multiple and interacting, reflecting women's career interruptions for caregiving, gendered occupational structures, institutionalized norms about caregiving work, and the selection of individuals, overwhelmingly women, into occupations that may pro- vide intrinsic rewards to compensate for the lack of economic remuneration.11 The result of these penalties is the imposition of private and largely invisible costs on caregivers in the form of forgone wages and earnings that are lower than what they might command, given their skills and experience, in other occupations. Prior Research Defining, measuring, and valuing caregiving work raises substantial concep- tual challenges. It also raises formidable empirical challenges that have limited our knowledge about the size of the child care sector and workforce, public and private expenditures for child care services, and the private costs that child care providers incur through their donation of uncompensated labor and through care penalties in their earnings. The first difficulty arises in the measurement of child care services and the size of the workforce. Because it is not provided through regular market exchanges, a substantial share of nonparental child care is invisible in conventional economic measures. Some nonparental care is provided as a market good in settings that are regulated by state and local government, including private child care centers, preschools, nursery schools, and licensed family child care homes. Other care is provided outside of the market through public education and welfare programs. Stillother, less formal forms of care are provided by home-based providers and by uncompensated family, friends, and neighbors. Data collected at the household level suggest that families use these more- and less-visible forms of child care almost equally. In the approximately two-thirds of 112 POLITICS & SOCIETY families with children younger than age six who use any care, children are about equally likely to be in child care centers or in relative care as their primary arrangement.12 Considering any use of relative care--as a primary arrangement or in combination with other forms of care--Snyder et al. estimate that more than a third of all children younger than age five are cared for by relatives on a regular basis while their parents are at work.13 Conventional economic measures are poorly designed to measure this diver- sity of arrangements. Warner describes recent efforts to study the child care sector as akin to observing an iceberg in which data on the "the formal licensed or regu- lated care that we see above the water line" capture only a small portion of the much larger, less-visible whole.14 The most visible market-based services are measured in Census data on establishments and workers covered by unemploy- ment insurance. When analysts compare these estimates to data from state regula- tory agencies, however, they typically find between 20 and 300 percent more employees at work in regulated child care settings. These estimates of the workforce more than double again if unregulated home-based providers are included through the use of self-employment tax information provided by the Internal Revenue Service. However, these expanded estimates still fail to capture the unknown number of individuals who are paid to provide care within the infor- mal economy, but do not report this self-employment income for tax purposes, or the family, friends, and neighbors who care for children on a regular basis without compensation. In a recent effort to address the limitations of these supply-side approaches, Burton et al. employ a demand-side approach to estimate the size of the child care workforce using household-level surveys of child care use.15 They estimate that the paid workforce caring for children younger than age six in the United States includes over 2.3 million individuals at a point in time--approximately 600,000 more than are identified in the Current Population Survey and more than twice the number tracked by the Bureau of Labor Statistics. They conclude that as a group, relative and nonrelative caregivers (other than family child care providers) make up nearly half of the paid child care workforce. Even more strikingly, they estimate that an additional 2.4 million individuals, 93 percent of whom are rela- tives, provide care for children younger than six during a given week without compensation. Given the invisibility of much nonparental child care, it is not surprising that knowledge about its costs is also limited and incomplete. Parents pay a portion of these costs through market exchanges; government pays another portion through social welfare and educational spending. Although it is possible to measure some of these expenditures using household-level survey data and aggregate data on government spending, it has been nearly impossible to combine these data to esti- mate the relative contributions of parents and government. If child care is measured as a market good, at least a portion of its costs can be estimated using data on parents'direct expenditures to purchase care or pay indi- MEYERS and DURFEE 113 vidual providers. Nearly all parents who use child care centers, family child care, or in-home caregivers pay for those services. An estimated 33 percent of those who use relative caregivers also pay for that care.16 How much do families pay for care? Giannarelli and Barsimantov, using the National Survey of America's Families (NSAF), estimate that about one-half of all families with employed mothers and children younger than age thirteen incur some child care expenses; among these families, expenditures average $286 per month, or about 9 percent of family earnings.17 This estimate is limited in several respects. It does not capture tax benefits that reduce parents'net expenses. It also fails to reflect expenditures by familiesin which mothers are not employed. Using the National Household Education Survey, Hofferth et al. estimate that although they are less likely to use nonparental care, more than a third of families in which mothers are looking for work or not in the labor force report some hours of nonparental care for preschool children.18 These families actually pay more per hour for this care than families in which mothers are employed, which may reflect their use of more expensive preschool programs or the higher costs of part-time care. These out-of-pocket parental expenditures capture only a portion of the com- pensated costs of nonparental child care. Federal, state, and local governments provide additional child care directly through public child care and early educa- tion services, for example, the federal Head Start and state pre-Kindergarten pro- grams. They provide other support through means-tested subsidies for parents' purchase of private services, primarily through the Temporary Assistance for Needy Families (TANF), Child Care and Development Fund (CCDF), and Social Services (SSBG) block grants. Less directly, governments make tax expenditures through policies that allow parents to deduct a portion of their private child care expenses from income taxes through the federal Child and Dependent Care Tax Credit and state-level tax credits in several states. Other government programs reduce the operating expenses of child care facilities through, for example, subsi- dized child care food programs and targeted tax credits. Data on these public expenditures are surprisingly limited and nearly impossi- ble to analyze