Who Cares for America's Children? (1990)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

8 The Child Care Market and Allernative Policies In response to a dramatic increase in the demand for out-of-home child care services in the past decade and a half, a diverse array of organized programs and informal arrangements has emerged, and a variety of public policies have been implemented to improve the child care system and its efficiency. In previous chapters we described the delivery and regulation of services and the mix of public policies that support and supplement the child care "market." By market we refer to the interaction of demand for child care (the number of parents who purchase or want to purchase care) and the supply of child care (the amount of child care available); both are influenced by government intervention (policies) in the market. Although the concept of an economic market is foreign to most early childhood professionals, concern about the inadequacy of many existing programs and arrangements, about shortages of services for selected children and families, and about the costs of care in short, issues of supply and demand- are familiar. In this chapter we review what is known about how well the child care market currently meets the needs of parents and children, and we explore several policy alternatives that have been proposed to improve it. Before we do so, however, some general observations are needed about the evidence that is available on these topics. Understanding of the child care market is at an early stage. In some respects the market fails to meet several of the economic conditions that characterize an efficiently operating market: lack of information available to consumers (parents); high transaction costs associated with changing child care arrangements; and resistance to profit maximization by some providers (not raising prices as demand increases). In addition, there are the costs or benefits of a program that are not (or cannot be) resected in the price paid by individual consumers, "externalities." For example, the benefit to society 227

228 WHO CARES FOR AMERICA'S CHILDREN? as a whole of better education for 4-year-olds is not accounted for in the price of a preschool enrichment program. It is also possible that there are several child care markets, rather than one. Distinguishing among various geographic markets or different product (program types) markets, or both, might explain some of the observed supply and demand phenomena, but such analyses have not been done. As a consequence, standard economic tools for measuring supply and demand and the related costs and benefits must be applied with caution. Assessing current practices and considering alternatives is also com- plicated by the dynamic and diverse nature of the child care market. It is one in which providers are a mixture of private for-profit firms, national chains as well as independent operators; private not-for-profit organiza- tions, such as neighborhood churches; public programs, such as Head Start; and individual family day care operators. It is a market in which funds come from the federal, state, and local governments, community groups, philanthropic organizations, employers, and parents. It in fact consists of many segmented, localized markets with little coordination and enormous turnover among providers and changing needs among consumers. It is also a rapidly expanding market and one in which many parents have difficulty obtaining adequate information about how to locate and arrange services that will meet their needs. The difficulties of analyzing the child care market are further com- pounded by the relative inadequacy of data on the current supply of, and demand for, child care services and by the lack of sophisticated analyses of the likely consequences of alternative policies. In short, although under- standing of the issues in this chapter has advanced substantially in recent years, it is still at a fairly rudimentary stage, due to the underdeveloped nature of the relevant analytic base. Nevertheless, a number of conclusions can be drawn by assembling the often fragmented, existing information, although many of these conclusions are highly qualified. A common perception about paid child care among parents, provid- ers, and politicians-is that there is a major, perhaps even severe, short- age of supply. However, there are three dimensions, often not clearly distinguished, to this perceived shortage. One dimension relates to the sufficiency of the number of places for the children of parents who wish to purchase care of the prevailing quality at market-determined prices: we refer to this dimension as availability. This kind of shortage may arise because demand is temporarily increasing faster than supply or because particular kinds of care are not available in certain locations. It is most vividly evident in the long waiting lists of many programs, as well as the high ratios of applicants to places in many private nurseries and preschools. The second dimension is affordability: Are the available places offered at prices that parents who need or want out-of-home child care can afford to

THE CHILD CARE MARKET AND ALTERNATIVE POLICIES 229 pay? The third dimension is quality: Regardless of the number of places or their affordability, what is the quality of care offered? AVAILABILITY OF CHILD CARE Although it might seem simple to count the number of child care places available, it is not. Many exist in unlicensed, unregulated centers and homes, and, as we discussed in Chapter 6, there is no national system to collect standardized data on the supply of child care services. It is also difficult to discuss availability without also considering the cost and quality of services. What is known about the availability of child care? The predominant form of nonparental care for all children 12 years old and under remains relatives, the majority of whom are not paid or are paid very little for their services. However, reliance upon care by relatives has been rapidly diminishing in recent years. In 1965 nearly two-thirds of nonparental care for children aged 5 or younger was provided by relatives; in 1985 about one-half was provided by relatives (Bureau of the Census, 1987~. Currently, the proportion is thought to be about 40 percent, and, as more and more women enter and remain in the labor force, the share is likely to decline even further. Nevertheless, relative care remains the least expensive form of nonparental care and an important resource for low-income families, who rely on immediate and extended-family members more than do middle- income families (McGroder, 1988~. And indeed, one study suggests that on one isolated objective indicator of quality caregiver/child ratio-care by relatives is superior, on average, to care by nonrelatives (Waite et al., 1988~. Only 18 percent of children under age 5 are in group or center care, most of which is licensed and, therefore, potentially countable from administrative data. About 32 percent are in family day care homes, approximately 60 to 90 percent of which are thought to be unregulated. And 11 percent of children under age 5 are cared for by nonrelatives in the child's own home. The supply of family day care and in-home babysitting is very difficult to measure. The existing data, however, suggest that the availability of nonrelative care differs according to children's ages and their special needs particularly needs related to economic disadvantage and disabilities. Infants and Toddlers The most common (and some argue the preferred) form of nonrelative care for children under age 3 is a family day care home, in which a provider looks after other people's children in her own home. For employed mothers

230 WHO CARES FOR AMERICA'S CHILDREN? during 1984-1985, 23 percent of infants (ages birth to 12 months) and 27 percent of toddlers (ages 12 to 36 months) were cared for in the home of an unrelated caregiver (Bureau of the Census, 1987~; only 14 percent of infants and 17 percent of toddlers were in child care centers. Center care for infants and toddlers is increasing but at what rate is not known. Thus, the predominant form of out-of-home care for infants and toddlers is family day care. Most family day care homes appear to operate in an underground market (see Chapter 6) in which prices are relatively low and caregivers do not pay taxes on their income from child care, although many parents who use this type of care currently receive a tax credit. In the panel's 1988 survey, state licensing offices reported 198,257 licensed family day care homes. If 10 to 30 percent of homes are licensed, as many observers suggest, there may be as many as 1.2 million family day care homes in the United States. How does this presumed supply relate to the demand for places? Hofferth and Phillips (1987) estimate that the number of licensed homes increased by about one-third between 1977 and 1986. But during the same period mothers of infants and toddlers entered the labor force at a much faster pace. The percentage of mothers with children under age 3 who were employed or looking for work rose from 32.6 in 1975 to 52.7 in 1988, an almost 62 percent increase (Bureau of Labor Statistics, 1988~. Moreover, the absolute number of young mothers rose considerably during this period, as the large baby-boom cohort reached the prime childbearing ages. Although each new family day care home presumably can care for more than one child, the increase in places in family day care homes does not appear to have matched the increase in the number of infants and toddlers of employed mothers. Looked at another way, the number of young children with mothers in the labor force has increased dramatically. For example, the number of infants with mothers in the labor force nearly doubled from 977,000 in 1975 to 1,796,000 in 1985 (Hofferth and Phillips, 1987), and in March 1988 there were 3.1 million children under age 2 with mothers in the paid labor force (data from Current Population Surrey). This increase in the number of employed mothers of infants and toddlers has also reduced the pool of potential providers of home care. Direct evidence of a shortage of infant care was provided by a recent survey of the child care market in three low- income urban areas, which found relatively little center-based care available to infants and no excess capacity of infant care either in family day care homes or in centers (Kisker et al., 1989~. Staff in resource and referral agencies consistently report that the highest demand is for places for infants and toddlers and that requests are more difficult to fill for them than for older children (Patricia Siegal, California Child Care Resource Referral Network, personal communication, May 23, 1988~. In a number of surveys, employed parents with infants have been more likely than parents of older

THE CHILD CARE MARKET AND ALTERNATIVE POLICIES 231 children to report difficulties in finding their current arrangements (see Galinsly [1988] for a summary). According to Grubb (1988), several recent commission and task force reports in California suggest that availability is more of a problem for parents of infants than for parents of toddlers and preschoolers. Thus, although aggregate national data do not exist, other evidence suggests that, in general, the supply of child care places for infants and toddlers has failed to keep pace with the demand. Preschoolers The rate of increase in the number of preschool children (ages 3 to 5) with working mothers was smaller than that for younger children: from 3,872,000 in 1975 to 4,984,000 in 1985, a 28 percent increase (Bureau of Labor Statistics, 1988~. Almost all 5-year-olds are enrolled in a school program, although fewer than half are in full-day programs (Kahn and Kamerman, 1987~. Among 3- and 4-year-olds, the predominant form of nonrelative care is some type of group care, for example, nursery school, prekindergarten, or a child care center. The number of licensed child care centers alone almost doubled over a 12-year period: from approximately 34,000 in 1976 (Coelen et al., 1979) to 64,879 in 1988 (panel survey). Although all licensed centers are not operating at full capacity, the number of available licensed center care places increased from approximately 1 million to 2.1 million during that period (Haskins, 1988~. Of course, it is possible that the number of places for preschoolers still is inadequate, despite the rapid growth of centers and the existence of part- day programs and family day care homes. Not all licensed centers operate full-day or full-year programs. Evidence suggests that many children are on waiting lists for places at child care centers, but caution must be exercised in interpreting this finding as an indication that child care is unavailable. These queues might be for ones in desirable locations or for ones that provide special opportunities for parents and children, such as especially gifted teachers or cost subsidies. If so, queues would not necessarily indicate an absolute lack of availability. Also, waiting lists tend not to be routinely updated and therefore may contain names of children who have since been placed in other care. A recent survey in three cities found that child care centers were operating at 92 percent of their capacity, but it also found significant (SO percent) unutilized capacity in family day care for preschool and school- age children (Kisker et al., 1989~. An important question is whether the unutilized capacity is accessible to parents in need of care now (or in the future). If it is not accessible due to a lack of information or inconvenient location, the increased demand for care for preschoolers

