Rethinking U.S. Child Care Policy
Demand for high-quality care will increase only when consumers have better information about child care and stronger economic incentives to purchase excellent care.
Child care in the United States is, by many standards, in poor shape. Commonly heard complaints include that today's system of child care endangers the well-being of children, causes financial hardship and stress for families, makes it next-to-impossible for low-income families to work their way off welfare, causes substantial productivity losses to employers, and prevents many mothers from maintaining productive careers in the labor force.
Because child care is a service that is bought and sold in markets, economics can provide a useful framework for thinking about its problems. As with any commodity, supply, demand, cost, price, and quality are key elements of market analysis. But with child care, quality plays a special role, because this characteristic may affect the development of the children who receive the care. Extensive research during the past 25 years documents a positive association between measures of child care quality and the social, emotional, and cognitive development of children. Although most of this research stops short of proving that the association is causal (there is ample reason to expect that children who would have developed well anyway will probably be placed in higher-quality care environments), a number of well-structured studies do support a causal relationship. These studies also show that the benefits of high-quality care are probably larger for children who are at risk of developmental delays as a result of living in poverty.
Analyzing the source of the quality problem in child care suggests a remedy. However, putting this remedy into action will require a fundamental reorientation of public policy, leading in turn to major shifts in the behavior of care providers and in the actions of parents.
Although there are no national data regarding quality, an intensive study by Suzanne Helburn of the University of Colorado of 400 day care centers in four states documented what many child care experts have been saying for a long time: "Child care at most centers in the United States is poor to mediocre, with almost half of the infant and toddler rooms having poor quality. Only one in seven centers provides a level of quality that promotes healthy development." Analysis of data from this and other studies reveals that child care is of low quality, on average, because of the unwillingness of many consumers to pay a high enough price to cover the cost of high-quality care, even though the cost to providers of improving the quality of care would be moderate. Higher-income children, on average, receive child care of roughly the same quality as lower-income children, indicating that the quality of child care is not a high-priority item for most consumers.
Even when high-quality child care is available, it is expensive in comparison to the income of low-income families. Families in poverty who pay for child care spend more than one-quarter of their income on that cost. Thus, high-quality care is very likely beyond the reach of most low-income families unless subsidized by the government. On the other hand, the large majority of U.S. families are not poor. The average family that pays for child care spends only 7 percent of its income on child care. These families could afford higher-quality care as easily as they could afford a nicer car. Research shows that day care centers can and do improve quality when the price of care rises. But centers have relatively little financial incentive to offer high-quality care, because consumers, on average, are not willing to pay much more for better care. To put it simply, the problem is not on the supply side, but on the demand side.
Workers who provide child care receive low wages, which leads to a high turnover rate among care providers. This instability contributes to low quality of care in a number of ways; for example, fewer workers hold their jobs long enough to build a good base of experience or to take part in extended training and education. And secure attachment of children to adults, an important element in child development, is more difficult when turnover is high. Low wages are due, in large measure, to the apparent willingness of many women to work as care providers for low wages. In day care centers, full-time teachers (almost all of whom are women) earn, on average, less than half the amount earned by other women of the same education and age. And family day care providers earn substantially less than do full-time teachers in day care centers. One-third of all teachers leave their jobs each year, a rate that is about three times higher than the average for all women. Naturally, these women would prefer higher earnings. But the fact that they are willing to supply their labor for such low monetary rewards, despite the fact that higher-wage jobs are available in other sectors of the economy, suggests that there are nonmonetary rewards to being a child care provider. Such rewards often include being able to care for their own children while working.
Consumers lack knowledge about important aspects of child care. Many parents cannot tell the difference between low-quality and high-quality care if they see it, and most of them do not really "see" the care provided anyway, because they typically drop off their children and head to work. Comparing the quality of child care arrangements as rated by parents and by trained observers shows that parents systematically overrate the quality of care, by a rather large amount. Survey evidence also suggests that the average parent does not visit many providers before selecting one. These are important clues about the source of the apparent unwillingness of many parents to pay a price high enough to cover the cost of high-quality child care.
In many respects, the child care market functions much better than is commonly believed. Shortages of care facilities are uncommon, and when they do exist, they are often limited to small segments of the market. Power is not concentrated in the hands of a few providers who are able to extract excess profits from consumers by restricting supply. Child care workers have low wages because they are willing to work for low wages, not because they are exploited by center owners or forced to subsidize consumers.