at the household or facilitylevel. Child care programs are funded by several federal, state, and local agencies housed in various social service, welfare, tax, and early education agencies and no public agency tracks all of these outlays. Meyers et al. estimate that federal and state spending for preschool services and means-tested subsidies, and federal tax expenditures for individual child care tax benefits, exceeded $20 billion as of 2000.19 This estimate fails to capture signifi- cant public expenditures, however, including subsidized food programs, state supplementation of federal Head Start services, and federal and state tax benefits for providers. Aggregate government spending data tell us something about what govern- ment contributes to the production of child care. They tell us little about the share of total costs that are paid by government, because they are difficult to link to 114 POLITICS & SOCIETY either household-level child care consumption or to establishment- and provider- level production of these services. In one of the few recent estimates of house- hold-level receipt of government assistance, Giannarelli and her colleagues use the National Survey of America's Families (NSAF) to estimate that about 12 per- cent of employed families with a child younger than age 13 receive some type of nontax assistance from government or other organizations.20 Because parents are unlikely to know the value of government subsidies and programs, however, the NSAF does not collect data on the amount of this assistance that can be compared to parental expenditures. Parental and government expenditures are even more limited as a measure of the costs of nonparental child care by their exclusion of the "donation" of labor by the unpaid family, friends, and neighbors who make up an estimated one-half of the total workforce caring for children. We know something about the number of children who are cared for in these arrangements, but little or nothing about the amount of labor that is supplied by uncompensated caregivers or its value in market terms. Estimating the cost of replacing the donated labor of uncompensated family, friends, and neighbors with paid child care would provide a more complete accounting of the public and private costs of child care. It would still underesti- mate the full value of this care, and its costs from the providers'perspective, if the wages paid to child care providers are lower than the wages they could earn in alternative employment--that is, if they incur a "care penalty" for their employ- ment in a caregiving occupation. The exceptionally low wages of the child care workforce suggest that these workers do pay some penalty for their occupational location. Child care workers are among the lowest paid workers in the U.S. labor market. Considering only the share of the child care workforce captured in BLS surveys ("child care workers" and "preschool teachers" in institutional settings), analysts at the Center for the Child Care Workforce estimate that only eighteen occupations have lower mean wages than child care workers.21 The earnings of licensed family child care pro- viders and individual providers are even lower. The wages of child care workers are particularly low in comparison to those of other workers with similar educa- tional backgrounds and, despite the substantial increase in demand for child care services in the 1980s and 1990s, were an estimated one dollar per hour lower in 1997 in real dollars than they were ten years earlier.22 The possibility that child care workers incur a "care penalty" in their wages suggests that these same individuals might command a higher wage in other occu- pations. If so, our estimates of the cost of child care based on current caregiver wages will underestimate the full costs and value of these services. In one effort to discern how this "care penalty" may bias estimates of the cost of producing child care, Helburn and her colleagues estimate an adjustment to the wages of workers in child care centers to reflect their predicted wages in other occupations, given their age, gender, education, and other characteristics.23 By their estimate, nearly MEYERS and DURFEE 115 20 percent of the full cost of providing these services is contributed by workers through "forgone wages"--that is, the difference between actual wages and the wages that they could command in other occupations. England et al. estimate that among all care occupations, child care workers incur the largest wage penalty-- 41 percent--relative to employment in occupations not involving care work after adjusting for human capital, occupational skills demands, and other occupation and industry characteristics.24 RESEARCH QUESTIONS AND APPROACH Although a majority of families use nonparental child care on a regular basis, much of the labor involved in providing this care is invisible in standard economic measures, its full social value may be underestimated, and the distribution of its costs among parents, government, and uncompensated caregivers is largely unknown. In this article we address some of these gaps in our knowledge by using unusually detailed family-level survey data to consider the contributions of par- ents, government, and uncompensated caregivers to the provision of nonparental child care. We go beyond existing studies of the use of child care arrangements, or of time devoted to caring for children, to estimate the visible and less visible costs associated with providing this care. Combining survey data with administrative data from a variety of sources, we estimate (1) the share of families that incur out-of-pocket expenses for non- parental care, receive any form of government assistance, and use unpaid care by family, friends, and neighbors on a regular basis; (2) average family expenditures, the value of government assistance, and the market value of the donated labor of unpaid caregivers under different assumptions about their compensation; and (3) the resulting share of the total economic value of nonparental care that is contrib- uted by parents, government, and uncompensated family, friend, and neighbor caregivers. CONTEXT: CHILD CARE IN NEW YORK CITY Conducting these analyses with data from a single city has the advantage of allowing us to use administrative data for detailed imputations of the value of assistance. The primary disadvantage is that the data represent only one popula- tion and policy environment, potentially limiting generalizability to the United States as a whole. New York City provides a case study of public and private child care costs in the context of relatively generous government assistance and relatively con- strained family resources. New York Stateis one of only six statesthathave autho- rized pre-Kindergarten services for all four-year-olds, although the delivery of these services has lagged behind the promise. It is one of twenty-six states that provide a child care tax credit and one of eleven in which this credit is refundable. 