232 WHO CARES FOR AMERICA'S CHILDREN? that will undoubtedly be prompted by implementation of the 1988 Family Support Act (FSA; see Chapter 6) could result in more obvious shortages. Other indirect evidence suggests that availability is less of a problem for preschoolers than for toddlers and, especially, for infants. After a comprehensive examination of national data on the availability of child care, Kahn and Kamerman (1987:14) concluded: Parents continue to complain about shortages, and most requests for help in finding care are for this age group [infants and toddlers]. Ibe supply of services for 3- to 5-year-olds appears to be quantitatively adequate. However, much of what is available is still only part-day, as parents seek full-day care and as many preschool programs, both full- and part-day, are more expensive than most parents can afford. In sum, the evidence of a shortage of child care places at prevailing prices for preschoolers is currently not persuasive, although availability undoubtedly varies by geographic region. School-Age Children Family day care homes are the dominant form of paid care for older children of working mothers. It is used most often by mothers who work full time and primarily for children 6 to 8 years old (Cain and Hofferth, 1987~. As detailed in Chapter 6, the number of before- and after-school programs is growing, both in public schools and in other community agencies and organizations, but the number of children who need such care appears to far exceed the available program places. Although it is very difficult to compare the supply and demand for school-age child care because of limited data, there is a serious concern about the large and growing number of children who are without adult supervision during nonschool hours. The Bureau of the Census (1987) reports that approximately 2.1 million elementary school and junior high school students are latchkey children. The U.S. Department of Labor (1988) concludes that this may well be the largest shortage in child care, and Hofferth (1988:564) suggests that as the current group of preschool children ages, there may be "a growing population of school-aged children who are unsupervised when they are not in school." Although school-age children who go unsupervised during non- school hours are not a new phenomenon, their growing numbers coupled with increasing incidence of drug use, youth violence, and other problem behaviors have made the care for these children a special concern. Children With Special Needs Child care providers, teachers, social workers, special educators, par- ents, and policy makers all believe there is a shortage of child care services

THE CHILD CARE MARKET AND ALTERNATIVE POLICIES 233 for children with special needs. Two types of special needs merit special attention: children from economically disadvantaged families and children with disabilities. Currently, there are no systematic data on the demand for or the supply of child care services for children with special needs; however, there are several indicators that a shortage exists. Economically Disadvantaged Children An estimated 25 percent of children under the age of 5 are living in poverty (Bureau of the Census, 1988~. In terms of chid care and development, these children and their parents have special problems. As discussed in Chapters 3 and 4, poor children have been shown to benefit from compensatory education programs (such as Head Start, Chapter I school-based programs for 4-year-olds, and child care services funded by the Social Services Block Grant [SSBG] program) and from programs that address nutritional and health needs (e.g., Head Start immunizations and U.S. Department of Agriculture food and nutrition programs). As these programs suggest, children from economically disadvantaged families are the target of several public intervention programs (see Chapters 6 and 7~. Rut many boor children are not served by these programs. Head Start, for example, serves fewer than 20 percent of the income-eligible population of 3- and 4-year aids despite increased funding since 1980. In 1981 it was estimated that SSBG programs served only 13 percent of the eligible children; since then the number of eligible children has grown, but funding has not (Reismon et al., 1988~. In one study, low-income women were more likely than others to report that they would work if affordable child care were available. The new welfare reform legislation acknowledges a shortage of child care for low-income families by specifically requiring that child care services be made available so that mothers with young children can participate in job training or seek employment. Implementation of the FSA may significantly increase the amount of care for economically disadvantaged children. The parents of these children also have special needs, some of which are not addressed by the current programs. For example, Head Start helps poor parents develop parenting skills, but it is primarily a part- day child development program, and it does not provide child care for parents working full-time. Yet 22 percent of all children aged 3 to 5 who live in poverty have mothers who work full time (Bureau of the Census, 1988~. Low-wage jobs, geographic location, irregular work schedules, and transportation needs constrain many low-income parents in finding child care. Public child care funding for low-income families varies dramatically by state. California and Massachusetts have made major commitments to ~-. rim

234 WHO CARES FOR AMERICA'S CHILDREN? low-income child care, but other states have not. For example, Blank and colleagues (1987) found that half of all the counties in Kentucky provide no child care assistance for low-income families. In New York City, publicly funded child care is available for only 20 percent of the eligible children (Blank et al., 1987~. Although there has been rapid growth in for-profit child care, partic- ularly by large corporate chains, few of these programs serve children in low-income families. Restrictions under the SS8G program make it difficult for these providers to cover their costs in many states, and therefore there is little economic incentive for new centers or family day care homes to open in rural areas or inner cities where there are large concentrations of low-wage jobs and poor families. Hours of service may also be an important issue affecting the availabil- ity of care for children from low-income families. In a study of employed mothers with children under 6 who receive support from Aid to Families with Dependent Children (AFDC), Sonnenstein and Wolf (1988) found that one-third required care after 5:30 p.m. and one-fifth required care after 8:30 p.m; 70 percent of that care was provided by relatives. As noted in Chapter 6, it is not known if these women work late hours because that is when inexpensive or free child care is available or if these are the only work hours available and they must use relatives because other types of care are in short supply (at any price) during these hours. Children With Disabilities There are very few data on the availability of care for children with handicapping conditions. Under national criteria specified in the Education for the Handicapped Act (P.L. 94-142), it is estimated that 1 to 2 percent of all infants will be born with some disabling condition (Scott, 1988~. Depending on the definition of disability and high risk, the numbers and costs of caring for those children vary tremendously. However, it appears that both the number of children and the need for out-of-home care has been increasing faster than the supply of such care. Public policies to dein- stitutionalize children with disabilities and require that they be integrated into programs with the least restrictive environments have exacerbated the need for specialized programs and caregivers. There are more children being diagnosed with serious emotional problems, and advances in modern medicine have lowered the death rate of high-risk infants. As more of these children live longer at home, the diagnostic, therapeutic, and medical costs of their care have increased. These increased costs may in turn necessitate more mothers seeking employment. It is not known whether parents of children with disabilities choose to provide full-time care themselves rather than seek employment and

THE CHILD CARE MARKET AND ALTERNATIVE POLICIES 235 out-of-home care or whether adequate services are simply not available. Under amendments to the Education for the Handicapped Act (P.L. 99- A57), decisions concerning whether or not to serve infants and toddlers are left to the discretion of the states. In Florida, for example, only a few districts serve 3- to 5-year-olds or those up to 2 years old; for the younger group, only visually impaired and hearing-impaired children are served. Scott (1988) concludes that, in Dade County, only 199 places are available annually for an estimated 265 to 530 handicapped infants and toddlers potentially in need of child care. When services for children with disabilities are available, they are used. Head Start requires its local programs to reserve 10 percent of their places for children with disabilities, and approximately 65,000 children with professionally diagnosed handicaps are now served by Head Start. In a small exploratory study, Fink (1988) found that a lack of child care programs for school-age children with disabilities resulted in employment problems for parents, especially for single parents. Evidence suggests that there may be a shortage of care for disabled children and that this shortage may be greatest for infants and toddlers, school-age children, and children from low-income families. In sum, for all children under age 6, the evidence suggests that, the younger the child, the more serious the availability problem. Finding a place seems to be most difficult for the parents of infants, somewhat less difficult for the parents of toddlers, and least difficult for parents of preschoolers. Finding places for school-age children and those with disabilities also appears to be difficult. All of these availability problems are compounded for children from economically disadvantaged families. If the places that are available are not affordable to most parents, they are not really available. AFFORDABILITY OF CHILD CARE Not all employed mothers pay cash for child care. In a sample of young employed parents using child care in 1985, 77 percent paid for care for their youngest child under 5; 57 percent paid for care for their youngest child over 5 (Hofferth, 1988~. The U.S. Department of Labor (1988) estimates that families who do pay for services, spend more than $11 billion per year: approximately $8.6 billion by married couples with both parents working and $2.5 billion by single working mothers. There are two significant aspects of the affordability issue: the absolute amount spent for care and the proportion of total income spent for care.

236 WHO CARES FOR AMERICA'S CHILDREN? Amount Spent For families whose youngest child was under 5 and who paid for at least 30 hours of child care per week in 1985, the amount spent was approximately $35 per child per week (Hofferth, 1988~. On an annualized basis, this totals $1,820, considerably less than the $3,000 estimate frequently cited (Clifford and Russell, 1989; Haskins, 1988~. However, the $1,800 figure may be low for several reasons. It is an average that includes care that is less than a full day; it averages the costs of center care, family day care, and relative care; and it does not adjust for large regional variations. In addition, the population used in the survey is from the National Longitudinal Survey of Youth (NLSY): parents in this survey are generally younger aged 20 to 27 than the majority of parents who purchase child care services. Older parents, who usually have higher incomes, typically spend more on care. Analysis of data from the Survey of Income and Program Participation (SIPP) found that 27 percent of the women surveyed paid more than $50 per week per child (Bureau of the Census, 1987~. Kisker and colleagues (1989) report the median total expenditure for those paying for care was $50 per week These estimates reflect the amount parents report spending: they do not reflect the actual costs to the provider or the fees they charge, since many parents benefit from public subsidies that reduce the amount they pay for child care. The differences in the amount paid for care reported in the NLSY are informative. In direct outlays, the least expensive type of care was that provided by relatives, about $30 per week; family day care homes and center or nursery care, about $37; and a babysitter in the home (the most expensive form of care), $42. On an hourly basis, relative care was also the least expensive, at about $1.14 per hour, and babysitter care was the most expensive at about $1.60 per hour. Family day care homes were $1.17 per hour, and center and nursery school care was about $1.40 per hour. For a 40-hour week 52 weeks per year, the fees paid ranged from $2,280 for family care to $3,200 for babysitter care. Despite the increased use of out-of-home care, weekly expenditures for child care appear to have risen only modestly in recent years (Hofferth, 1987~. Not unexpectedly, Hofferth (1987) and Brush (1987) both found that the number of children in a family and the mother's employment status were the most significant predictors of how much parents paid for care. Mothers employed full time obviously paid more than mothers employed part time because they purchased more hours of care. In addition, mothers living in metropolitan areas, those with higher educations, those who are white, and those living in families with higher earnings were all likely to pay more money than others for child care.

