238    Child Gare in Britain, or, How Do You Restructure Nothing?
Van der Eyken.W. 1987. The DHSS Undet-Bves Initiative 1983-1987': Final Report. London:
DHSS. Wilding Paul  1997. "The Welfare State and the Conservatives." Political Studies 45: 716-2 6.
CHAPTER 10
More Can Be Less:
Child Care and Welfare Reform in the United States
Denise Uiias Levy and Sonja Michel
In the United States today, child care provision is divided into two distinct sectors, public and private. Such an arrangement is not, of course, surprising in a welfare state regime that has been categorized as "liberal" (Esping-Andersen 1990). Nor should it be surprising to discover that this dual system has produced deep inequities in the quality, accessibility, and affordability of child care services. What is perhaps unexpected is that, despite the overall inadequacy of the system, the rate of full-time labor-force participation among American women, including mothers, is currently one of the highest in the Organization for Economic Cooperation and Development (OECD) countries.1 This can be explained by a congeries of factors, including the demand for labor produced by the general upsurge in the U.S. economy (a trend that may now be reversing), relatively well-enforced antidiscrimination legislation in the area of employment, and, since 1996, a shift in public assistance policy "from welfare to workfare" that has pushed thousands of low-skilled women into the labor force. The fact that the majority of U.S. mothers, including those with very young children, are employed outside the home should not, however, be taken as an indication that, despite its flaws, the American child care system "works." Rather, it suggests that wage-earning parents at all income levels have learned to "make do"—to cope with the everyday stresses and long-term consequences of inadequate provisions, but at a significant toll on the quality of their private lives and the lives of their children (Hochschild 1997).
In both the public and private sectors, several different types of services may be available, ranging from in-home care to formal child care centers,2 but the methods of payment differ. In the private sector, middle- and upper-income parents choose and pay for services directly, while in the public sector, poor and low-income parents must find child care centers or family day care providers that will accept state-issued vouchers to be reimbursed at fixed rates, or they may, in some instances, receive reimbursement for individual arrangements with kith or kin. While it appears that the private sector is wholly self-supporting, in fact it, too, is subsidized indirectly by the federal government through a dependent care tax credit and incentives to employers who establish child care services. Perhaps the more salient differ-
240    More Can Be Less
ence is that in the private sector, the state supports the independent, freely made decisions of parents to place their children in child care in order to seek paid employment or pursue educational or vocational goals, while in the public sector, the state subsidizes child care as part of a mandatory system of job training and employment (workfare) for parents seeking public assistance. In this instance, child care is used as a lever of a compulsory public policy.
This chapter will first briefly review the history of child care policy that preceded recent developments, highlighting public and private developments and the relationship between early childhood education and child care. The chapter goes on to analyze the politics of the linkage between child care and workfare and the ongoing opposition to universal policy. In the final sections, we examine in detail the deployment of contemporary public policy within the context of welfare reform, focusing on its impact on poor and low-income mothers and children and the implications for universal provisions.
American Child Care: A Very Brief History
The current arrangement of child care in the U.S. is the outcome of a long and vexed history of social provision characterized, as Sonya Michel has argued elsewhere (1999), by a strong man-as-breadwinner orientation (Lewis 1992), and both fueled by and constitutive of the policy logic typical of a public/private, liberal, or "residual" welfare state regime. Throughout its history, child care—like much of American social policy—has also reflected deep racial and class divisions. Beginning with the charitable day nurseries of the nineteenth century, child care in the U.S. has almost always been targeted toward the poor as a means of "helping them help themselves." Although the federal government took responsibility for providing services during periods of national crisis—first for the children of the unemployed during the Great Depression and then for children of defense workers during World War II—these policies did not lead to permanent public, universal provisions. For several decades after the war, the issue of state-sponsored child care lay more or less dormant. It was not until the 1960s and 1970s that it once again appeared on the public agenda, this time in conjunction with efforts to reform public assistance. In a series of amendments to the Social Security Act, the U.S. Congress foreshadowed the linkage between public provisions and mandatory employment that would become law in 1996.