116 POLITICS & SOCIETY Although spending for means-tested child care assistance is also high in absolute terms--New York City provided an estimated $230 million in means-tested sub- sidy assistance in fiscal year 19992000--the city is similar to other jurisdictions in being unable to serve all eligible families and rationing assistance. In terms of comparability to the United States as a whole, analyses by the Urban Institute indicate families in New York State are about as likely as those nationwide to have a child younger than five in nonparental care and, on average, use similar care arrangements.25 Relative to other families, however, families with total incomes under 200 percent of poverty in New York are less likely to be using center-based care and more likely to use relative care as a primary arrangement. This may reflect the "bricks and mortar" difficultiesof establishing new child care facilities given the City's density and real estate prices. The proportion of families with an employed mother and a child younger than thirteen who receive some form of nontax child care assistance from government or other organizations is higher in New York (16 percent) than nationwide (13 percent), which is consistent with the availability of both subsidy and preschool assistance.26 Family resources may also be more constrained in New York City than in the country as a whole. Child care costs are higher in New York than they are for the remainder of the United States; in 1999, families in New York State who paid for child care spent an average of $348 per month, more than the average of $303 spent by families nationwide.27 The proportion of children living in poverty in New York State is also higher than the rest of the country. Nearly one-quarter (22 percent) of New York children lived in poor families in 1998, compared to 18 percent of children nationwide.28 DATA AND MEASURES Data and Sample Data for this analysis are taken from waves II and III (1999 and 2002) of the New York Social Indicators Survey (NYSIS) and from a variety of city and state administrative sources. The NYSIS is a telephone survey administered to a ran- dom sample of New York City households by the Columbia University Social Indicators Research Center. The approximately thirty-minute survey covers a number of aspects of individual and family well-being, including adequacy of and satisfaction with institutional supports, family economics, living conditions, edu- cation, health, and employment. Each wave of the NYSIS collectsdatafrom a rep- resentative sample of approximately 1,500 randomly selected families in the City using computer-assisted telephone interview (CATI) technology and random- digit dialing (RDD). For our analyses, we use data on 646 families with at least one child between the ages of zero and five. "Family," as defined by the NYSIS, includes a respondent, his or her resident spouse or partner, and any resident chil- dren between the ages of zero and seventeen. All estimatesare reported atthe fam- MEYERS and DURFEE 117 ily level and weighted, using Census data, to represent families with children in New York City.29 Measures Waves II and III of the NYSIS included a special module that collected unusu- ally detailed questions about all nonparental child care arrangements, child care expenditures, and receipt of governmental assistance (including tax credits, pub- lic preschool, and subsidies) in the prior twelve months. We make use of these data to estimate the amount of various types of nonparental child care used in the prior year along with the receipt of various forms of financial and nonfinancial assistance for this care. Because parents are unlikely to know the value of many forms of government assistance, we use administrative data for New York City or State, as appropriate, to impute the value of various benefits. To estimate the value of unpaid care provided by family, friends and neighbors, we estimate the total hours used and impute both the replacement cost (of purchasing equivalent mar- ket services) and the value of the caregivers' labor using alternative assumptions about caregiver compensation. Child Care Arrangements Nonparental care is defined as regular care for children by someone other than parents. Information was collected from parents with any children younger than thirteen on each type of nonparental care used for any child(ren) in the family and on the number of child(ren) in that form of care. Respondents were asked how many months the child(ren) were in each type of care and whether it was on a full- or part-time basis. The timethatindividuals devoted to providing care--for exam- ple, as uncompensated child care providers--was defined as the time that they cared for one or more children in the family. Parental Out-of-Pocket Expenditures Parental expenditures are estimated directly from survey responses to detailed questions about out-of-pocket payment for all forms of care, for all children, in the prior twelve months. We adjust these expenditures by subtracting the amount reimbursed through Federal or statechild care tax credits. Item-missing values are imputed using family demographics, including family structure, income level, education level, race, ethnicity, and immigration status. Governmental Assistance The receipt of specific forms of government assistance is estimated directly from survey questions about the use of any public preschool or other child care programs, the receipt of any subsidies (by parents or providers), and the use of federal and state child care tax credits in the previous twelve months. Questions about government assistance were asked for each type of care arrangement, a 118 POLITICS & SOCIETY more complete accounting of this assistance than has been available in other sur- veys. The value of government assistance is calculated by imputing and summing the value of tax credits, subsidies and/or vouchers, and Head Start/public pre- school for families reporting the receipt of governmental assistance. The value of tax credits is estimated using the TAXSIM model provided by the National Bureau of Economic Research. The TAXSIM program applies federal and state tax rules to estimate the value of both refundable and nonrefundable child care tax benefits using family-level data on income, family composition, child care expenditures, and other family-level characteristics. The value of child care subsidies is estimated by first calculating the annual hours of subsidized care received by all children in the family in the prior twelve months and then imputing the maximum subsidy reimbursement availablein New York City on the basis of the most recent New York Market Rate Survey. The sur- vey is conducted by New York State to estimate the market costs of various forms of care in each region of the state, and is used to set reimbursement rates for assis- tance through the TANF, CCDF, and other subsidy programs. These imputations adjust for the regional price of care by type of setting, the age(s) and number of child(ren), whether that care was provided on a full- or part-time basis, and the number of months each child was in the subsidized setting. To estimate the value of Head Start and public preschool services, we use survey data to determine how many children were enrolled in either type of program during the prior twelve months, the number of months enrolled, and whether for a full- or part-day pro- gram. We impute the value using governmental cost methods--that is, distribut- ing total public expenditures (for direct services) among all recipient children to determine costs on a per-child basis. Value of Care Donated by Family, Friends, and Neighbors A substantial share of the families in our sample report using "free" care pro- vided by family, friends, and neighbors on a regular basis for one or more of their children. To estimate the cost of this care we employ two approaches: estimating the replacement cost of purchasing equivalent market services and estimating