THE CH LD CARE MARKET AND ALTERNATIVE; POLICIES Proportion of Income 237 Among all families paying for care for children under age 5, the cost averages about 10 percent of family income (Grubb, 1988; Hofferth, 1988). This represents a substantial expense, comparable to the share of income most families spend on food. Among low- and moderate-income families, however, the burden of child care expenses is much heavier. Data from the 1985 NLSY showed that child care expenses were 30 to 50 percent of the family incomes of those earning under $5,000 per year; 15 to 20 percent for those earning $5,000 to $9,999; 10 to 15 percent for those earning $10,000 to $14,999, 5 to 10 percent for those earning $15,000 to $49,999; and under 5 percent for those with incomes of more than $50,000 (Hofferth, 1988~. For a single parent who earns the minimum wage and pays a caregiver $30 per week for one child, the data suggest that that family is spending 22 percent of its gross income on child care. Hofferth (1988) finds that, overall, poor families paid an average of 23 percent of their incomes on child care; nonpoor families paid only 9 percent. Although low-income employed families do receive some subsidized care and although they tend to use the least expensive form of out-of-home care, their relative expenditures are vastly larger than those of higher income families. The potential burden of out-of-home child care on low-income house- holds is even greater when children are infants or if they have disabilities. One study has found that infant care costs run, on average, one-third higher than the costs of care for preschoolers (Grubb, 1988~. Head Start estimates the additional cost of serving a child with disabilities (compared with a child without disabilities) at $1,000 per child per year (Brush, 1988~. Additional staff (a major cost component) and services required for children with disabilities account for this difference. Hofferth (1988) also found that single-parent families tend to pay a larger proportion of their incomes on child care than do two-parent families, in part because they have lower earnings and less flexibility to reduce their expenses- for example, by working different shifts. As indicated in Chapter 2, shift work is surprisingly common among two-parent, two-earner families (Presser, 1988~. One study found that in one-third of all such families with children under 6 in which both parents worked full time, one parent worked other than a regular day shift (Presser and Cain, 1983~. Although out-of-home child care expenses absorb 20 percent or more of the gross income of working poor families who use it, these costs are not borne by a large proportion of all poor households, since the proportion of the poor who work full time all year and use out-of-home care is small. Paradoxically, this fact suggests that the burden of out-of-home care is even more important than the data suggest: it implies that child care costs may discourage work altogether for some parents.

238 WHO CARES FOR AMERICA'S CHILDREN? In response to a hypothetical question about whether they would work if affordable child care were available, 26 percent of nonworking mothers with preschoolers said they would seek employment if child care were available at a "reasonable cost." This response was more prevalent among women from low-income families, unmarried women, black women, and women who had not finished high school (O'Connell and Rogers, 1983~. Blau and Robins (1986) found that low-income women who faced higher child care costs were actually less likely to enter the work force than those who faced lower costs. Moreover, employed women were also more likely to leave paid employment if they faced higher child care costs than were those who had lower costs. This evidence suggests that the cost of care is a significant constraint on employment for some women, particularly for those with low incomes. An important unresolved question is whether out-of-home child care costs are most appropriately considered in relation to total family income or to the wages (or potential wages) of the spouse earning less in a two-adult household. This conceptual issue is of particular importance when considering middle-income couples. Although they may spend up to 10 percent of their total income on child care, that amount typically represents 25 percent of the wife's income (or potential income) (Waite et al., 19884. If the prevailing view is that the costs of child care should be borne by women's earnings net of taxes, then out-of-home child care costs can be a disincentive to employment for many women in middle-income households as well as for those in low-income households. Determining the "appropriate" share of a family's budget to be devoted to child care is not a scientific matter. It entails value judgments that only individual families can make in light of their needs and preferences and other budget constraints. Confronted with the same income and child care costs, one family might choose eagerly to make the expenditure to allow both parents' careers to progress, another family might bear the expense reluctantly out of real or perceived financial necessity, and another might happily or unhappily decide that one parent should withdraw from the labor force. But when the costs of child care and the average share of income actually going to child care are as substantial as they are for many low-income working families, there is no choice possible. Many people would consider this situation a serious problem although just how serious depends on the strength of their views about the benefits of out-of-home child care and of working parents. Overall, we conclude that the lack of affordable child care is a major public concern for low- and even many moderate-income families. It is especially significant for single parents, for young families with infants, and for families with children with disabilities.

THE CHILD CARE MARKET AND ALTERNATIVE POLICIES QUALITY OF CHILD CARE 239 Although employed parents pay a substantial proportion of their in- comes for child care, it is not clear how many are purchasing services of adequate, let alone, high-quality, care. As Chapters 3 and 4 detailed, academic research and best professional practice provide the bases for establishing broad parameters of the quality of child care services. They suggest ranges for group size, staff/child ratios, and other characteristics of child care settings that constitute a minimum threshold for safe and developmentally sound care. In light of that knowledge our assessment raises important concerns regarding the quality of care many children are receiving. As discussed In Chapter 6, the wages and benefits of child care workers are very low. The Child Care Staffing Study (Whitebook et al., 1989) estimates that full-time teachers and aides earn, on average, less than $10,000 annually. No current comparable figures exist for family day care providers, but the evidence suggests that their wages are even lower, although there may be some nonmonetary compensations, such as not paying taxes or caring for their own children at the same time. Fuchs (1988) found that child care workers earn only about two-thirds as much per hour as other women with comparable levels of education. And Hartmann and Pearce (1989) and Whitebook and colleagues (1989) found that workers with more education are not consistently paid higher wages than those with less education. Turnover among child care workers is extremely high. According to U.S. Department of Labor statistics, the annual turnover rate is 42 percent for workers in group programs and 59 percent for those who work In other people's homes (Eck, 1984~. These rates are twice the average for all employed persons; they are comparable to the rates for gas station attendants. Consistency of care is an important component of quality (see Chapter 4~. High staff turnover, undoubtedly fueled by low wages, makes it difficult to provide consistent and, therefore, adequate care. Staff/child ratios are also important to quality, yet many states' regula- tions governing center care do not require staff/child ratios that reflect what scientific knowledge and best practice indicate is appropriate for infants and toddlers. Raising wages, especially for caregivers with more education, has been shown to increase the quality and stability of staff. But raising wages is expensive, since staff salaries constitute between 60 and 85 percent of the operating budgets of centers and nearly 90 percent of the budgets of family day care providers (Coelen et al., 1979; Fosburg, 1981; Lombardi, 1988~. The magnitude of the increases in cost necessary to improve the quality of out-of-home child care was best demonstrated by Clifford and Russell

240 WHO CARES FOR AMERICA'S CHILDREN? (1989~. These researchers estimated the cost per child of care in four hypothetical center care settings that offer full-day care and include infants. Model 1 was an "ideal" center that met the accreditation standards set by the National Association for the Education of Young Children (see Appendix B). Salaries were relatively high $20,000 per year for teachers and $10,000 to $14,000 for assistant teachers although still lower than the average salary of elementary school teachers (Grubb, 1988~. Staff/child ratios were relatively low-for example, 1:4 for infants, 1:8 for 3-year-olds, and 1:12 for school-age children. Modest amounts were budgeted for staff development ($125 per staff member per year), parental involvement ($10 per child per year), and so forth. These costs are similar to those of a full-day Head Start program (Brush, 1988~. Model 2 retained the higher standards and salaries for teachers, staff development, and parental in- volvement, but increased the staff/child ratios- for example, 1:6 for infants, 1:12 for 3-year-olds, and 1:20 for school-age children. Model 3 retained the staff/child ratios of Model 1, but decreased the funds for salaries ($12,000 per year for teachers and $8,320 to $10,000 for assistant teachers), staff development ($50 per staff member per year), and parental involvement ($5 per child per year). Model 4 included both the lower salaries of Model 3 and the higher staff/child ratios of Model 2. Clifford and Russell (1989) maintain that Model 4 typifies existing programs. Their estimate of the cost per child per year for Model 4 is $2,937, an amount at the high end of the $1,800 to $3,000 average cost discussed above. Raising the standards for staff/child ratios but keeping salaries low Model 3-raises the estimated cost by 27 percent, to $3,743 per child. Raising salaries but maintaining the higher stafI/child ratios Model 2 raises the cost relative to Model 4 by 37 percent, to $4,030 per child. The ideal center of Model 1 raises the cost by 79 percent, to $5,267 per child. Even the highest standard (Model 1) pays teachers only $20,000 and allots just one staff member for every four infants. But raising all centers to this standard would require a massive increase in fees charged to par- ents, in government subsidies, or both. Even instituting the more modest improvements In quality of Models 2 and 3 would place the cost of child care beyond the reach of many more parents than are excluded from the market today unless large new subsidies were forthcoming. In family day care homes, increased wages would have an even greater affect since wages account for 90 percent of the cost of care. Parents, Quality, and Costs A major issue in the child care debate is who should determine whether a child's care is of adequate quality. If one wishes to subsidize child care

THE CHILD CARE MARKET AND ALTERNATIVE POLICIES 241 expenses and believes that parents should determine the quality of care for their children, one strategy would be to provide subsidies to low- and moderate-income parents, to refrain from regulating the market, and to let parents make their own decisions about how much to pay and what level of quality to purchase. This strategy is consistent with the widely held value that it is parents' right and responsibility to determine what forms of child care and patterns of childrearing are best for their children. However, research provides some contraindications to this approach to public policy. One recent study examined whether parents pay more for the charac- teristics of child care that are associated with higher quality- regulatable characteristics such as the caregiver/child ratio, group size, physical fa- cilities, and education of the caregiver (Waite et al., 1988~. Analyses of the 1985 NLSY data revealed that parents who purchased better quality care typically did not pay more than other parents. This finding can be interpreted to mean either that better quality care was no more expensive than lesser quality care or that parents did not value the characteristics of quality care as highly as other characteristics (such as location and hours of operation) and therefore were unwilling to pay for them. In fact, better quality care now appears to be subsidized in part by staff, who accept low wages, which in turn leads to high turnover. Part of the reason wages are low may be because parents, even upper income parents, value other factors, including costs, more than characteristics associated with quality in choosing child care arrangements. Studies of parents' search for child care show that surprisingly few parents visit more than one program before deciding where to enroll their children. In one study of parents using proprietary centers, 9 percent of the parents did not even visit the center in which they ultimately enrolled their children (Bradbard et al., 1983~. Kisker and colleagues (1989) reported that in three cities only half of par- ents using out-of-home child care for their preschoolers visited more than one provider before making their selection. For many parents, the over- riding considerations in choosing child care arrangements were convenient location, cost, and hours that matched the parents' work schedule. In a study of callers to a resource and referral service, however, 77 percent were found to have visited more than one setting (Bogat and Gensheimer, 1986~. Powell and Eisenstadt (1980) found that the parents most interested in using resource and referral services are well-educated single parents, new to their residential area and living in neighborhoods with relatively few contacts between neighbors. These parents rely on print as a source of information and tend to engage in a wider search than people who do not use resource and referral services. These preliminary studies suggest that many parents may not be able to weigh considerations of program quality in comparison with considerations of cost, location, and hours. Furthermore, many parents are not well i