The lack of public child care notwithstanding, the postwar decades witnessed a significant rise in employment among mothers, including those with preschool children, which in turn prompted the growth of market-based child care services. This trend was aided by several federal measures, including the child care tax deduction passed in 19S4 (and converted to a child care tax credit in 1972), as well as a variety of incentives to employers to set up or sponsor services for their employees, beginning in 1962. Market-based services were of several types: while all charged fees,
Denise Urias Levy and Sonya Michel    241
some were commercial enterprises, while others were run by voluntary or nonprofit organizations (often denominational), many of which transformed existing part-day nursery schools into full-day child care centers in response to community demand.
The place of nursery schools in this history bears further explanation. Nursery schools, along with kindergartens, followed their own trajectory in the United States, one that was (in contrast to some of the other cases discussed in this volume) quite distinct from that of child care—and deliberately so (for a full discussion, see Michel 1999, chaps. 1—3). As part of the early childhood education/child development movement, nursery school and kindergarten innovators, advocates, and practitioners were among the first groups of American women who self-consciously sought to professionalize, and as such did not want their movement to become tainted with the odor of charity and poverty that clung (literally, in some cases) to the day nurseries. With their impetus, early childhood education forged ahead in the early twentieth century, leaving child care behind. As kindergartens were incorporated into the public school system and nursery schools (mostly fee-based) became the darlings of the middle class, day nurseries and child care centers gained a reputation for being "custodial warehouses" that only the poor would use as a last resort.
Early childhood educators believed that children of all classes should have the benefits of nursery schools, and in the 1930s they convinced New Deal policymakers to create a network of federally supported Emergency Nursery Schools for the children of the unemployed; these were, however, dismantled when the Depression was over. It was not until the mid-1960s that reformers were finally able to establish a permanent public program of early education for the poor: Head Start. The most successful and enduring of President Lyndon Baines Johnson's Great Society programs, Head Start has repeatedly demonstrated its educational value, but because of the way it was conceived and funded, it has not, until quite recently, been considered part of the public child care system. As we shall see below, the fact that it had its own funding stream has, no doubt, helped preserve Head Start, but this has also created problems of coordination with other public programs for young children.
Linking Child Care and Workfare
The 1960s saw several initiatives to reform public child care, all of which grew out of political concerns about the sharp rise in—and changing racial composition of— the welfare rolls. Because of deeply entrenched practices of racial segregation and discrimination throughout the country but especially in the south, the federal public assistance program Aid to Families with Dependent Children (AFDC) had, since its establishment in the 1930s, served primarily white families—a profile that hardly reflected the racial composition of America's poor (Bell 196S, pt. 1). But starting in the late 1950s, impoverished African Americans, as a result of a series of court cases brought by the strengthening civil rights movement, began to gain access to the as-
242    More Can Be Less
sistance to which they were entitled by federal law (the following discussion is drawn from Michel 1999, 243-47; see also Mink 1998, chap. 2, and Kornbluh 2000, pt. 2). Between 1960 and 1967, the proportion of AFDC recipients had shifted from 86 percent white to 46 percent nonwhite.
AFDC had initially been based on the principle that the state should support mothers whose families lacked a male breadwinner so that they could remain at home caring for their children—a principle that fit well with America's general adherence to the "male-breadwinner" logic. As long as the majority of AFDC recipients were white, the policy proceeded without challenge.3 But as the racial balance shifted, many politicians, especially those from the south, bridled at the idea that African-American women—who had "always" worked—were now going to receive public support to remain "idle." Not coincidentally, a new congressional consensus began to form around the idea that mothers should now be encouraged to become "self-sufficient"4 and that child care would be a critical element of any policy designed to free mothers to pursue work or job training.
Congressional initiatives to link child care and welfare first produced the Social Security Amendments of 1962, which provided grants-in-aid to state welfare agencies for the development and support of licensed child care agencies (the following discussion is drawn from Levy 2000, chap. 3). Notably, these measures were intended to "encourage" welfare recipients to become self-sufficient, but they did not make work mandatory. Not so the 1967 amendments that created the Work Incentives program (WIN), granting child care funding to states that expelled welfare recipients who rejected job offers—including mothers, regardless of the age of their children. Despite the harsh tone of this legislation, Congress soon became aware that there were nowhere near enough child care facilities to serve every mother who was supposed to work—regardless of how much funding was made available. This fact emboldened liberal opposition to requiring mothers of young children to work or enroll in job training and led to a de facto abandonment of the program.