the opportunity costs or forgone wages of caregivers in terms of alternative alloca- tions of their under different assumptions about hourly wages. We first estimate the total number of hours of uncompensated care that fami- lies received in the prior year, using survey questions about the number of chil- dren cared for and the length of the care arrangement (average weekly hours and the number of weeks used) in the prior twelve months. Our first approach to esti- mating the market value of this care assumes that the replacement cost would be similar to the cost of using family child care. As with child care subsidies, we use the New York Market Rate Survey to impute the value of family child care, adjust- ing for the amount of care provided. We use the rates for preschoolers (threefive years old) because the NYSIS does not contain specific information as to which child was cared for in these arrangements. Based on the number of hours reported, MEYERS and DURFEE 119 we apply the rate for either full-time ($103 per week) or part-time ($67 per week) care for each child in care. Our second approach assumes that uncompensated caregivers face opportu- nity costs when allocating their time to unpaid care work rather than market work. The NYSIS does not provide sufficiently detailed data on caregivers to estimate the hourly wage that they would be likely to command in other occupations. Instead, we first impute their potential earnings at the New York State minimum wage of $5.15 per hour. We assume that this is the minimum that they would earn were they to allocate their time to alternative employment. By using the minimum wage we assume that, if paid for providing care, indi- viduals would incur the "care penalty" associated with working in child care. To estimate the private costs associated with this penalty we also impute a higher hourly wage for their time. In the absence of data with which to predict an individ- ual wage, we apply the "living-wage" level that New York City has recently adopted as a minimum for all city employees and contractors.30 When enacted in 2002, the living-wage rate was $8.10 per hour for employees with health benefits, with an additional allowance of $1.50 per hour for employees who do not have health benefits. Wage-based imputations are adjusted for the number of weeks that parents reported using the care and set the providers'hours at either full-time (forty hours per week) or part-time (twenty hours per week). Unlike imputations based on the cost of purchasing family child care, the wage-based estimates do not adjust for the number of children in care. We calculate the share of average costs that are paid by parents and govern- ment and the estimated value of donated caregiving time as the share of the weighted average, at the family level, contributed by each. RESULTS Sample Characteristics In the NYSIS sample, weighted to represent the population of families with any children younger than six in New York City, about one-half of families have children younger than three (Table 1). Nearly 40 percent of families are headed by Hispanic parents and about one-quarter are headed by both white non-Hispanic and African American parents. Just over one-quarter of the families have income at or below the federal poverty line. In about one-half of allfamilies,mothers were currently employed, in school, or in job preparation activities. Sample character- istics are similar for those families who did and did not use some form of nonparental care, with the exception of higher rates of maternal employment (59 versus 48 percent) among those reporting some child care use. 120 POLITICS & SOCIETY Use and Payment for Nonparental Child Care When we consider all forms of care used on a regular basis for children youn- ger than six, the large majority of New York City families with young children use some form of nonparental care during the year (Table 2). Among all families with young children, more than 80 percent used some form of care. About one-third of all families, and nearly one-half of those using any care, use formal care in a child care center, family child care home, or early education or preschool programs. Substantially more families--63 percent of all and 84 percent of child care users--use care that is provided by family, friends, or neighbors. The majority of parents who use child care pay something for this care out-of- pocket (59 percent); 20 percent pay the full amount. Fewer of these families receive some form of assistance from government (55 percent). Federal and state tax credits are the most common form of assistance (35 percent), followed by enrollment of children in public Head Start or pre-K programs (24 percent). Con- siderably fewer familieswho use care receivegovernment subsidies (12 percent). As described above, the large majority of all families who use any nonparental child care use some care provided by family, friends, and neighbors. Although parents pay for a portion of this care, and a small share is subsidized by govern- ment, the majority is donated without charge to either parents or government. MEYERS and DURFEE 121 Table 1 Sample Characteristics (Means and Standard Deviations, Weighted), New York City Families with any Child Aged Zero to Five All Families (N = 646) Families Using Nonparental Child Care (N = 526) Standard Standard Mean Deviation Mean Deviation Any infants in family (age 0 to 2) 0.55 0.50 0.54 0.50 Youngest child older than 3 0.45 0.50 0.46 0.50 Number of children age 0 to 5 in family 1.10 0.37 1.10 0.37 Respondent is Hispanic 0.38 0.49 0.35 0.48 Respondent is black, non-Hispanic 0.26 0.44 0.31 0.46 Respondent is white, non-Hispanic 0.22 0.41 Respondent's education < high school degree 0.21 0.40 0.14 0.34 Respondent's education = high school degree 0.35 0.48 0.37 0.48 Respondent's education = some college or more 0.43 0.50 0.49 0.50 Family earnings less than 100% Federal Poverty Level 0.34 0.47 0.37 0.48 Family earnings 101200% Federal Poverty Level 0.23 0.42 0.21 0.41 Family earnings 201400% Federal Poverty Level 0.28 0.42 0.25 0.44 Family earnings more than 400% Federal Poverty Level 0.20 0.40 0.16 0.37 Mother is employed, in school, or in job training 0.49 0.50 0.58 0.49 Respondent is in Wave 2 0.51 0.50 0.49 0.50 Source: New York City Social Indicators Survey, Waves II and III. Note: Means calculated using weighted data. One-third of all families, and nearly one-half of child care users, use some care that is provided without charge by family, friends, or neighbors. Value and Distribution of Child Care Services Detailed data on the hours and types of care used for each child in the family in the prior year allow us to estimate the total economic value of care under various assumptions (Table 3). If we estimate the value of nonparental child care on the basis of parental expenditures alone, New York Cityfamiliespay a littlemore than $2,000 on average per year. This is lower than the annualized estimate that Giannarelli et al. report for New York State,31 which may reflect the restriction of their sample to families with employed mothers (who may use and pay for more care) and their inclusion of care for children older than six (who may be in care for fewer hours per day). When we include our estimates of the value of government assistance, the average value of child care per family is closer to $4,000 per year and expendi- tures for this care are split nearly evenly between parents and government. Our estimates of the share of all child care