Yet, there is a fundamental failure in the child care market. This failure is not due to some defect in the internal workings of the market. Rather, it is caused by an externality: Market participants (parents) either do not bear all of the costs of their child care decisions or they make these decisions without understanding their consequences, or both. In general, the remedy for a market failure caused by an externality or an information problem does not lie in regulation or public takeover of the market. Instead, the remedy lies in providing appropriate information to market participants so they can make well-informed decisions and in finding a way to internalize the externality by giving the participants an incentive to seek high-quality care.
Thus, demand for excellent child care will not increase unless consumers have sound information about the quality of care as well as stronger incentives to purchase better care. The most important determinant of child development and well-being is no doubt the quality of parenting, not the quality of child care. But child care quality can be an important factor, particularly for low-income children, and it is probably more susceptible to change through public policy than is the quality of parenting. However, today's child care policies are badly flawed.
For one thing, the majority of child care subsidy funds are available only to employed parents, but it is not obvious what sort of problem in the child care market is related to employment. Subsidies that require employment increase the demand for care but do not increase the quality of care demanded. Tax-based subsidies that are available to middle- and upper-income families--the Dependent Care Tax Credit and the Exclusion of Employer-Provided Dependent Care Expenses--place no restrictions on the quality of care obtained, so it cannot be argued that they have the goal of improving quality. The subsidies encourage employment of both parents in two-parent families and of the single parent in one-parent families, but it is not immediately clear why society should wish to provide such encouragement.
The case of employment-related child care subsidies for low-income families may seem easier to rationalize. The goal of such subsidies is to help families achieve and maintain economic self-sufficiency as an alternative to dependence on welfare. For many low-wage parents today, employment is not very rewarding financially, especially when cash transfers from the government are reduced dollar for dollar as earnings increase. The cost of child care and other work-related expenses can make the net financial reward from employment so low as to make welfare a more attractive alternative.
However, low employee wages typically are a result of low skills, and child care subsidies do not directly address this connection. These subsidies do make employment more attractive, and if skills improve through on-the-job training and experience gained by being employed, then the subsidies would indirectly address the problem of low skills and help families escape poverty and welfare dependence in the long run. But there is no evidence that the typical low-wage job provides the training and experience that lead to improved skills, and this means employees are not likely to receive significantly higher wages as they continue to work. In this case, child care subsidies must be continued indefinitely in order to make employment attractive, and the goal of economic independence is not achieved.
Another problem is that one of the primary current means by which government provides child care subsidies to low-income families, the Child Care and Development Fund, which resulted from welfare reform efforts of 1996, places few restrictions on the quality of child care that can be purchased with the subsidies, and it places no emphasis on improving the development of low-income children.
Clearly, low-income families face an employment problem: Their low labor market skills mean that even full-time employment frequently will not be enough to lift them out of poverty. But this is not a child care problem, and it cannot be solved by child care subsidies. Such subsidies do not result in improved labor market skills, and as a result they simply substitute in-kind transfers for cash transfers under welfare. Education and training are the most appropriate policies for improving the labor market skills of low-income workers.
Principles for public policy
The following principles reflect research findings about the child care market as well as judgments about the goals that child care policy should try to achieve:
Child care policy should be neutral with respect to employment. There are no compelling economic or moral reasons for society to encourage employment of both parents in a two-parent middle-class family. There may be a more compelling case for encouraging single parents to achieve economic independence through employment. But a child care subsidy is at best an indirect approach and at worst an ineffective approach to accomplishing this goal. Indeed, employment-related child care subsidies are likely to have the unfortunate side effect of increasing the amount of low-quality child care experienced by children from low-income families. Instead of subsidizing the employment of parents, government should, if anything, subsidize the costs of raising children, without favoring market costs for child care over the foregone earnings cost of a parent who stays home to care for a child.
Child care policy should provide information to parents about the benefits of high-quality child care, about how to discern the quality of care, and about how to find high-quality care. As in all markets, an informed consumer drives improved performance of service providers.
Child care policy should provide incentives for parents to choose high-quality care. Even if parents are generally aware of the developmental benefits of high-quality care, they may not value those benefits much compared to other things they can buy. For example, parents may feel that their own influence on the development of their children can make up for the effects of low-quality care, or that the developmental outcomes measured by standard assessments are less important than, say, religious values, respect for authority, and other intangible attributes. If consumers are given sufficient incentives to choose high-quality care, then providers will have an incentive to offer such care.
Child care policy should encourage the development of programs to help providers learn how to improve the quality of care. An essential feature of a competitive market is that firms can prosper only by offering the services consumers are willing to pay for. Thus, direct subsidies to providers should not be necessary. Providers will have an incentive to increase quality in response to consumer demand, but they may lack the knowledge to upgrade quality.
Child care policy should be progressive, with benefits being larger for children in poor families. Because children in poor families are at greater risk of developmental delays and the problems that result from such delays, the benefits of high-quality child care are therefore likely to be larger for them. Equity considerations also favor a progressive child care policy.