In 1969 a new approach to poverty emerged from the administration of Richard Nixon: the Family Assistance Plan (FAP) (Levy 2000, 18-19; Michel 1999, 249-51; Kornbluh 2000, chap. 3). This bill was intended to replace WIN with a guaranteed annual income and significantly expand job training and child care facilities. Simultaneously, congressional Democrats proposed a series of measures, including the Comprehensive Child Care Development Act of 1971, designed to increase and improve child care provisions for all American families. Nixon, while clearly favoring child care for the poor, opposed the liberals' universal proposal and vetoed the bill when it reached his desk. In the meantime, the FAP foundered in Congress because of disagreements over the principle of a guaranteed income and the amount of funding it would entail.
For the next two decades, the administration's bifurcated stance cast a long shadow over child care initiatives. Child care legislative initiatives were muted for most of the 1970s and suffered in the 1980s from growing disagreement about the role of government in family life. These disagreements prevented passage of com -
Denise Urias Levy and Sonya Michel    243
prehensive or universal child care legislation and continued to restrict federal support for child care to policies that served as work-incentive instruments within the context of welfare reform. Even those, however, fell short of their potential because state and local bureaucrats failed to implement them by drawing down federal funds and organizing adequate services (Michel 1999, 251-59).
The link between child care and welfare reform was further reinforced in 1988, when President Ronald Reagan signed into law the Family Support Act (FSA).This measure rephrased the principles underlying public assistance by articulating the view that welfare should be based upon a "social contract" that emphasized the mutual obligation and responsibility of both poor parents and the government. Under this contract, parents would be primarily financially responsible for their children, but the government would assist by granting benefits (cash subsidies, health insurance) to low-income, two-parent families while channeling the adults into the Job Opportunities and Basic Skills (JOBS) program.
The JOBS program reinstated the principles underpinning the ill-fated WIN program but attempted to make them work by emphasizing education and training and expanding support services like child care to facilitate participation. It also extended WIN's reach to new groups of recipients by mandating the participation of mothers with children aged three to five and of all teenage mothers, irrespective of the age of their children, who had not completed high school and had no work experience (these two groups had previously been exempted from most work mandates). On paper, the new program implied significant new commitments to provide child care to AFDC recipients who pursued education or training designed to improve their employability.
Two programs were created to fulfill these commitments: AFDC-Child Care and Transitional Child Care (TCC).The first required states to guarantee child care to any AFDC recipient who participated in state-approved education or training activities or who accepted or retained employment. The second required states to guarantee child care for up to twelve months to families who had received AFDC and its child care benefits in the past but were no longer eligible due to increased hours of employment or earnings. Although there was evidence at the time that these commitments could not be met within the existing structure of child care services, legislators supported the FSA as an alternative to a more comprehensive piece of child care legislation then pending in Congress, the Act for Better Child Care (ABC, discussed below). Indeed, later studies showed that both programs in many ways failed to live up to the commitments made in the law.
By the mid-1990s, only about 13 to 15 percent of the eligible children of adult enrollees in JOBS were being served—although three-fourths of the programs claimed to be providing child care subsidies or helping to arrange child care for all or almost all participants who needed assistance (U.S. General Accounting Office 1995a and 1995b). This was in part due to parents' difficulties in finding child care providers who were accessible (given the shortage of public transportation) and available during nontraditional hours of work, or flexible enough to accommodate
244    More Can Be Less
part-time JOBS participation hours (most child care centers operate—and charge
fees__on the assumption that slots will be filled full-time). The shortage of facilities
led to the creation of long waiting lists which, in some states, included thousands of children (U.S. General Accounting Office 1995b). Also, because funds were scarce, states in practice often exempted welfare recipients from work requirements or limited participation in training programs to parents with school-aged children or those who could easily find free child care.