using families who receive assistance are higher than those reported in prior studies, which is likely due to our ability to do a more complete accounting for public child care programs, subsidies, and tax ben- efits. Our estimate of the share of direct expenditures paid by government pro- 122 POLITICS & SOCIETY Table 2 Frequencies of Child Care Usage, Costs, and Receipt of Assistance (Means and Standard Deviations, Weighted), New York City Families with any Child Aged Zero to Five All Families (N = 646) Families Using Nonparental Child Care (N = 526) Standard Standard Mean Deviation Mean Deviation Uses formal child care 0.35 0.48 0.47 0.50 Uses Head Start or public pre-K 0.18 0.38 0.24 0.43 Uses care provided by family, friends, or neighbors 0.63 0.48 0.84 0.36 Uses care donated by family, friends, or neighbors 0.36 0.48 0.49 0.50 Pays some amount out-of-pocket for nonparental child care 0.44 0.50 0.59 0.49 Pays all costs out-of-pocket for nonparental child care 0.15 0.36 0.20 0.40 Receives child care subsidy 0.09 0.28 0.12 0.32 Receives child care tax credit 0.25 0.42 0.35 0.46 Receives any governmental assistance with nonparental child care 0.41 0.49 0.55 0.50 Receives governmental assistance with formal child care (other than Head Start or public pre-K) 0.07 0.26 0.10 0.30 123 Table3 EstimatedValuesofParental,Government,and"Donated"ContributionstoChildCare(CalculatedattheFamilyLevel,Weighted),NewYorkCityFamilieswithanyChild AgedZerotoFiveUsingNonparentalChildCare(N=526) TotalValue TotalValueofChildCare TotalValueofChildCareDonatedby ofChildCareProvidedFamily,Friends,ValueofEstimated PaidOut-ThroughandNeighbors,ParentGovernmentCostof TotalValueof-PocketGovernmentIfValuedatExpendituresAssistanceDonatedCare ofChildCarebyParentsAssistanceMarketRatesasShareofasShareofasShareof (PerFamily)(PerFamily)(PerFamily)(PerFamily)WeightedTotalWeightedTotalWeightedTotal MeanMeanMeanMeanMeanMean Parentalout-of-pocketexpendituresonly$2,041$2,041$1,949------------------------ Parentalexpendituresplusvalueof governmentassistance$3,990$2,041$1,949------51percent49percent------ Parentalexpenditures,governmentassistance, plusdonatedcare... valuedatthecostoffamilychildcare$5,788$2,041$1,949$1,78335percent34percent31percent ...valuedatthecostofpayingtheprovider atminimumwage$6,792$2,041$1,949$2,77030percent29percent41percent ...valuedatthecostofpayingtheprovider theNewYorkCity "LivingWage"$9,684$2,041$1,949$5,66221percent20percent59percent Source:NewYorkCitySocialIndicatorsSurvey,WavesIIandIII. vides a high estimate of the distribution of these costs between parents and gov- ernment, given the relatively extensive provision of public preschool in New York City and the availability of child care tax credits in the state. When we add our estimates of the economic value of the "invisible" care donated by family, friends, and neighbors, we reach substantially different con- clusions about public and private contributions to child care. If the child care cur- rently donated by family, friends, and neighbors were replaced with family child care purchased at the average market rate, the total economic value of the child care used would increase by about one-third per family. This estimate reflects the very low wages earned by child care workers, however, and family child care pro- viders in particular. If the family, friends, and neighbors who provide care to New York City families earned even the New York State minimum wage of $5.15 per hour for the labor they currently provide for free, the total value of nonparental care would be more than $6,700 per family per year, with more than 40 percent of this total provided by these caregivers. If these providers were paid what New York City deems a living wage for city employees and contractors, the total value of child care would be nearly 2.5 times higher than the expenses currently covered by parental and government expenditures. Government assistance accounts for only 20 percent of this total. Nearly 80 percent of the economic value of non- parental child care is provided privately, by parents (21 percent) and by the esti- mated wages of family, friends, and neighbors who provide care (59 percent). DISCUSSION Much of the adult time and labor that are devoted to the care and nurturing of young children is rendered invisible by conventional measures of economic activ- ity and output. Most of this labor is provided by parents and its value is entirely uncounted in measures, such as the GDP, that define economic productivity in terms of market exchanges. To an increasing extent, families are substituting nonparental child care for a portion of their own caregiving labor. Like parental caregiving, much of the labor provided by these nonparental caregivers remains invisible because they are not compensated through conventional market ex- changes and because, when paid, they incur a caregiving penalty in their wages. As a result, although some of the costs of nonparental child care are captured in data on parental and government expenditures, these data substantially underesti- mate the total labor devoted to child care, its economic value, and, arguably, its social value. We estimate that about one-third of New York City families with young chil- dren use some form of "formal" care (in public or private centers, preschools, and family child care homes) during the year and nearly two-thirds use care provided by family, friends, or neighbors on a regular basis. Some of this care by relatives and friends is compensated by parents and/or through government subsidies. But one-third of all families with young children, and about one-half of those using 124 POLITICS & SOCIETY child care, receive some regular care that is "donated" by these family, friends, and neighbors without direct charge. Assessing the market value of this invisible "donated" labor substantially alters our estimates of the total economic value of nonparental child care services, as well as the share of the value provided by public and private parties. When we consider only parental out-of-pocket payments and government expenditures for services, subsidies, and tax credits, the value of child care is almost equally dis- tributed between parents and government. If the family, friends, and neighbors who care for children without direct charge were compensated at the rate of fam- ily child care providers, however, we estimate that they would provide about one- third of the total economic value of care. Because family child care workers are among the lowest paid workers in the U.S. workforce, valuing the donated labor of family, friend, and neighbor caregivers at market rates for family child care may underestimate both the market value of their labor, if allocated to other employment, and the social value of their work. If these caregivers could earn even the minimum wage for their hours of work in other employment, they are "donating" more than 40 percent of the total value of child care. At this wage, many of these individuals would still earn poverty-level incomes, even with full- time employment. New York City has established a "living wage" standard for city employees and contractors that is $19,968 for full-time, full-year employ- ment. If currently uncompensated caregivers were able to earn this wage for their "donated" hours of caregiving, their donated labor would account for nearly 60 percent of the total value of care. Under this assumption, only about 20 percent of the value of all child care is contributed by