Child care policy should be based on incentives, not regulations. Regulating an industry such as child care, with its hundreds of thousands of providers, is likely to be either very costly or ineffective. Evidence suggests that current regulations imposed by the states are not very effective at improving the quality of care being provided. Of course, states should not be discouraged from regulating basic safety and health aspects of child care. But with federal policy, financial incentives can be more flexible than regulations, and well-designed incentives can be self-enforcing rather than requiring a monitoring bureaucracy.
Child care policy should presume that well-informed parents will make good choices about the care of their children. Government can provide the best available information to inform parental decisionmaking, along with incentives for parents to make good choices. But government should not limit the freedom of parents to arrange care for their children as they see fit (again, subject to regulations regarding neglect and abuse). Not all parents will want to take advantage of subsidized care in preschools and family day care homes, no matter how high the quality of care provided. Some parents will prefer care by a relative or close friend; some will prefer care in a church-based setting that emphasizes religion; and some will prefer care by a babysitter in the child's home. Although these choices may not be optimal in fostering child development, government should not coerce parents to raise children in a particular way. Parents should remain the decisionmakers.
A proposal for reform
Many previous reform proposals contain creative and useful ideas. The following proposal, which rests on the guiding principles outlined above, borrows liberally from previous proposals and adds some new ideas:
- Provide a means-tested child allowance. Each family would receive from the federal government a cash allowance for up to two children, from birth through age 17. The allowance could take the form of a refundable tax credit, requiring that a family file a tax return to claim the allowance. Refundability means that even a family with no tax liability is eligible for the credit, so it is of value to low-income families (unlike most child allowances in the current tax code). The value of the allowance should decline as the level of family income rises, with the allowance phased out entirely for high-income families. There would be no restrictions on the use of the allowance. The money could be used to pay for child care, food, housing, medical care, or other items that directly benefit children. But it could just as easily be used for other purposes--for example, to subsidize nonemployment by one of the parents, enabling the parent to stay home to care for his or her children. The rationale for subsidizing child-rearing costs is twofold: On moral grounds, children should be taken care of; and on efficiency grounds, children are future workers and citizens, and society has an interest in their healthy development.
- Subsidize the cost of accreditation to care providers. The National Association for the Education of Young Children, as well as several other organizations, will for a fee assess the operations of day care centers and preschools that want to have their services accredited. These organizations should be subsidized so that they can carry out their assessments at no cost to care providers. In addition, accrediting groups should develop a standardized system that offers several levels of accreditation, so that providers unable to qualify for the highest level of accreditation could nevertheless be certified as providing certain levels of care. In this system, providers would be either (1) accredited as offering care of excellent quality; (2) accredited as offering care of good quality; or (3) unaccredited, meaning that they satisfy state regulatory standards but do not reach the higher level of performance required for a higher rating. Participation by providers would be voluntary.
- Inform all new parents of the benefits of high-quality care, as well as how to recognize and find excellent care. The simplest way to accomplish this would be to give a booklet and video with such information to mothers when they are in the hospital to give birth. The materials should describe, within the limits of scientific evidence, the consequences for child development of high- and low-quality care, and they should illustrate in vivid terms what a high-quality child care arrangement is like, in contrast with a low-quality arrangement. The materials also should describe the accreditation system for care providers (emphasizing that accreditation is certified by independent agencies), and they should provide information on how to contact resource and referral agencies and other sources of information about the local child care market.
- Provide a means-tested care voucher, for up to two children per family, with a value that depends on the quality of the care provider at which it is redeemed. Vouchers would be worth more if used at an accredited provider. For example, a low-income family might receive a voucher that covers 30 percent of the average cost of unaccredited child care, 60 percent of the average cost of good-quality care, and 100 percent of the average cost of care accredited as of excellent quality. This differential gives families an incentive to seek high-quality care, and thus it also gives providers an incentive to offer high-quality care in order to attract consumers with greater purchasing power. The value of the vouchers would decline as family income rises, with the value dropping to zero for high-income families. The voucher would be of no value if a family does not purchase child care or pays a relative for child care. This condition is unavoidable if the system is to best provide incentives for use of high-quality care. Because the vouchers would not require employment, they would encourage use of high-quality care by both employed and nonemployed parents to enhance child development. Families that did not use their voucher would still receive the cash benefits provided under the child allowance part of the system.