The Opposition to Universal Child Care
The problem of child care in the late 1980s was not, however, limited to the welfare population. By this time, the proportion of mothers with children under fifteen participating in the labor force had grown to 60 percent. In 1988, more than 10.S million children under age six, including nearly 6.6 million infants and toddlers under age three, had wage-earning mothers. Yet the supply of child care remained small and the range of options limited, with cost a constraint affecting both factors. For low-income two-earner families who were not eligible for public subsidies, child care for just one child could consume more than a quarter of their total household income, 43 percent if they needed full-time infant care. This in turn limited parental options. While 31 percent of mothers of higher socioeconomic status were likely to enroll their children in a child care center or preschool, only 18 percent of low-income parents turned to such services. Within this latter group, relatives (other than one of the parents) were the most common source of care, which meant that a large percentage of children were in unlicensed home settings, often under inadequate conditions of safety, and with limited educational and developmental opportunities (Hayes, Palmer, and Zaslow 1990, 159-62; Michel 1999, 259-64).
Since the late 1960s, child care advocates had been trying to make a case for comprehensive child care legislation as the adequate response to these trends, and to convince policymakers that the care and education of all young children—not just the poor—should become a public responsibility. They attempted to explain that the existing system was inadequate to meet the needs of all parents, including those who could afford to pay more for child care. Some Congressional leaders were responsive to these concerns,5 and many governors began to take an interest in child care issues, including then governor Bill Clinton of Arkansas, who was chairman of the National Governors' Association (Hofferth 1993).
Despite the visibility of and interest in child care issues in the late 1980s, proposals for comprehensive legislation could not gain the necessary traction in Congress. For example, in 1987, supporters introduced the ABC bill; this was debated well into 1988, even after the FSA was passed, and although it had the endorsement of more than 135 national groups, it met with fierce opposition from legislators, who remained divided about creating a new federal program—especially after President George H. W Bush threatened to veto any child care law envisioning such an action.
Denise Unas Levy and Sonjra Michel    245
If passed, the ABC bill would have dramatically shifted the political rationale for public child care by defining eligibility for subsidies on the basis of earnings rather than enrollment in welfare, using the state median income as a cutoff with a sliding fee scale. In addition, it would have established a federal funding stream for child care while preserving parental choice regarding the type of provision and offering a diverse delivery system using grants, contracts, and vouchers; and maintaining Social Services Block Grant funding for child care. The bill also mandated federal child care standards and required each state to develop its own child care plan, coordinate resources, establish an interagency committee of all bodies concerned with child care, and strengthen basic health and safety protections.
At the root of much opposition to the bill were political differences about the proper approach to funding child care programs. Democrats wanted to create federal grants to states to subsidize low-income families while improving child care for all families, while Republicans favored a combination of tax credits and vouchers, with benefits being targeted at low-income families only. In addition, some Republicans preferred policies that encouraged (nonwelfare) women to stay home with their children, at least while the children were very young. Finally, Democrats wanted to create national standards for child care, while Republicans argued that standards should remain a state responsibility.6
After two more years of introducing and debating dozens of competing child care bills without finding common ground, legislators were able to settle on a compromise child care package that retained many aspects of the original ABC bill but accommodated Republican demands for devolution to the states. The winning legislation, finally passed in 1990, had four major components: in addition to expanding funding for Head Start, it established a new entitlement program that expanded child care assistance under Title IV-A for families "at-risk" of falling into welfare dependency (At-Risk Child Care, or ARCC); expanded the Earned Income Tax Credit to low-income families (to alleviate the burden of their child care costs); and initiated a new program called the Child Care and Development Block Grant (CCDBG).
CCDBG allocated states the funds to create child care for eligible children, subject to a sliding scale based on family income.7 Although all children under thirteen in families with incomes below 75 percent of the state's median income were eligible for services, they were not guaranteed. Indeed, funds were relatively modest. While the original ABC bill had envisaged resources on the order of $2.5 billion per year, Congress authorized only $750 million for 1991. Funding rose to $925 million in 1993, but the scope of the program remained limited. In 1991, only 571,095 children were being served, whereas an estimated ten million or more children under age thirteen were living below the federal poverty level (Levy 2000, 27), and millions more at the upper end of the sliding scale might have benefited from child care services.