government. Our analysis of the visible and invisible "costs" of child care is unusual in that it captures the full range of contributions from parents, government, and unpaid caregivers. It is still limited in some important respects. Although we have detailed data on family-level receipt of government assistance, our estimates do not include other forms of assistance--including food subsidies and tax benefits--that are received directly by some child care centers and family child care providers. We adjust our estimates of the market value of donated caregiving time in order to demonstrate the extent to which market prices may undervalue paid caregiving work. We do not adjust for the possible undervaluing of the child care labor that is purchased through parental payments and government assistance. Doing so would raise both the total estimated value of child care and the relative contribu- tions of parents and government. These estimates are also limited in using data from a single city that may not be representative of other parts of the country. In particular, given the relatively gen- erous preschool, tax credit, and subsidy assistance available in New York City, the share of the value of child care provided through government assistance may be higher in our sample than they would be in other regions. Parents in New York also pay more for child care, on average, than parents in other regions, rendering our estimates of relative share more representative. MEYERS and DURFEE 125 Although limited in some respects, this analysis provides one of the first esti- mates of the visible and invisible "costs" of nonparental child care. It suggests if we fail to recognize and value the "invisible" labor of uncompensated caregivers, we will seriously underestimate the total economic value of nonparental child care and the private contributions--primarily by family members--to its provi- sion. In doing so, we further devalue this work and reify the conceptualization of it as private and costless. NOTES 1. F. L. Sonenstein, G. Gates, S. R. Schmidt, and N. Bolshun, Primary child care arrangements of employed parents: Findings from the 1999 National Survey of America's Families (Washington, D.C.: The Urban Institute, 2002). 2. A. Burton, M. Whitebook, M. Young, D. Bellm, C. Wayne, R. Brandon, and E. Maher, Estimating the Size and Components of the U.S. Child Care Workforce and Care- giving Population (Washington, D.C.: Center for the Child Care Workforce, 2002). 3. N. Folbre, The Invisible Heart: Economics and Family Values (New York: The New Press, 2001). 4. P. England, "Emerging Theories of Care Work," Annual Review of Sociology 31 (2005): 38199. 5. Ibid., 395. 6. L. Goldschmidt-Clermont and E. Pagnossin-Aligisakis, Measures of Unrecorded Economic Activities in Fourteen Countries (New York City: United Nations Development Programme, Human Development Report Office, 2001); J. Gershuny and J. P. Robinson, "HistoricalChangesintheHouseholdDivisionofLabor,"Demography25(1998):53752. 7. K. Cloud and N. Garrett, "A Modest Proposal for Inclusion of Women's Household Human Capital Production Analysis of Structural Transformation," Feminist Economics 2 (1996): 93119; Folbre, The Invisible Heart: Economics and Family Values. 8. N. Folbre, "Children as Public Goods," The American Economic Review 84 (1994): 8690; P. England and N. Folbre, "The Cost of Caring," The Annals of the American Acad- emy of Political and Social Science 561 (January 1999): 3951; J. C. Gornick and M. K. Meyers, Families That Work: Policies for Reconciling Parenthood and Employment (New York: Russell Sage Foundation, 2003). 9. J. C. Gornick, M. K. Meyers, and K. Ross, Public Policies and the Employment of Mothers: A Cross-National Study. Working Paper No. 140 (Syracuse, N.Y.: Maxwell School of Citizenship and Public Affairs, Syracuse University, 1996); A. Crittenden, The Price of Motherhood: Why the Most Important Job in the World Is Still the Least Valued (New York: Metropolitan Books, 2001); J. Waldfogel, "Understanding the 'Family Gap'in Pay for Women with Children," Journal of Economic Perspective 12 (1998): 13756; M. J. Budig and P. England, "The Wage Penalty for Motherhood," American Sociological Re- view 66 (2001): 20425. 10. P. England, M. J. Budig, and N. Folbre, "Wages of Virtue: The Relative Pay of Care Work," Social Problems 49 (2002): 45573. 11. England, "Emerging Theories of Care Work." 12. S. Hofferth, K. Shauman, and R. Henke, Characteristics of Children's Early Care and Education Programs: Data from the 1995 National Household Education Survey (Washington, D.C.: U.S. Department of Education, 1998); Sonenstein et al., Primary child care arrangements of employed parents: Findings from the 1999 National Survey of Amer- ica's Families. 126 POLITICS & SOCIETY 13. K. Snyder and S. Adelman, The Use of Relative Care While Parents Work: Findings from the 1999 National Survey of America's Families (Washington, D.C.: Urban Institute, 2004). 14. M. Warner, "The Threefold Importance of Child Care: The Need for a New Eco- nomic Framework," 7. Background Paper for the May 2005 Cornell Workshop, Articulat- ing the Economic Significance of the Childcare Sector. 15. Burton et al., Estimating the Size and Components of the U.S. Child Care Workforce and Caregiving Population. 16. Hofferth et al., Characteristics of Children's Early Care and Education Programs: Data from the 1995 National Household Education Survey. 17. L. Giannarelli and J. Barsimantov, Child Care Expenses of America's Families. Occasional Paper No. 40, Assessing the New Federalism Project (Washington, D.C.: The Urban Institute, 2000). 18. Hofferth et al., Characteristics of Children's Early Care and Education Programs: Data from the 1995 National Household Education Survey. 19. M. K. Meyers, D. Rosenbaum, C. Ruhm, and J. Waldfogel, "Inequality in Early Childhood Education and Care: What Do We Know?," In K. Neckerman, ed., Social Inequality. (New York: Russell Sage Foundation, 2004). 20. L. Giannarelli, S. Adelman, and S. Schmidt, "Getting Help with Child Care Expenses," Occasional Paper No.62.,Assessingthe New Federalism Project (Washington, D.C.: The Urban Institute, 2003). 21.Center forthe Child Care Workforce. Current DataonChild Care Salaries andBen- efits in the United States. (Washington, D.C.: Center for the Child Care Workforce, 2002); M. Whitebook, C. Howes, and D. Phillips, Worthy Work, Unliveable Wages: The National Child Care Staffing Survey 19881997 (Washington, D.C.: Center for the Child Care Workforce, 1998). 22. D. Blau, The Child Care Problem: An Economic Analysis. (New York: Russell Sage Foundation, 2001). 23. S. Helburn, M. L. Culkin, C. Howes, D. Bryant, R. Clifford, D. Cryer, D. et al., eds., Cost, Quality, and Child Outcomes in Child Care Centers (Denver: Department of Eco- nomics, Center for Research in Economic and Social Policy, University of Colorado at Denver, 1995). 24. P. England, M. J. Budig, and N. Folbre, Wages of Virtue: The Relative Pay of Care Work. 25. Sonenstein, Gates, Schmidt, and Bolshun, Primary child care arrangements of employed parents: Findings from the 1999 National Survey of America's Families. 26. Giannarelli, Adelman, and Schmidt, Getting Help with Child Care Expenses. 27. Ibid. 28. L. Fender, C. T. O'Brien, T. Thompson, K. Snyder, and R. Bess, Recent Changes in New York Welfare and Work, Child Care, and Child Welfare Systems. (Washington D.C.: The Urban Institute, 2002). 29. For information on NYSIS sample, survey, and weighting procedures see http:// www.columbia.edu/cu/ssw/projects/surcent/. 