Balancing benefits and costs
This new system would replace all current federal child care subsidies, as well as all tax deductions and credits for children, including the income tax exemption for children and the child tax credit. Some programs that provide child care for low-income children but are explicitly development-oriented and have no employment requirement, such as Head Start and Title IA of the Elementary and Secondary Education Act, could be integrated into the system. The system also would replace the Temporary Assistance for Needy Families (TANF) program, which provides cash assistance to low-income families with children but which includes employment requirements and time limits.
Because the proposed system is neutral with respect to employment, it would not replace programs that are explicitly intended to reward employment, such as the Earned Income Tax Credit and job training and education programs. If society considers it desirable for low-income single mothers to be employed, then the voucher part of the system provides considerable resources that these mothers could use for child care. If further encouragement of employment is desired, the government would have to provide resources from another source.
The cost of the new system will depend on the value of the child allowance and the child care voucher. To illustrate, I have calculated the cost based on several assumptions. I assumed that the cash allowance would be set at $5,000 per low-income child, with the allowance being reduced as annual family income rises, and phased out at $64,000. The child care voucher would be worth $6,000 for a low-income preschool age child in the highest-quality child care, with the value of the voucher decreasing with family income, child age, and child care quality. (I also made some assumptions about how many families will redeem the vouchers.) In this scenario, the total cost of the system is $208 billion per year. However, after accounting for savings due to the elimination of several current federal programs, the net annual cost is about $95 billion. This is obviously a very large sum, but it is a realistic estimate of the cost of a rather sweeping solution to the child care problem.
Are the benefits from such a radical revamping of child care policy as large as the costs? We simply lack the information needed to quantify the long-run benefits of improvement in the quality of child care on such a large scale to answer this important question. Some encouragement is offered by the Perry Preschool Study, a small-scale randomized study of very intensive preschool and other social services for low-income children. The study found that the long-run benefits to the government, in the form of lower spending on crime, special education, and welfare, exceeded the amount of government support provided for the program. This conservative approach to evaluation does not take into account the benefits to the children themselves in the form of enhanced education and earnings. The fact that the program actually saved the government money suggests that high-quality preschools may have significant benefits to society as well as to the participants. However, it is not clear whether it was the high-quality preschool itself that made such a big difference or the other social services provided, such as weekly home visits. Nor is it clear whether the benefits would be as large for children who are not as deprived as the children who were enrolled in the study.
Certainly, much more well-designed research on the benefits of high-quality child care is called for. But such research will take many years to produce results. In the meantime, policymakers need advice. Based on economic analysis, the best course for today is to proceed with implementing the new system, because the risk of suboptimal development for millions of children under the current system is not worth running.
W. Steven Barnett, "New Wine in Old Bottles: Increasing Coherence in Early Childhood Care and Education Policy," Early Childhood Research Quarterly 8, no. 4 (1993): 519–558.
Barbara Bergmann, Saving Our Children from Poverty: What the United States Can Learn From France (New York: Russell Sage Foundation, 1996).
Tricia Gladden and Christopher Taber, "Wage Progression Among Less Skilled Workers," in Finding Jobs: Work and Welfare Reform, eds. David Card and Rebecca Blank (New York: Russell Sage Foundation, 2000).
Suzanne W. Helburn, Cost, Quality and Child Outcomes in Child Care Centers, Technical Report (Denver, Colo.: Department of Economics, University of Colorado at Denver, 1995).
Lynn A. Karoly, Peter W. Greenwood, Susan S. Everingham, Jill Houbé, M. Rebecca Kilburn, C. Peter Rydell, Matthew Sanders, and James Chiesa, "Investing In Our Children: What We Know and Don't Know About the Costs and Benefits of Early Childhood Interventions" (Santa Monica, Calif.: RAND Report MR-898-TCWF, 1998).
Michael E. Lamb, "Nonparental Child Care: Context, Quality, Correlates, and Consequences," in Child Psychology in Practice, eds. I. Sigel and K. Renninger, Handbook of Child Psychology, fifth ed., W. Damon, series ed. (New York: Wiley, 1998).
John M. Love, Peter Z. Schochet, and Alicia L. Meckstroth, Are They in Any Real Danger? What Research Does--And Doesn't--Tell Us About Child Care Quality and Children's Well-Being (Princeton, N.J.: Mathematica Policy Research, May, 1996).
James Walker, "Funding Child Rearing: Child Allowance and Parental Leave," The Future of Children 6, no. 2 (Summer/Fall 1996): 122–136.
Edward Zigler and Matia Finn-Stevenson, Schools of the Twenty First Century: Linking Child Care and Education (Boulder, Colo.: Westview Press, 1999).
David Blau (email@example.com) is professor of economics and fellow of the Carolina Population Center at the University of North Carolina at Chapel Hill. He is the author of The Child Care Problem: An Economic Analysis (Russell Sage Foundation, 2001).