The legislative process that produced CCDBG underscored two important lessons about U.S. politics. First, it showed that attempts to create comprehensive federal programs for child care outside the scope of welfare reform would lead to
246    More Can Be Less
profound and irreconcilable controversies among the two parties; but second, it revealed that a block grant approach could provide a framework to accommodate this conflict. By giving states the responsibility for making difficult policy decisions, with outcomes that could be thought of as best reflecting local preferences, the act still allowed for an incremental move toward universal provisions.8
Child Care and Welfare Reform
While the 1990 legislation temporarily slowed debates over child care and welfare, it fully addressed neither issue. Thus both returned to the congressional agenda in 1996, with child care presented as part of the larger effort to reform public assistance. In passing the Personal Responsibility andWork Opportunity Reconciliation Act (PRWORA) in 1996, the United States initiated a major restructuring of its principal program of social assistance for lone mothers and low-income parents, Aid to Families with Dependent Children (AFDC). Instead of supporting such families to care for their own children at home, as AFDC had done, the new program requires a majority to seek paid employment within a specified length of time and sets a lifetime limit for receiving public assistance.9 Following the trend toward "federalism" or devolution to the states initiated by President Reagan and exemplified in measures like the CCDBG, the measure provided funding through two major block grants to the states, one for temporary (cash) assistance to needy families (TANF), the other for child care (the Child Care and Development Fund, or CCDF).
In crafting the legislation, Congress was at first reluctant to confront the unprecedented demand for child care that would inevitably be generated by PRWORA's employment mandates. When "workfare" mandates become fully operational in 2002, an estimated one million children (in addition to the 3.3 million low-income children already being served under previous laws) will require subsidized services (Long and Clark 1997, 5 and 10, table 3; U.S. Bureau of the Census 1998). But after President Clinton twice vetoed the bill for lacking adequate funding for child care, Congress added $4 billion over five years to the child care block grant. Despite the additional funding, however, the new child care system has proven to be less than workable. There are a number of reasons for this: in addition to a lack of sufficient funding, there is wide variation in the bureaucratic practices that determine access to child care at the local level; inadequate compensation and benefits for child care providers, coupled with minimal support for improving program quality; and difficulties in coordinating child care with other public programs for early childhood education. We will examine each of these in turn.
Funding
Even with the additional spending authorized by Congress, there is not enough money in the system to care for all the children in low-income families who are by law eligible for, if not entitled to, subsidized child care. According to the law, all chil-
Oenise Urias Levy and Sonya Michel    247
dren under age thirteen in families with incomes below 85 percent of the state's median income are eligible for services. Currently, states can offer services to only a fraction of those who are eligible. According to one recent estimate, although current federal expenditures have climbed to nearly $12 billion, this amount is only about half of what is needed to meet the full demand (Democratic Staff 2001).
Under the terms of PRWORA, several preexisting sources of federal funding for child care for poor and low-income families were consolidated into one block grant to individual states—the CCDF. Although the CCDF potentially makes available more child care funding than all previous federal sources combined (the $2.97 billion appropriated for this fund in 1997 represented an increase of $600 million, or 27 percent, over the total federal child care funding for the previous year),10 it is designed in such a way that a significant portion of the funds are controlled by state-level politics. Funds are divided into three categories: mandatory, matching, and discretionary, with $1.2 billion apportioned among the states based on their previous federal (Title IV-A) funding (see n. 3), $0.77 billion based on states maintaining their previous Title IV-A matches (the so-called maintenance-of-effort or MOE requirement), and an annually appropriated amount of discretionary spending ($ 1 billion for 1997), which is distributed according to states' economic and demographic features.11
In such a structure, state-level decision making becomes the key determinant of how much child care will be available and how it will be distributed. A number of factors—all of which vary from one state to another—come into play. For example, two analysts of child care policy argue that these decisions depend on level of need (which is related to child poverty), fiscal capacity (ability to tax; per capita income), and fiscal effort (willingness to devote resources to a service such as child care)—in other words, political climate (Douglas and Flores 1998). Denise Urias Levy (2000) points to a cluster of "policy levers" used with great flexibility by states that determine the distribution of provisions among poor and low-income families and also affect the shape and content of public child care overall.