30. New York City Administrative Code, Section 6-109. 31. Giannarelli, Adelman, and Schmidt, Getting Help with Child Care Expenses. Marcia K. Meyers is Professor of Social Work and Public Affairs at the University of Washington. She earned a Master of Public Administration at Harvard University and a Master of Social Work and a doctorate from the University of California Berkeley. Dr. Meyers's research focuses on public policies and programs for vulner- MEYERS and DURFEE 127 able populations, with a particular focus on issues of poverty, inequality, and policy implementation. Current research projects examine the impact of U.S. state policy regimes on the labor force participation of mothers, on inequality in access to early childhood education and care, and on disposable family income. In collaboration with researchers at the Rockefeller Institute of Government, she is studying the front-line delivery of welfare reforms in eleven sites around the country. She is coau- thor of Families That Work: Policies for Reconciling Parenthood and Employment, and her papers have appeared recently in the Journal of Policy Analysis and Man- agement, Journal of European Social Policy, Social Service Review, Journal of Pub- lic Administration Research and Theory, Social Science Quarterly, and Demography. Alesha Durfee is an assistant professor in Women and Gender Studies at Arizona State University. Her research explores the ways that structural inequalities both constrain and are reinforced by individual behavior, and assesses the impacts of social policies implemented in an attempt to ameliorate these inequalities. Her sub- stantive interests include child care utilization and policy,domestic violence and the justice system, and welfare reform. 128 POLITICS & SOCIETY 1. F. L. Sonenstein, G. Gates, S. R. Schmidt, and N. Bolshun, Primary child care arrangements of employed parents: Findings from the 1999 National Survey of America's Families (Washington, D.C.: The Urban Institute, 2002). 2. A. Burton, M. Whitebook, M. Young, D. Bellm, C. Wayne, R. Brandon, and E. Maher, Estimating the Size and Components of the U.S. Child Care Workforce and Caregiving Population (Washington, D.C.: Center for the Child Care Workforce, 2002). 3. N. Folbre, The Invisible Heart: Economics and Family Values (New York: The New Press, 2001). 4. P. England, “Emerging Theories of Care Work,” Annual Review of Sociology 31 (2005): 381–99. 5. Ibid., 395. 6. L. Goldschmidt-Clermont and E. Pagnossin-Aligisakis, Measures of Unrecorded Economic Activities in Fourteen Countries (New York City: United Nations Development Programme, Human Development Report Office, 2001); J. Gershuny and J. P. Robinson, “Historical Changes in the Household Division of Labor,” Demography 25 (1998): 537–52. 7. K. Cloud and N. Garrett, “A Modest Proposal for Inclusion of Women's Household Human Capital Production Analysis of Structural Transformation,” Feminist Economics 2 (1996): 93–119; Folbre, The Invisible Heart: Economics and Family Values . 8. N. Folbre, “Children as Public Goods,” The American Economic Review 84 (1994): 86–90; P. England and N. Folbre, “The Cost of Caring,” The Annals of the American Academy of Political and Social Science 561 (January 1999): 39–51; J. C. Gornick and M. K. Meyers, Families That Work: Policies for Reconciling Parenthood and Employment (New York: Russell Sage Foundation, 2003). 9. J. C. Gornick, M. K. Meyers, and K. Ross, Public Policies and the Employment of Mothers: A Cross-National Study . Working Paper No. 140 (Syracuse, N.Y.: Maxwell School of Citizenship and Public Affairs, Syracuse University, 1996); A. Crittenden, The Price of Motherhood: Why the Most Important Job in the World Is Still the Least Valued (New York: Metropolitan Books, 2001); J. Waldfogel, “Understanding the ‘Family Gap'in Pay for Women with Children,” Journal of Economic Perspective 12 (1998): 137–56; M. J. Budig and P. England, “The Wage Penalty for Motherhood,” American Sociological Review 66 (2001): 204–25. 10. P. England, M. J. Budig, and N. Folbre, “Wages of Virtue: The Relative Pay of Care Work,” Social Problems 49 (2002): 455–73. 11. England, “Emerging Theories of Care Work.” 12. S. Hofferth, K. Shauman, and R. Henke, Characteristics of Children's Early Care and Education Programs: Data from the 1995 National Household Education Survey (Washington, D.C.: U.S. Department of Education, 1998); Sonenstein et al., Primary child care arrangements of employed parents: Findings from the 1999 National Survey of America's Families . 13. K. Snyder and S. Adelman, The Use of Relative Care While Parents Work: Findings from the 1999 National Survey of America's Families (Washington, D.C.: Urban Institute, 2004). 14. M. Warner, “The Threefold Importance of Child Care: The Need for a New Economic Framework,” 7. Background Paper for the May 2005 Cornell Workshop, Articulating the Economic Significance of the Childcare Sector. 15. Burton et al., Estimating the Size and Components of the U.S. Child Care Workforce and Caregiving Population . 16. Hofferth et al., Characteristics of Children's Early Care and Education Programs: Data from the 1995 National Household Education Survey . 17. L. Giannarelli and J. Barsimantov, Child Care Expenses of America's Families . Occasional Paper No. 40, Assessing the New Federalism Project (Washington, D.C.: The Urban Institute, 2000). 18. Hofferth et al., Characteristics of Children's Early Care and Education Programs: Data from the 1995 National Household Education Survey . 19. M. K. Meyers, D. Rosenbaum, C. Ruhm, and J. Waldfogel, “Inequality in Early Childhood Education and Care: What Do We Know?,” In K. Neckerman, ed., Social Inequality . (New York: Russell Sage Foundation, 2004). 20. L. Giannarelli, S. Adelman, and S. Schmidt, “Getting Help with Child Care Expenses,” Occasional Paper No. 62., Assessing the New Federalism Project (Washington, D.C.: The Urban Institute, 2003). 21. Center for the Child Care Workforce. Current Data on Child Care Salaries and Benefits in the United States . (Washington, D.C.: Center for the Child Care Workforce, 2002); M. Whitebook, C. Howes, and D. Phillips, Worthy Work, Unliveable Wages: The National Child Care Staffing Survey 1988–1997 (Washington, D.C.: Center for the Child Care Workforce, 1998). 22. D. Blau, The Child Care Problem: An Economic Analysis . (New York: Russell Sage Foundation, 2001). 23. S. Helburn, M. L. Culkin, C. Howes, D. Bryant, R. Clifford, D. Cryer, D. et al., eds., Cost, Quality, and Child Outcomes in Child Care Centers (Denver: Department of Economics, Center for Research in Economic and Social Policy, University of Colorado at Denver, 1995). 24. P. England, M. J. Budig, and N. Folbre, Wages of Virtue: The Relative Pay of Care Work . 25. Sonenstein, Gates, Schmidt, and Bolshun, Primary child care arrangements of employed parents: Findings from the 1999 National Survey of America's Families . 26. Giannarelli, Adelman, and Schmidt, Getting Help with Child Care Expenses . 27. Ibid. 28. L. Fender, C. T. O'Brien, T. Thompson, K. Snyder, and R. Bess, Recent Changes in New York Welfare and Work, Child Care, and Child Welfare Systems . (Washington D.C.: The Urban Institute, 2002). 29. For information on NYSIS sample, survey, and weighting procedures see http:// www.columbia.edu/cu/ssw/projects/surcent/. 30. New York City Administrative Code, Section 6-109. 31. Giannarelli, Adelman, and Schmidt, Getting Help with Child Care Expenses .