242 WHO CARES FOR AMERICA'S CHILDREN? informed about how to identify high-quality child care and what to look for when visiting programs. Because of the relative newness and decentralized nature of the child care delivery system in the United States, it is often difficult for parents to obtain adequate information. Looking for out-of- home care is a new and unfamiliar task that may be difficult for many parents (Grubb, 1988~. Accordingly, the expansion of services that provide better information to parents as well as to providers might improve the quality of care and the match between consumers and providers without significantly affecting affordability. BENEFITS AND COSTS OF ALTERNATIVE POLICIES This evaluation of the child care market suggests that the current mix of child care services does not adequately meet today's child care needs. For some children and families, services are not available at any cost; for others, the cost of available care is prohibitive; and for many, the quality of care is less than research suggests is needed to protect children's health and safety and foster their social and cognitive development. Although we have identified numerous local initiatives to begin to address these inadequacies, they are isolated and do not appear to be sufficient to meet identified needs. There is a need for a broad expansion of federal and state support and involvement. The call for government action is not new, but since 1987 child care has received growing attention. Indeed, more than 100 bills were introduced in the 101st Congress in 1989; the Senate and House each passed a version of the Act for Better Child Care, but they failed to reconcile differences in conference. This movement has been echoed by numerous initiatives at the state and local level (Stephen and Schillmoeller, 1987~. For policy makers, the challenge is to define socially desirable and politically feasible child care objectives and to help meet them with policies and programs that divide the costs and the benefits of child care policies in ways that are fair and efficient. In attempting to achieve each goal one must ask several questions: Who will benefit? How much will it cost? Who should pay? Is it the best way to achieve the benefit? Is the benefit worth the cost? National Objectives and Alternative Policies The first step in assessing the benefits and costs of alternative policies is to understand the relationship between stated objectives and alternative types of policies. The panel has identified the following range of objectives for child care policy, which affect the supply of and the demand for nonparental as well as parental care:

THE CHILD CARE M~4RKET AND ALTERNATIVE POLICIES 243 1. Quality: to improve the quality of out-of-home child care services; 2. Quantity: to increase the availability of child care services, es- pecially care for infants, school-age children, and children with disabilities; 3. Specialized care: to increase the availability of high-quality com- pensatory care for children from economically disadvantaged and stressed families; 4. Costs: to increase the affordability of high-quality care for children from moderate- and low-income families; 5. Parental choice: to increase parents' choices about employment and the type of child care they use; 6. Parental labor force participation: to increase parents' labor force participation; 7. Labor force stress: to reduce the stress on parents participating in the paid labor force to attain or maintain their economic self- sufficiency. These objectives are intended to improve the current environment for children and families. Some, however, may be mutually inconsistent, and to achieve all of them within a short time frame would be economically infeasible. Policy makers must make choices based on the costs and benefits of these alternatives. As described in Chapter 7, there are three general types of policies to consider: support for consumers (parents), support for providers, and support for the infrastructure of the child care system. There are four po- tential funding mechanisms for this support: general revenues, earmarked revenues, tax expenditures (lost revenues), and mandated programs (which would result in reliance on private funding). There are many variations on these basic types of policies and funding mechanisms; we examine several selected combinations. The behavioral consequences of alternative policies are not well understood and are in need of further research. Evidence from the United States and several European countries, however, suggests that government policies in the child care area will have few undesirable effects, or unintended consequences in birth rates, fertility patterns, divorce rates, and labor force productivity (Bane and Jargowsky, 1988; Cherlin, 1988~. Rather, the successful pursuit of policies to improve child care would have generally salutary consequences for children, their parents, and society as a whole. Although the response to any specific policy cannot be predicted with certainty, responses that might be expected, based on economic models, enable us to make some general predictions about the benefits and the costs of various policy options (Connally, 1988; Fuchs, 1988; Garfinkel, 1988~. In a very simplified way, Table 8.1 presents the key types of policies

244 WHO CARES FOR AMERICA'S CHILDREN? and their expected effects. For example, the effect of a consumer subsidy, such as a child allowance on quality of out-of-home care is unclear. Parents might use the money to buy higher quality nonparental care. But if the allowance is used to enable parents to stay home or to buy more hours of care at the same quality level, then the policy does not help improve the quality of nonparental child care. In contrast, a provider subsidy for service delivery should help meet five of the objectives. It may limit parental choice, however, since it only benefits those who purchase nonparental care, and even for those parents it limits their choice of types of care. The costs of selected policies are summarized in Table 8-2. Although limited by the amount of available research, the following discussion reviews the state of knowledge regarding likely benefits and costs of various policy changes and points out the areas of substantial uncertainty that must be considered in decision making. Consumer Subsidies Consumer subsidies provide resources to parents, increasing their hex- ibility in making decisions about whether to provide care themselves or purchase it in the market. Such subsidies may or may not be tied to earn- ings, employment, or actual use of paid child care. Consumer subsidies do not directly affect the supply or quality of child care. These subsidies assume that parents will make the best choice of care for their children and that the market will respond to changes in the demand for out-of-home care. We consider several different forms of consumer subsidies, the likely consumer and provider responses, and the related benefits and costs. Child Allowances Various forms of child allowances are the most generally supportive forms of family policy and are quite common in other Western industrial- ~zed countries. Under the most generous policies, financial resources are made available to all parents with children without regard to their income or employment-related child care expenses. The basic rationale for this approach is to support families' ability to rear children and to increase parental options with respect to child care, while remaining neutral with respect to mothers' employment and family structure, neither encouraging nor discouraging dual-earner, single-earner, two-parent, or single-parent family arrangements. Such an allowance could be funded through general tax revenues. Fuchs (1988) estimates that providing all families an allowance of $2,000 per child under the age of 12 would cost about $83 billion annually. (In contrast, we have conservatively estimated the gross monetary costs of

245 Cal Cat AD ._ o Ct - C~ Ct C) ._ - o it> - ~a .a C) A, o ._ - o Pa C) Go m ¢ CQ ._ C) . _ - ~D  . . ._ ._ U) V) Cal C5 '_ .O AUK An Ct CD -t 70 0~ ~ ydm U. S ~ =._ .= Cal ~ S ~ Cat = ;- O ~ O O ~ ~ 5 C: ·_ Ct ~ it> ~4 ~ 3 ~ Us EM ct 4 - C) ~ ~ ~ - C) .g .= ~ .E o ~ V) ~ V) - C;S 62 X .= _ ~n c) :> Ct5 '- c) c.) c: c' ~ o + + + o + + + + l l l l o  c`. c`. + + o+ o + + + + c`eo + o + + + +l + + o l c~.o+ * + c`- + c`- ++ + o o ~c~ ~: O > ~t)~D O u~ ) .~ ~ ~ ~ ~ ~ ~c) m .~ o ~ ~ o o ) ~ v .~ ~ ~ ~ $- ~ "o o :O ~ ° ~ ~ ~ ~ Ct ~ ca C<S $_, C) ~ o ~ 3 ~ ~ o ,., ,~ == o - ~: ._ ._ C~ ~> Ct ~4 ._ o C~ ~: ._ o ._ Ct _ .= C5 - o b0,~= 1_~ b~ ~ ~ · C~ ~ - - .O ~: ~,~ u, =~= ~ ( ct ~ ~ ~ ~o ~ ~ - ~ ~ ~- o o o - , o ~ ~ ~ ~ cD c> + l oc`- *

246 WHO CARES FOR AMERICA'S CHILDREN? providing adequate care to all children under the age of 13 at $126 billion isee Chapter 7] the result of providing $4,000 to families for every child aged newborn to 5 and $2,000 for every child aged 6 to 12.~i With such an allowance, parents could more easily afford to provide care themselves or to purchase care of adequate quality at prevailing market rates. The net cost to government of providing child allowances would actually be considerably less than the gross cost since it would presumably replace the current personal income tax exemption for children, as well as several current child care subsidies and general income support programs (e.g., AFDC, the dependent-care tax credit, and the earned income tax credit). Moreover, in most countries that have a child allowance, it is subject to income taxes and, therefore, families in the upper income brackets pay back a portion of the benefit. Just how much less the net cost would be is impossible to say, however, since it would depend on other policy changes. The overall result of a policy of child allowances would be a substantial redistribution of income from adults and families without children to those with children. An alternative and less ambitious form of child allowances could be achieved by greater reliance on current tax expenditures, through increased personal income tax exemptions or refundable tax credits. The United States currently has a personal tax exemption for dependent children and a refundable tax credit for low-income families with earnings (the earned income tax credit). More general-income support for families with children could be achieved by raising the personal tax exemption for dependent children: this policy would benefit all families who pay significant income taxes, although larger and higher income families would benefit more than smaller and middle-income families. Converting the exemption to a flat credit would result in the same benefit to middle- and higher-income fam- ilies with the same number of children. A 1986 study (Cherlin, 1988) recommended an increase in the deduction from $2,000 to $5,000. Espen- shade and Minarik (1987) estimate that doubling the personal exemption for dependent children from $2,000 to $4,000 would cost about $19 billion per year. One criticism of higher tax exemptions and tax credits is that they provide little benefit to low-income families who have little or no income tax liability; the benefits disproportionately go to upper income families who may not need them as much. One way to change this effect is to make the tax credit refundable: If parents do not owe a tax equal to the amount of the credit, they would receive a check from the government for the amount of the credit. 1b reduce the costs of making the tax credit 1 This is a conservative estimate because it does not take into account the higher costs of child care for children with special needs, which are difficult to quantity; those costs could potentially add tens of billions of dollars to the estimate.