A full understanding of all the state or subnational political contexts that determine child care funding decisions requires discussion of a large body of empirical research that is beyond the scope of this chapter, since each one is unique (for detailed analysis, see Levy 2000, chap. S). We can, however, offer a few generalizations. A state's political disposition enters the formula at two points, one historical, one current. This means that to a certain extent a state's current funding status may be delimited or "contained" by its previous funding history. Levy (2000) has found that, to some extent, current levels reflect historical trends regarding social spending: states that have tended to be liberal (in the North American sense) in the past, such as those of the northeast, midwest, and far west, and those of a more conservative bent, such as those of the southeast, have (with a few notable exceptions) generally both remained true to type on this issue.12
Yet, prior history is not all-determining, and current conditions can still have telling effects. For example, most states, by refusing to spend their own money for
248    More Can Be Less
child care, stand to lose a significant proportion of their potential federal funding as well, thus reducing their total budget for child care.13 At the same time, every state stands to gain by maximizing its matching funds, with an average increase of SO percent (Long and Clark 1997, 3). In many states, however, this would require increasing its own child care funding by 70 percent or more—something their funding histories suggest they may well be reluctant to do. In 1994, for example, twenty states fell short of allocating sufficient matching funds to draw down their maximum Title-IVA funding (Long and Clark 1997, 9—10, table 2), while twenty-one allocated enough to draw down the full amount (Adams and Poersch 1996).
Whatever its political predilections, a state's current fiscal capacity and the level of child care need also come into play. As Toby Douglas and Kimura Flores (1998) point out, "many of the states with higher rates of child poverty tended to have low fiscal capacities" (2). Thus states like Mississippi, Louisiana, New Mexico, and West Virginia, which rank 50th, 49th, 48th, and 47th, respectively, in terms of child poverty, also rank low in terms of per capita personal income (SOth, 40th, 47th, and 49th). Three of these states also rank low or medium-low in terms of "fiscal effort" (willingness to tax): Louisiana (44th), Mississippi (29th), and West Virginia (28th). In sharp contrast, New Mexico ranks Sth on this scale (reflecting the political strength of its prominent Latino/a contingent). To take a somewhat different example, New York state, which is 40th in child poverty, ranks very high (4th) in terms of per capita income and first in fiscal effort (Douglas and Flores 1998, 8, table 2). Taken together, these three factors can either reinforce one another (as in the cases of Louisiana, Mississippi, and West Virginia) or offset one another (as in New Mexico and New York), but if fiscal capacity is too low, as in New Mexico, the effect of even high fiscal effort is minimized.
An accounting of total allocations, however, tells only part of the story. As Levy (2000, chap. 4) has discovered, states can, by manipulating certain policy levers, fine-tune the amount and quality of child care available, and how and to whom it is distributed; indirectly, these three levers can affect parental choice and access to quality child care. The levers themselves comprise income eligibility limits, parent copayments, and maximum payment rates to providers. According to the law, families earning incomes up to 8S percent of the state median income (SMI), whose heads are working or preparing to work, are eligible for child care assistance under the CCDF.1+States may lower the income ceiling, thus reducing the number of families to be served. About half of the states have adopted a level equal to or lower than 65 percent of the SMI (in twenty-one, it falls between 40 and 59 percent). Of the remainder, about half set it between 65 and 80 percent, and only eight at the maximum rate of 85 percent (Levy 2000, 43). In calculating these ceilings, many states sought rough equivalency with the federal poverty level for a family of three, which was generally considerably lower than the SMIs.15
States' discretion to "manipulate" this policy lever and adjust it to their own needs and preferences has also been shown to be greater with the 1996 reform. The great autonomy granted by the reform led to a broader spectrum of choices. Of
Denise Urias Levy and Sonya Michel    249
particular interest is the variety of approaches taken by states to deal with the insertion of very low-income families into the new program. Moreover, there has been an increased dispersion in eligibility limits established by states as well as a general decline of such limits in several states (Levy 2000, 46—51).
With regard to copayments, the law states that each family is supposed to contribute to the cost of child care services, although this requirement may be waived for families at or below the federal poverty level, or under special circumstances. All but eight states have some sort of waiver in place, but regulations vary widely. Thus, in three states, a single parent with a child in center-based care, earning $12,000 a year, would pay nothing, while in six others, she would pay more than $100 per month (Levy 2000, 52—56).16 Here, too, there is considerable variation in the approaches taken by states when they designed their child care copayment policies, with a clear indication that copayment levels are used as a lever to limit the number of families served by the system. The higher the level of copayments, the more money available for child care and thus the wider the distribution of services over the eligible population. If the level of copayments is too high, however, eligible families will be unable to meet it and instead will seek alternative—and often lower-quality—forms of care.