PY - 2006/3

Y1 - 2006/3

N2 - Although the majority of young children now spend time in nonparental child care, we know relatively little about who provides this care and how its costs are distributed among parents, government, and other family members. In this article we use data from a survey of New York City families with children younger than six to estimate the contribution of parental expenditures, government assistance, and the market value of "donated" caregiving time by family, friends, and relatives. We conclude that uncompensated caregivers provide a substantial share of child care that is "invisible" in conventional economic measures.

AB - Although the majority of young children now spend time in nonparental child care, we know relatively little about who provides this care and how its costs are distributed among parents, government, and other family members. In this article we use data from a survey of New York City families with children younger than six to estimate the contribution of parental expenditures, government assistance, and the market value of "donated" caregiving time by family, friends, and relatives. We conclude that uncompensated caregivers provide a substantial share of child care that is "invisible" in conventional economic measures.

KW - Caregiving

KW - Child care

KW - Gender

KW - Social policy

KW - Subsidies

UR - http://www.scopus.com/inward/record.url?scp=32944460840&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=32944460840&partnerID=8YFLogxK

U2 - 10.1215/00161071-29-1-109

DO - 10.1215/00161071-29-1-109

M3 - Review article

AN - SCOPUS:32944460840

SN - 0032-3292

VL - 34

SP - 109

EP - 128

JO - Politics and Society

JF - Politics and Society

IS - 1

ER -

Who pays? the visible and invisible costs of child care

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this