THE CHILD CARE MARKET AND ALTERNATIVE POLICIES TABLE 8-2 Estimated Costs of Proposed Policy Alternatives 247 Policy Alternatives Annual Costs (billions) Source Consumer Subsidies Child allowance $4,000/child (0-5 yr) $126.0a Panel on Child Care $2,000/child (6-13 yr) Policy $2,000/child (0-15 yr) 83.oa Fuchs (1988) b $700/child (0-12 yr) 33.0 Barnes (1988) Doubled personal tax exemption 19.0 Espenshade and Minarik (1987) Refundable child care tax credit 3.4 Barnes (1988)b Cap on dependent care talc credit ~.4 Sen. Orrin Hatch and Rep. Nancy Johnson (1988) Tax credit for low-income families 2.2 President Bush (1989)C Unpaid 10-week parental leave 0.1 General Accounting Office (1988) Fully paid Month parental leave 5.0 Frank (1988) Provider Subsidies Universal child care 126.0d Panel on Child Care Policy Extending Head Start to full day, 2.3 Brush (1988) full year for existing clientele Expanding current Head Start to serve 5.7 all eligible ~ and ~year~lds Brush (1988) and Panel on Child Care Policy Expanding Head Start to all eligible 8.7 Brush (1988) and Panel children, full day, full year on Child Care Policy Educational services for ~yearmlds 1.0 Sen. Edward Kennedy (1988)C Expanded services 25 Sens. Christopher Dodd and Orrin Hatch (1989) c Infrastructure Subsidies 0.4 - aThese are gross costs. The net costs would likely be considerably less since offsets would occur from elimination of personal income tax exemptions for children and other federal program costs. bcRobin Barnes, The Urban Institute, personal communication, Aug. 3,1988. dProposed legislation. Would not be used by everyone, unless mandated like public schools; these are also gross costs.

248 WHO CARES FOR AMERICA'S CHILDREN? refundable to low-income families, some observers have suggested that the tax credits be phased out at higher levels of family income (Robins, 1988~. The costs of a hypothetical refundable child allowance in the form of a $700 tax credit for all children under the age of 15 have been estimated by Robin Barnes (The Urban Institute, personal communication, Aug. 3, 1988~: approximately 30 million families would receive an average of $1,100 each, for a total cost of $33 billion annually. Simultaneously eliminating the current personal exemption and the dependent care credit would restore $23 billion to the federal government, resulting in an overall net cost of $10 billion. Raising the level of the credit would obviously increase the cost substantially, but it could be phased out at higher income levels in order to reduce the cost. A refundable tax credit could also be linked to employment and income, as is the current earned income tax credit. In his fiscal 1990 budget, President Bush proposed a $1,000 refundable per-child tax credit that would go to families with children under the age of 4 and annual earnings of between $8,000 and $20,000. At an estimated eventual annual cost of $2.2 billion, this credit would be refundable and available to all income-eligible families with children, regardless of whether one or both parents worked and regardless of the auspices of child care. This policy would be limited to low-income families, but it is neutral as to mothers' employment and form of child care. It is far less expensive than any policy providing support to all families regardless of income or employment, although, as many observers have noted, $1,000 does not go very far toward the purchase of high-quality child care in most localities. Elements of this approach were incorporated in the Senate and House versions of proposed child care legislation in the 101st Congress. The benefits of a child allowance differ for different groups. For ex- ample, an allowance of $1,000 per child would bring little additional benefit to parents on welfare, because the allowance would tend to substitute for the current welfare benefit. Among working mothers, one would expect a slight decrease in employment, both in the number of mothers who work outside their homes and in the number of hours they work. Some mothers would choose to care for children themselves, thus decreasing slightly the need for nonparental or paid child care. Some employed parents might use the benefit to increase the quality of the child care they purchase. A1- lowances would clearly benefit single mothers who must work Low-income families would benefit if the allowance is provided as a direct payment in addition to current welfare benefits or as a refundable tax credit. There is no guarantee, however, that any of the money would be used for children at all. Clearly, child allowances and tax exemptions or credits that do not vary with parental income are expensive and only modestly redistribute

THE CHILD CARE MARKET AND ALTERNATIVE POLICIES 249 income among income groups, although the redistributive effects depend on what they replace and how they are taxed. Overall, resources would be transferred from individuals and families without dependent children to those with children. Such a change may be responsive to a growing perception that children are an important national resource in which societr as a whole should invest. But much of the benefit of tax exemptions or credits as they are currently structured accrues to middle- and upper income families. In a time of large budget deficits, policy makers as well as taxpayers must weigh the importance of a strong symbolic statement in support of all families with children against the needs of many low- and moderate-income families. Policies that limit eligibility on the basis of income, however, have other drawbacks. Some observers believe that they perpetuate a stigma against support for children. Rather than a national policy in support of all children, beneficiaries are limited to those defined as economically disadvantaged, perhaps regardless of how high the eligibility limit is set. "Means testing" also introduces incentives and disincentives that influence employment behavior. Fuchs (1988:134) provides the following example: For instance, suppose a woman is eligible for a $4,000 allowance for her children if she makes only $10,000 per year, but the allowance falls to $2,000 if her income rises to $15,000 per year. Her effective tax on the extra $5,000 of earnings would be about 60 percent, because she loses $2,000 in benefits as well as paying about $1,000 in additional income tax and social security tax. Ibis high marginal tax lowers her incentive to seek a better job or to work harder. Another disadvantage of varying the allowances inversely with income is that it distorts the prenatal effect toward low-income families. Subsidy for Paid Child Care The second major form of consumer subsidy is directly linked to nonparental child care expenses related to employment. As we have noted, the current dependent care tax credit subsidizes employed parents who purchase child care services: child care is regarded as a work expense, at least part of which is credited in calculating a family's income tax liability. This approach is a form of tax expenditure. Subsidies available to low- income families participating in AFDC or the Job Raining Partnership Act (JTPA) program, which offer work-related income-disre~ard Programs. Operate in a similar way. The current dependent care tax credit and income disregard programs and any expansion of these programs reduce the effective price of child care, thus benefiting employed parents with young children. Lower costs presumably would induce some parents to increase their use of nonparental c, ~,

250 WHO CARES FOR AMERICA'S CHILDREN? care, that is, encourage mothers to work; others who are already employed may use the cost savings to purchase higher quality care. There is no direct link between employment-related tax credits and the quality of care, however, unless the availability of the credit is limited to child care services that meet specified standards. Current tax credit benefits are distributed to an estimated 9 to 12 percent of working families who pay for child care and claim the credit to reduce their income tax liability. The costs are distributed across all other taxpayers. In other words, the estimated $300 to $700 average family benefit to those using the credit is paid for by an average "tax" of $40 to $80 on nonusers. On a much smaller scale, the redistribution of child care resources associated with the low-income disregard programs similarly involves transferring a small amount of resources from a large number of taxpayers to relatively few beneficiaries. Policies that support increased parental employment entail a number of benefits that accrue to taxpayers in general: higher total income tax receipts; potentially enhanced economic growth due to greater stability in the work force arising from workers' use of more reliable (and generally more costly) child care; and lower income transfer costs due to the greater labor force participation of parents. In addition, there may be long-term secondary effects of higher work-related income of parents and more high- quality child care for children. The 1988 Family Support Act reflects a growing public consensus that low-income mothers should increase their labor force participation. Yet it also acknowledges that this goal will not be attained unless child care is available and affordable. The work requirements of F3A are likely to increase the demand for out-of-home child care over the coming several years. However, current budget estimates of funds that will be allocated for child care are low. There is considerably less consensus, however, on the goal of maintaining or increasing labor force participation of all mothers, especially those in middle- and upper middle-income families. Critics of the current tax credit argue that it penalizes women who stay at home to care for their children. Several proposals have been put forward to extend the credit to parents who stay home, making it similar to a child allowance and neutral toward women's employment. Such a change would disproportionately benefit affluent families, since they are more likely to have a spouse at home who would receive this additional benefit (see Marr, 1988~. However, it would also enable at least some women who are employed to decrease their hours of work or to leave the labor force altogether. Critics also argue that tax credits unnecessarily benefit middle- and upper income families who have less need for support than low-income

THE CHILD CARE MARKET AND ALTERNATIVE POLICIES 251 families. Other alternatives, what Marr (1988) calls "targeting" bills, at- tempt to reduce the tax credits for upper income families or increase the credits for low-income families. For example, one proposal would eliminate the current dependent care credit for those making more than $75,000. The congressional Joint Committee on luxation estimates a savings of about $375 million a year one-tenth of the cost of the current credit. This relatively small savings would be achieved by penalizing working women in families with two middle or high incomes. Many critics argue that if this change were implemented, a tax credit that was intended to enhance women's opportunities to seek employment outside their homes would be- come another antipoverty policy. Another proposal would make the current credit refundable up to $400 per child under 6 years of age and cap the refund at the level of the household's Social Security payroll tax liability. Such a policy would provide benefits to about 20 percent more households than benefit from the current dependent care tax credit, at a cost about 25 percent higher than the current credit. Unlike child allowances, child care tax credits guarantee that the benefit is spent on (nonparental) child care. They also provide benefits to a wide range of working parents while supporting parental flexibility in the choice of care. They do not directly influence quality, however, and may present a cash flow problem for low-income families who must pay for care on a weekly basis but wait until the end of the tax year for reimbursement. A system of child allowances or refundable tax credits would more directly address the needs of low-income families. None of these forms of support, however, directly address the issue of quality. The amounts of the subsidies per family are modest in relation to the amounts families actually pay for child care and, therefore, they may have only marginal effects on the affordability of care for low-income families. Parental Leave A third form of consumer subsidy, intended to provide working par- ents with more choice and to reduce the stress of work-family conflicts, is parental leave. Proposed policies range from unpaid leave of a few weeks, which guarantees job security and extended medical coverage, to the generous paid leave policies characteristic of many European countries. Current proposals for unpaid parental leave mandate government policies, but the costs would be paid by employers and employees, not by taxpayers. Arguments in support of parental leave policies relate to the develop- mental needs of children and parents, women's equity in the work force, and the labor force efficiency. Proponents argue that parental leave policies will result in better parenting and childrearing practices by giving parents more time to establish relationships with their infants and reducing the