The third policy lever, maximum payment rates to providers, affects the supply, accessibility to needy families to subsidized child care, and quality of services. High rates are likely to help in making more centers—and centers of better quality— available to children. However, given budgetary constraints, they may limit the number of children served, since the cost per child is likely to be higher. Under previous public programs, states were required to set payment rates at 75 percent of the local market rates for services, based on biannual surveys. The current law grants states flexibility to establish whatever payment rates they consider "fair" and conducive to equal access, and, until 1998, did not require regular surveys (Levy 2000, 64). In 1998 more than half the states failed to meet the 75th percentile standard as calculated for 1996 (Greenberg, Lombardi, and Schumacher 2000, 7). While the regulations now mandate regular surveys, they set no specific level for minimum (or maximum) payments; rates can thus range from less than $300 per month in the lowest five states to more than $600 in the highest five. Presumably, higher rates of compensation will promote the creation of centers and ensure their quality, but, given budgetary constraints, they may end up limiting the number of children being served, since the cost per child will be higher. By the same token, however, lower payment rates allow states to stretch child care dollars over a larger number of children, but they may also lower the supply of child care by deterring existing centers from opening up slots to subsidized children and discouraging individuals from becoming home-based providers. In addition, they depress staff salaries and benefits, leading to high turnover and lower quality overall.
States have another policy lever in the area of work requirements; that is, they can ease their own child care burden as well as that of TANF families by offering more latitude to parents with newborns and infants. Infant care is not only scarce
250    More Can Be less
but costly (because adult to child ratios are supposed to be quite high—normally no more than two infants to one adult), and some states have reasoned that it is cheaper to exempt the parents of newborns from mandatory employment or training rather than pay for infant care (providing, in effect, a form of paid maternity or parental leave, which is otherwise not mandated by law in the United States). Federal law allows states to provide such exemptions until a child reaches the age of one, but here again, states vary, with some interpreting long-term exemptions as a form of encouragement to poor families to have additional children.17Thus, although a few states offer the maximum exemption for each child, others allow only the minimum—thirteen weeks or three months—and some also set a lifetime limit of twelve months for the parent (Waller 1997, 6).
A final area of leverage affects the eligibility of families at different stages in their relationship to public assistance. As noted above, current law combines four previous programs—CCDBG, AFDC-Child Care, At-Risk Child Care, and Transitional Child Care—into a single block grant—the Child Care and Development Fund. Three of those programs guaranteed services to all families that fit specific criteria—AFDC-CC, for parents who were receiving aid who were working or participating in training or education; TCC, which provided services for 12 months to those no longer eligible for AFDC; and ARCC, for low-income parents who were "at-risk" of requiring public assistance. Moreover, funding for these programs was open-ended; that is, while states were required to put up matching funds in order to draw down federal dollars, there was no maximum for how much they could access. Under PRWORA, these separate funding streams have disappeared and the programs' constituencies have become absorbed into the larger group of those who are eligible for benefits under federal law but not entitled under state regulations. Each of the three groups must, in effect, compete for a share of limited child care funds, and while each presents a compelling case, one or more groups is inevitably granted lower priority. Most often, that group is the welfare "leavers."
States have the option of providing twelve months of transitional child care and shifting unused TANF funds from direct public assistance to child care as welfare rolls fall. About three-fourths of the states have done so (Schumacher and Green-berg 1999, 4), but recent studies of former recipients have found that in many states, fewer than half of those responding were receiving child care subsidies. The reasons they gave included lack of information about eligibility or a feeling that it was "too much trouble to apply" (Schumacher and Greenberg 1999, ii). While a majority of those without subsidies turned to relatives or neighbors, this type of care is not always reliable, making it difficult for former recipients to retain their hard-won employment.