252 WHO CARES FOR AMERICA'S CHILDREN? inevitable stress that accompanies combined work-family roles after the birth of a baby (Zigler and Frank 1988~. Parental leave policies increase parental choice about whether to work outside the home or to work at home caring for a young child. Extended paid leaves are expected to increase the number of parents, primarily mothers of infants, who would temporarily leave their jobs and care for their children themselves, thus easing the demand for infant care services. Paid leave, however, could be expected to have a much stronger impact than unpaid leave on reducing the demand for infant care since un- paid leave is not a real option for many low- and moderate-income mothers or fathers. At the same time, paid leaves can be expected to promote equal opportunity for women workers by increasing their attachment to the labor force and their seniority in their jobs. These outcomes, in turn, would lead to greater promotion opportunities and higher relative expected wage rates (higher than if women left the work force completely, but lower than if they did not have or take any parental leave). Higher family income may have secondary effects on the type and quality of child care that parents purchase. The primary beneficiaries of parental leave policies are working parents of very young children. Proposed benefits include extended job security, continued health insurance coverage, and, under some proposals, wage replacement income. Parents who use parental leave purchase less out-of- home child care during the period when they are on leave, which helps offset their wage loss. Job guarantees also prevent some women from having to participate in a welfare program. The parents who use and benefit from parental leave also pay some costs, however. They experience an immediate loss of income if leaves are unpaid or paid at an average rate that is lower than their employment earnings; they may also receive lower wage rates in the future because of less work experience. In other words, future earnings may be lower than if parents had not taken time off. Unlike the other subsidies discussed so far, parental leave policies may affect the behavior of employers. Proponents of such policies argue that they promote greater attachment of workers to their employer and, thereby, increase the stability of the work force and may induce associated efficiencies in the workplace. Critics argue that such policies encourage an increase in the incidence of employee leaves, thereby imposing significant direct costs and productivity losses on employers because of the need to hire temporary workers. The net result, they argue, may be increased consumer costs, lower overall wages to compensate for higher nonwage labor costs, or discrimination against women of childbearing age. Cost estimates for different types of parental leave vary enormously depending on the length of the leave, whether or not it is paid, and the number and type of benefits that are maintained during the leave time. It is

THE CHILD CARE MARKET AND ALTERNATIVE POLICIES 253 very difficult to estimate the costs, because employers are already providing many types of leave (see Chapter 7) and because there are no reliable data on who would use various kinds of leave and how employers would compensate for the workers on leave. For the unpaid leave version of the proposed Family and Medical Leave Act, however, the General Accounting Office (GAO) used several national surveys and carried out its own survey of 80 firms in two metropolitan labor markets to estimate its costs and effects (General Accounting Office, 1987~. The GAO estimated the cost to firms with 35 or more employees of a 10-week unpaid leave to care for a new baby to be $102 million annually. Approximately 931,000 parents would be covered. In recent testimony, GAO officials estimated a 30 percent increase in costs for health insurance, increased births, and expanded employment (General Accounting Office, 1989~. They further found that employers were more likely to reallocate work than hire replacement workers for those on leave. In general, they did not conclude that reallocating or replacing workers would result in a significant loss of output or higher costs. Frank (1988) provides rough estimates for a range of paid parental leave options. Based on labor force, age, and birthrate data, these estimates range (in 1983 dollars) from a low of $1.25 billion annually for 3 months' leave with 50 percent earnings replacement to slightly more than $5 billion for 6 months' leave with 100 percent earnings replacement. This estimate does not take into account what employers are currently providing, but few offer paid leave for any extended period of time. Spalter-Roth and Hartmann (1988) also point out that not having parental leave has a cost to women workers and to society. They also estimate an annual loss of $607 million in wages and time for women who leave the work force. They also estimate an additional $108 million cost to taxpayers in increased public assistance for women no longer able to support themselves, having quit their jobs to care for their infants. Under current proposals for unpaid leave, the costs of continued health insurance benefits, administration, and the need to reallocate work or for temporary employees to fill vacancies are borne by the employer. Fuchs (1988), however, argues that "employer-provided benefits" is a misleading term, that the costs of proposals that rely solely on employer mandates are likely to result partly in higher prices for goods and services and partly in lower wages. They would have little effect on profits unless foreign competition prevented higher prices and resistance to wage reductions required increased investment. Because the utilization of infant child care services is not equal among workers across industries, the effects would be uneven. In apparel manufacturing, for example, mothers of young children constitute 15 percent of the work force, but in firms manufacturing durable goods they account for only 4 percent and in mining and construction less

254 WHO CARES FOR AMERICA'S CHILDREN? than 2 percent. The costs weigh more heavily on consumers of certain products and on employees in certain industries (Fuchs, 1988~. In another approach, researchers at the Yale Bush Center (Zigler and Frank, 1988) recommend 3 months of paid leave at 75 percent of salary to be supported through an insurance fund with employee and employer contributions, similar to the New Jersey disability plan (see Chapter 7~. Overall, it appears that the costs to employers of current unpaid leave proposals are modest. Expanding leave to cover small employers, longer leaves, or wage replacement, however, would necessitate considering various cost-sharing alternatives. Provider Subsidies Provider subsidies directly affect the ability of and the incentives for individuals to enter the child care business as well as the level, quality, and price of the services they offer. In contrast to consumer subsidies, which are directed primarily at increasing the affordability of child care and increasing parental choice, provider subsidies have more targeted objectives to increase the supply of care for special categories of children and families, and to improve the quality of available services. Head Start, for example, is designed primarily to help poor children prepare for school, and the Child Care Food Program is designed to promote better nutrition among poor children who are in child care. Provider subsidies direct resources to specified groups and increase public control over the characteristics of care that is offered, but they also limit parents' flexibility in choosing programs and arrangements. Universal Child Care Universal child care is the provider-subsidy equivalent of the full-scale child allowance. Child care would be available to every family, in a manner similar to the public school system, with some minimum level of quality guaranteed. Since high-quality child care costs, on average, between $3,000 and $5,000 per child, the gross cost of universal child care would be similar to a child allowance of the same amount available to all children regardless of income and employment status. However, unlike child allowances, not every family would make use of a universally available service. Many parents would continue to care for their own children or make private arrangements. In addition, the estimated gross costs would be reduced by savings from eliminating some current programs. The cost of providing universal child care could be met through sliding fees to parents and some combination of federal, state, and local funding.

THE CHILD CARE MARKET AND ALTERNATIVE POLICIES 255 A child allowance gives maximum support to parental choice, whereas universal child care limits choice and emphasizes quality, availability, and affordability. It would also encourage women's labor force participation as a means of economic self-sufficiency and reduce the stigma of programs for poor children. Such a policy would provide maximum support for worker stability and increase the available supply of labor for employers. Research suggests that comprehensive early childhood programs can have short-term as well as long-term benefits to children, families, and society (see Chapter 4~. They can also be designed for facilities in neighborhoods where-they are most needed. Another alternative is to consider only school-age children. The prob- lem of care for school-age children would be reduced if the school day were extended to more accurately match the typical work day and if vacation and summer holidays were similarly cut back; of course, teacher hours and salary costs would rise. As for universal child care proposals, the money would be used to provide care directly. Sentences for Children in Low-Income Families Several alternative policies have been proposed to improve services for children in low-income families. One option is to extend the current Head Start program to make it a full-day, full-year program. This would enable parents to work full time, if they don't already, or it would enable them to place their children in high-quality care throughout the day rather than having to coordinate arrangements that may differ in significant respects. Brush (1988) estimates that the added cost of extending Head Start to a full-day, full-year program would be 1.5 times the current expenditures or about $2.3 billion annually. Another option is to make Head Start available to all income-eligible children. Currently, the program serves fewer than 20 percent of the income-eligible 3- and 4-year-olds. How many children would attend Head Start if it were universally available is not known, but there is considerable evidence that the current low participation rate resects, in part, the fact that there are not enough spaces available to meet parental demand for the program. If Head Start were available to all income-eligible children, we estimate that it would cost $5.7 billion as a part-time program and $8.7 billion as a full-time program. There are estimates of how much this cost would be offset (e.g., by increased long-term work-force productivity, reduced welfare expenditures) for the parents of Head Start children. Such long-term effects are hard to assess, but for the Perry Preschool Project, begun in 1962, it is estimated that for every $5,800 spent on comprehensive child development services per child, there was a long- term savings of $28,000 (in constant 1981 dollars discounted at 3 percent

256 WHO CARES FOR AMERICA'S CHILDREN? annually) (Weikart, 1989~. Whether or not these savings would be realized on a national scale is not known, but they give some indication of the long-term benefits of high-quality care for young children. A variation of the proposal to expand Head Start is to expand an array of public prekindergarten programs for 4-year-olds. As discussed in Chapter 6, a number of states have launched such initiatives, and federal legislation to provide added subsidies is also pending. For example, one proposed federal bill would authorize $1 billion a year for a full-day school program, building on the Head Start and state compensatory education programs already in place. The program would be voluntary, with a funding formula based on the number of children in the state (or community) aged newborn to 5 years who live in families below the poverty line, the number who live in single-parent families, and the number in families with both parents in the labor force. This formula reflects a commitment to serve poor children from families with no employed parents as well as children from low-income working families. Similar specific programs could also be designed for school-age children and children with disabilities. A third and more comprehensive approach was presented in a major bipartisan child care initiative the Act for Better Child Care (ABC). Initially introduced in the 100th Congress, it would target approximately 75 percent of a $2.5 billion budget to subsidize child care programs for families at or below 115 percent of a state's median income. A 20 percent state match would be required. The legislation proposed a block grant approach, combined with income targeting and direct provider grants to increase the supply of child care. The bill also contained provisions for some consumer subsidies, such as sliding-fee scales and vouchers. Quality guidelines were mandated, and states were required to coordinate child care resources and services. Despite significant negotiation and compromise, the bill failed to pass in the first session of the 101st Congress; it is expected to be reconsidered in the second (1990) session. We believe that provider subsidies for services to low-income families and other categories of children who are underserved in the current market would probably raise the quantity and overall quality of care. Parental choice, however, would be less than if the same sums were provided through direct consumer subsidies. Increased self-sufficiency through increased maternal employment should also result in reduced welfare costs, although gains would be small if proposed expenditures are small. For all of the proposed provider subsidies, the most likely source of federal funding would be general revenues, with the financial burden spread across all taxpayers, although Watts and Donovan (1988) propose using projected surpluses from the Social Security trust fund. However, these funds are already being used to offset the general fund deficit, so that any use to pay for child care would be equivalent to using general revenues

THE CHILD CARE MARKET AND ALTERNATIVE POLICIES 257 and would increase the current budget deficit. Direct services could also be funded through the public schools by using the local tax base, although it can be argued that federal funding is essential since fiscal capacity varies considerably across communities. Infrastructure Subsidies A third type of subsidy supports the infrastructure of the child care system. Such policies do not provide financial benefits directly to families, nor do they finance the direct provision of services to children. Rather, they address the more general questions of quality and efficiency through increased training and wages for caregivers, expanded planning and coor- dination, improved standards and regulations, and extended resource and referral services. Related infrastructure supports include liability insurance pools and provider networks. By themselves, such policies may be less expensive than consumer or provider subsidies. Many of the increased costs are borne by providers and consumers (unless other subsidies are also available). Since there is currently no federal child care policy, and most initiatives are undertaken at the state and local levels, there are no accurate data from which to project the costs of investments in the child care infras- tructure. They would, however, directly affect the cost and quality tradeoff discussed earlier: they may improve quality and efficiency, but increase costs and reduce availability if implemented in the absence of additional subsidies. The primary goals of caregiver training and wage subsidies are to encourage individuals to become child care workers, to increase their skills, to increase their tenure, and, therefore, to improve the quality of child care that is provided in a variety of programs and settings. Although child care workers in centers generally have some formal child development training, a high proportion of workers in family day care homes have limited formal education and little or no formal child care training (Coelen et al., 1979; Fosburg, 1981; Kisker and Strain, 1988~. Increasing the supply of trained providers is likely to increase the wages of child care workers and, hence, increase the cost and affordability of care. Implementing a comprehensive policy for training child care workers would involve initial as well as recurring costs. There are a substantial number of current providers who would benefit from basic training in child care, as would new child care workers. A less intensive program of in- service training would benefit all workers in centers, schools, and family day care homes on an ongoing basis. Increased wages, however, are a significant and continuing cost (Clifford and Russell, 1989~. Only a few states have explored the possibility of providing subsidies directly earmarked to increase the wages of child care workers (see Chapter

258 WHO CARES FOR AMERICA'S CHILDREN? 7~. Massachusetts, for example, allocated supplemental funds to Head Start programs for salary enhancement through a grant program. To reinforce the intent that supplemental funds be used to increase staff salaries, the state established suggested hourly minimum rates for several positions. Initial findings of a study of the impact of this allocation reports reduced staff turnover and an increased ability to recruit qualified staff (Goodman et al., 1988), but the full impact of wage subsidies on the quality, availability, and affordability of care is unknown. Increasing staff salaries through wage subsidies in one segment of the child care market is likely to create competition for qualified staff, which may result in higher wages and higher fees for parents in programs not receiving wage subsidies. Programs unable to charge higher fees and therefore to provide higher wages may have to hire less qualified staff. The goal of subsidies for service planning and coordination is to improve the efficiency of the child care market. Planning and coordination efforts at the local and state levels focus on identifying needs and available program resources, coordinating programs, and allocating funding across myriad departments and jurisdictions. These planning efforts may bring together the public and private sectors in an effort to increase the supply of services, enhance the provision of resource and referral services, and more efficiently allocate funds (see Chapter 6~. Infrastructure support for the development of standards and the im- provement and enforcement of regulations is intended to increase the quality of care children receive. As detailed in Chapters 3 and 4, many current state regulations fail to reflect what research and best professional practice suggest is necessary to protect children's health and safety and to enhance their social, emotional, and cognitive development. Efforts to establish federal child care regulations have a long and beleaguered his- tory; efforts to encourage states to adopt more stringent regulations have been limited. Although there is convincing evidence that increasing the stringency of regulations and their enforcement improves the quality of care, there is widespread disagreement about whether national standards or regulations are feasible, whether the federal government can or should enforce child care regulations, how the states could be induced to adopt and implement more stringent regulations, and whether it is possible to effectively regulate family day care homes as well as child care centers. If agreement could be reached on national standards of quality for child care and the federal government then endorsed such standards for states to use as the basis for their regulations, the likely edect would be to reduce poor-quality care by establishing a minimum threshold for services. It would also raise the costs of care. To encourage states to act, the federal government would have to link any existing or new child care subsidies to states' incorporation of the specified standards in their regulatory system.

THE CHILD CARE MARKET AND ALTERNATIVE POLICIES 259 However, as several observers have noted, unless sufficient public funds are available both to help providers meet new regulations and to assist low-income families in purchasing care, the unfortunate side effect of more stringent regulations is likely to be a reduced supply of affordable child care services. Regulations that establish recommended levels for staff/child ratios, group size, and physical facilities could as much as double the current average costs of care and of caregiver training, putting them on a par with high-quality Head Start (Brush, 1988~. More stringent regulations will tend to discourage unlicensed providers from entering the regulated market and might encourage some currently licensed providers to go underground. One analysis reported by the Heritage Foundation (1988) found that regulations in the proposed House version of the ABC bill in the 100th Congress would result in the closing of roughly 20 percent of current child care centers, primarily those in the private sector, and would replace them with publicly funded child care centers. An alternative interpretation, however, is that the centers would raise their fees rather than close. If increased costs are not offset by public subsidies, many parents would be unable or unwilling to pay the increased costs of purchasing care, thus reducing their flexibility of choice and, perhaps, leading them to place their children in lower quality care. The administrative costs of regulations vary with the extent of en- forcement activity. Critics of stronger regulations point to the difficulties of widely varying parental views about what constitutes quality care and the inadequacy of current enforcement efforts. They also note that it is extremely difficult to effectively enforce services provided in family day care homes. An alternative approach has been to link standards for quality to the provision of technical assistance, resource and referral services, and provider subsidies for family day care providers (e.g., the Child Care Food Program benefits and Work Incentive Program tWIN] child care subsidies). If the financial incentives are substantial enough, and technical support Is available, there is evidence that providers are willing to comply with ap- plicable standards, thereby improving the quality of the services they offer (see Chapter 6~. Regulations and standards are similar to parental leave in that they can be mandated but are not necessarily funded by the government. If national guidelines were mandated without funding, the costs of meeting them would be borne by providers, who would presumably pass them along to consumers. Enforcement costs, however, would be borne by the government, that is, taxpayers. Resource and referral services can complement regulations and serve as an alternative mechanism to increase the quality of care by offering consumer education to parents and technical assistance and training to

260 WHO CARES FOR AMERICA'S CHILDREN? providers. Parents generally have limited knowledge of the child care op- tions available to them, and they choose largely on the basis of convenience or recommendations from friends (Kisker and Strain, 1988; Liebowitz et al., 1988~. Educating parents about the factors that affect quality and pro- viding information on available programs will enhance their ability to make informed decisions to select high-quality programs. Some proponents of resource and referral services claim that they will encourage parents to be- come effective monitors and so fill the gap between appropriate standards and enforcement of regulations. Resource and referral services can provide critically needed support and assistance to family day care providers, who tend to work in isolation, to lack efficient mechanisms for filling staff vacancies, and to have limited access to training and technical assistance (Kisker and Strain, 1988~. The costs of establishing and maintaining resource and referral services are relatively low in comparison with the costs of other mechanisms to enhance the quality of care. In addition to improving the quality of care through consumer education and provider training and assistance, resource and referral services are a source of valuable information about the supply of and demand for services that is essential for state and local planning and coordination. Although there are insufficient data to estimate the costs of specific programs, there are general estimates associated with various proposals to strengthen the child care infrastructure. For example, in the ABC proposal (cosponsored by Senators Christopher Dodd and Orrin Hatch in the 101st Congress), approximately 22 percent of the originally proposed $2.5 billion authorization was earmarked for investments in infrastructure, including state-level planning and coordination. CONCLUSIONS In the 1950s most child care was provided free and "off the books" by at-home mothers and relatives; since then it has increasingly been replaced by paid nonrelative care. During the 1950s and 196Os, a great deal was written about the economic value of the services that women were performing in the home. Now the great expense of replacing the quality and quantity of those services is becoming apparent. There is also increasing recognition that even in the 1950s child care was not free; it was paid for by women in lost wages and by society in lost tax revenues. Inhere is now much greater public awareness and discussion of the current child care market and alternative public policy responses to the perceived child care problem. Discussion about shortages must include three linked but distinct con- cerns: availability, affordability, and quality. The number of places available

THE CHILD CARE MARKET AND ALTERNATIVE POLICIES 261 is probably of most serious concern for children aged 0-2, for children from low-income and, especially, single-parent families, and for children with disabilities. But availability alone is of little import unless the places avail- able are affordable and of adequate quality. Most parents are paying a nontrivial proportion of their family income for care. But the burden is much heavier for low- and moderate-income parents, among whom the share of income for child care may approach the share of income for housing. And this burden is magnified for single parents, for parents of infants, and for parents of children with disabilities. Improving the quality of child care will inevitably raise the cost. Increasing wages to levels that would reduce the extremely high rate of staff turnover and implementing standards that would reduce the number of children per caregiver now allowed by some states would be very expensive. It is hard to see how low- and moderate-income families would be able to afford high-quality care without substantially more assistance from government or employers. This evidence demonstrates the very difficult tradeoffs in the child care market, among availability, affordability, and quality. Improving staff/child ratios is expensive, and raising staff salaries is even more expensive. Yet current salary levels and staff/child ratios are generally not adequate for the kind of child care that research and best professional practice suggests is safe and developmentally sound. Thus, intervention In the market such as stricter licensing and regulation may be desirable from the standpoint of improving quality, but it would be likely to aggravate the problem of affordability, and it might reduce availability. How should the United States find the optimal balance of cost and quality? One way is to ensure that parents have adequate resources, and allow each family to make its own decisions. Most American parents want to retain a high degree of independence from government in choosing employment and child care, and the panel agrees that parental choice should be a key feature of public policy. Current policies do in fact provide some subsidies for most families, and this approach could be expanded through tax credits and child allowances. It is possible, however, that sometimes the choices parents make for the care of their children do not meet the criteria necessary to achieve a safe, healthy, and developmentally sound environment. For example, should parents have the choice of placing an infant in a child care center in which one worker cares for six or more infants at a time (as eleven states allow), even though it is known that such stafI/child ratios are not good for children? The U.S. government already supports low-income families in improving the intellectual and social development of their children and strengthens parenting skills through programs such as Head Start. Indeed, one of the common political rationales for expanded child care programs is to improve the health and life chances of children at risk by building

262 WHO CARES FOR AMERICA'S CHILDREN? on the positive results of Head Start. The government also intervenes in family choices to their and the larger socie~cy's benefit through the public school system. It is not a large step to argue for more intervention in the child care market, at least on behalf of the youngest children, regardless of family income, with particular emphasis on those whose needs are greatest. Because of the limited research available, this panel cannot fully ex- plore the many policy alternatives available to address the child care needs we have identified. On the basis of our review of the available research and our evaluation of the current system and selected policy alternatives, however, we can make specific recommendations for research and policy. REFERENCES Bane, M.J., and P. Jargowsky 1988 Links between government policy and family structure: What matters and what doesn't. In ~ Cherlin, ea., We Changing American Family and Public Policy. Washington, D.C.: The Urban Institute Press. Blank, H., A. Wilkins, and M. Crawley 1987 State Child Care Fact Book 1987. Washington, D.C.: Children's Defense Fund. Blau, D.M., and P.K Robins 1986 Fertility, Employment and Child Care Costs: A Dynamic Analysis. Paper presented at the annual meeting of the Population Association of America, San Francisco. Bogat, G.~, and L" Gensheimer 1986 Discrepancies between the attitudes and actions of parents choosing day care. Child Care Quarterly 15~3~:159-169. Bradbard, M., R. Endsley, and G Readdick 1983 How and why parents select profit-making day care programs: A study of two southeastern college communities. Child Care Quarterly 12~2~:160-169. Brush, L" 1987 Usage of Different Kinds of Child Care: An Analysis of the SIPP Data Base. Paper prepared for the Social Services Policy Division, Planning and Evaluation, U.S. Department of Health and Human Services, Washington, D.C. 1988 Projecting the Costs of Full Day Child Care from the Costs of Head Start. Paper prepared for the Panel on Child Care Policy, Committee on Child Development Research and Public Policy, Commission on Behavioral and Social Sciences and Education, National Research Council, Washington, D.C. Bureau of the Census 1987 Who's Minding the Kids? Current Population Reports, Series P-70, No. 9. Washington, D.C.: U.S. Department of Commerce. 1988 Poshly in the United States 1986. Current Population Reports, Series P-60, No. 160. Washington, D.C.: U.S. Department of Commerce. Bureau of Labor Statistics 1988 Marital and Family Characteristics of the Labor Force: March. Unpublished data. Cain, V., and S. Hofferth 1987 Parental Choice of Self Care for School Age Children. Paper presented at annual meeting of the Population Association of America, Chicago.

THE CHILD CARE MARKET AND ALTERNATIVE POLICIES Cherlin, A., ed. 263 1988 Lee Changing American Family and Public Policy. Washington, D.C.: The Urban Institute Press. Clifford, R., and S. Russell 1989 Financing programs for preschool-aged children. Theory into Practice. 2~1, Winter):19-27. Coelen, C., F. Glantz, and D. Calore 1979 Day Care Centers in the US. A National Profile 197~1977. Cambridge, Mass.: Abt Associates. Connelly, R. 1988 Utilizing Market Child Care: An Economic Framework for Considering the Polipy Issues. Paper prepared for the Panel on Child Care Policy, Committee on Child Development Research and Public Poligy, Commission on Behavioral and Social Sciences and Education, National Research Council, Washington, D.C. Department of Economics, University of Vermont. Eck, A. 1984 New occupational separation data improve estimates of job replacement needs. Monthly Labor Review 107~3~:3-10. Espenshade, T., and J. Minarik 1987 Demographic implications of the 1986 U.S. tax reform. Population and Devel opment Review 13:115-127. Fink, D. 1988 A Quick Fight All the Way: A Report on the Need for Child Care Among Parents of School-Age Children with Handicapping Conditions. Working paper No. 178. Wellesley, Mass.: Wellesley College Center for Research on Women. Fosburg, S. 1981 Family Day Care in the United States. Summary of Findung>;. Cambridge, Mass.: Abt Associates. Frank, M. 1988 Costs, financing, and implementation mechanisms of parental leave policies. Pp. 315-325 in E. Zigler and M. Frank, eds., The Parental Leave Crisis: Toward a National Policy. New Haven, Conn.: Yale University Press. Fuchs, V. 1988 Women's Attest for Economic Equality. Cambridge, Mass.: Harvard University Press. Galinsly, E. 1988 The Impact of Child Care Problems on Parents on the Job and at Home. Paper presented at the Child Care Action Campaign Conference, Wingspread, Racine, Wisc. Bank Street College of Education, New York. Garfinkel, I. 1988 The Potential of Child Care to Reduce Poverty and Welfare Dependence. Unpublished paper. Institute for Research on Poverty, University of Wisconsin. General Accounting Office 1987 Parental Leave: Estimated Costs of H.R 925, The Family and Medical Leave Act of 1987. Washington, D.C.: U.S. Government Printing Office. 19 ~Parental Leave: Estimated Cost of Revised Parental and Medical Leave Act. Washington, D.C.: U.S. Government Printing Office. 1989 GAO's Cost Estimate of the Family and Medical Leave Act of 1989 (H.R 770). Washington D.C.: U.S. Government Printing Office.

264 WHO CARES FOR AMERICA'S CHILDREN? Goodman, I., J. Brady, and B. Desch 1988 A Committment to Quality: The Impact of State Supplemental Funds on Mas- sachusetts Head Start. Newton, Mass.: Education Development Center, InG Graham, M., and K. Scott 1988 The Fiscal Impact of Definitions of High Risk for Education of Infants and Toddlers. Unpublished paper, University of Miami. Grubb, W.N. 1988 Choices for Children: Policy Options for State Provision of Early Childhood Programs. Unpublished paper. School of Education, University of California, Berkeley. Hartmann, H., and D. Pearce 1989 High Skill and Low Pay. The Economics of Child Care Work. Washington, D.C.: Institute for Women's Policy Research. Haskins, R. 1988 What day care crisis? AEI Joumal on Govemmenz and Society Ref~lanon 2:13-21. Heritage Foundation 1988 The "ABC" child care bill: An attempt to bureaucratize motherhood. Issue Bulletin No. 145, October 6. Washington, D.C.: Heritage Foundation. Hofferth, S. 1987 Child Care in the U.S. Statement before the Select Committee on Children, Youth, and Families, July 1. The Urban Institute, Washington, D.C. 1988 The Current Child Care Debate in Context. Paper prepared for the 1987 annual meeting of the American Sociological Association (revised). The Urban Institute, Washington, D.C. Hofferth, S., and D. Phillips 1987 Child care in the United States, 1970-1995. Joumal of Mamage and the Family 49~3~:559-571. Kahn, A., and S.B. Kamerman 1987 Child Care: Facing the Hard Choices. Dover, Mass.: Auburn House. Kisker, E., and M. Strain 19~ Child Care Markets: A Brief Look at the Markets for Child Care in Camden and Newark, New Jersey. Notes for presentation at annual meeting of the Association for Public Policy Analysis and Management. Mathematica Policy Research, Inc., Princeton, N.J. Kisker, E.E., R. Maynard, ~ Gordon, and M. Shain 1989 The Child Care Challenge: What Parents Need and What Is Available zn Three Metropolitan Areas. Princeton, N.J.: Mathematica Policy Research. Leibowitz, A., L" Waite, and C. Witsberger 1988 Child care for preschoolers: Differences by child's age. Demography 25~23:205- 220. Lombardi, J. 1988 Child Care Workers. The Hidden Subsidy in the Child Care Delivery System. Paper prepared for the Panel on Child Care Policy, Committee on Child Development Research and Public Policy, Commission on Behavioral and Social Sciences and Education, National Research Council, Washington, D.C. Child Care Employee Project, Alexandria, Va. Marr, M. 1988 The Child Care Crzszs: Are Tax Credits the Answer? An Arzalysis of Seven Child Care Tax Credit Bills. Washington, D.C.: Citizens for Tax Justice.

THE CHILD CARE MARKET AND ALTERNATIVE POLICIES 265 McGroder, S.M. 1988 A Synthesis of Research on Child Care Institution Patterns. Paper prepared for the Panel on Child Care Policy. Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Health and Human Services. O'Connell, M., and C.C. Rogers 1983 Child Care Arrangements of Working Mothers: Rune 1982. U.S. Bureau of the Census, Current Population Reports, Series P-23, No. 129. Washington, D.C.: U.S. Department of Commerce. Powell, D., and J. Eisenstadt 1980 Finding Child Care: A Study of Parents' Search Processes. Detroit, Mich.: The Merrill-Palmer Institute. Presser, H. 1988 Shift work and child care among dual earner American parents. Journal of Mamage and the Family 50:133-148. Presser, H., and W. Baldwin 1980 Child care use and constraints in the United States. In ~ Horberg, ea., Women and the World of Work. New York: Plenum Press. Presser, H.B., and V.S. Cain 1983 Shift work among dual-earner couples with children. Science 219:876-879. Reismon, B., A. Moore, and K Fitzgerald 1988 Child Care: The Bottom Line An Economic and Child Care Policy Paper. New York: The Child Care Action Campaign. Robins, P. 1988 Federal support for child care: Current policies and a proposed new system. Focus 11(2):1-9. Scott, KG. 1988 The Fiscal Impact of Definitions of High Risk for Education of Infants and Toddlers. Unpublished paper. University of Miami. Sonnenstein, FL., and D.^ Wolf 1988 Caring for the Children of Welfare Mothers. Paper presented at the annual meeting of the Population Association of America, New Orleans, April 21-23. Spalter-Roth, R., and H. Hartmann 1988 Unnecessary Losses: Costs to Americans of the Lack of Family and Medical Leave. Washington, D.C.: Institute for Women's Policy Research. Stephan, S., and S. Schillmoeller 1987 Child Day Care: Selected Federal Programs. Congressional Research Service, Library of Congress, Division of Education and Public Welfare. Washington, D.C.: U.S. Government Printing Office. U.S. Department of Labor 1988 Child Care: A Workforce Issue. Report of the Secretary's Task Force. Washington, D.C.: U.S. Department of Labor. Waite, C, ~ Leibowitz, and C. Witsberger 1988 What Parents Pay for Child Care and Child Care Costs. Unpublished paper. Rand Corporation, Santa Monica, Calif. Watts, H., and S. Donovan 1988 What Can Child Care Do for Human Capital? Paper presented at the Child Care Action Campaign, Wingspread, Racine, Disc. Department of Economics, Columbia University. Weikart, D.P. 1989 Quality Preschool Programs: A Long-Term Social Investment. Occasional Paper 5. New York: lhe Ford Foundation.

266 WHO CARES FOR AMERICA'S CHILDREN? Whitebook, M., D. Phillips, and ~ Howes 1989 Mono Cares? Child Care Teachers and the Quality of Care in America. Executive Summary, National Child Care Staffing Study. Oakland, Calif.: Child Care Employee Project. Zigler, E., and M. Frank, eds. 1988 The Parental Leave Crisis: Toward a National Policy. New Haven, Conn.: Yale University Press.