Although PRWORA directs states to coordinate services for all categories of low-income families under one "lead agency," the practice of prioritizing one group over another can often create discontinuities in child care as families exhaust their TANF benefits and/or begin earning wages that exceed state-set income maxima for publicly supported child care. The irrationality of such arrangements becomes evident if
Denise Urias Levy and Sonya Michel    251
we look at an individual case. In 1997, Christine Ferguson, aWal-Mart cashier in Union Township, Ohio, found that her earnings of $6.80 an hour put her above the state's income maximum (which was set at 12S percent of the federal poverty level). When her county ran out of funds for child care, Ferguson and 109 other low-income parents were summarily cut off, and her child care costs jumped from $65 to $400 a month—an amount she simply could not afford. Reluctantly pulling her child out of care, Ferguson deplored her plight: "I'm really glad [President] Clinton wants to do this welfare reform—I think it's time. But you're going to send someone back to welfare if you take their child care" (quoted in Waller 1997, 1).
Ferguson's situation was the result of her state's interpretation of federal welfare requirements. Unlike most of its neighbors, Ohio had decided not to reallocate TANF funds to child care for low-income families as the number of TANF recipients fell, but instead lowered matching funds in order to reduce taxes.18Thus, Ferguson's county did not have enough funds to provide child care for all of its low-income, non-TANF families. In some states, former recipients are compelled to go back on TANF in order to receive child care benefits, thus keeping their "clocks" running and coming ever-closer to the five-year limit for benefits—precisely the course Ferguson was reluctant to take.
Bureaucratic Impediments and Local Conditions
Whatever the funding levels and priorities for their allocation, access to subsidized child care is ultimately determined at the local level, where widely varying bureaucratic practices can determine who receives services and who does not. Under previous "workfare" measures, states had an "affirmative responsibility" to provide services to all eligible families, and when these were not available, work requirements were waived for the families affected. Under PRWORA, this relationship is somewhat different; lone parents must still receive an exemption from work requirements if child care is not available for any child under the age of one, but states are no longer mandated to provide services (for the politics of this, see Levy 2000, 28—39). In order to receive an exemption, recipients must demonstrate that they cannot obtain child care "for one or more of the following reasons: a) unavailability of appropriate child care within a reasonable distance from the individual's home or work site; b) unavailability or unsuitability of informal child care by a relative or under other arrangements; c) unavailability of appropriate and affordable formal child care arrangements" (Greenberg 1998, 2-3). Although states themselves can be penalized for refusing to grant exemptions in these instances (risking a loss of up to S percent of their CCDF block grants), this rule does not serve as an effective deterrent, since those who administer the rules arbitrarily or improperly are not directly affected—while their clients are.
The attitudes of local officials and their willingness to extend themselves on behalf of their clients can determine access to child care and clients' prospects for success under the mandatory work program. Take, for example, Illinois. From a "macro" perspective, the state appears to be strongly committed to providing child care for
252    More Can Be Less
its poor and low-income residents. It drew down 100 percent of its federal funds in 1994 (for a total child care budget of over $ 149 million), and from 1997 to 2000 its governors have consistently requested and won significant increases in state appropriations for child care (12 percent for 2000) on the basis of projected growth in the number of children requiring services (IllinoisWelfare News 2000a, 6).19 In turn, these triggered increased allocations from the federal government. The present governor has also consistently shifted TANF monies from employment and training services to child care (Schumacher et al. 2001, 35), suggesting that, at least in the state of Illinois, welfare reform is moving in the direction intended by PRWORA's architects, with funding following need as recipients move from welfare to work.
If we look beyond these state-level indicators, however, we find anecdotal evidence to suggest that child care funds and services may not be reaching needy families in a timely fashion, thus preventing TANF applicants and recipients from complying with work requirements. Further, clients are being sanctioned—unfairly—as a result. Welfare rights advocates have found repeated instances in which caseworkers summarily deny benefits to such individuals, even though this practice is specifically prohibited by federal law. In Illinois, caseworkers are required by state welfare policy to refer all TANF applicants and recipients to child care referral agencies and arrange for payments to be made as soon as an appropriate slot has been located. But instead, according to IllinoisWelfare News (2000b), caseworkers are, in effect, telling clients, "Getting someone to watch your kids so you can go to work is your problem and not mine" (6). While this is technically true under federal statutes, its outcomes defy the spirit of PRWORA. To take just two examples:
A Bloomington, Illinois, TANF applicant was unable to attend a job interview at a fast food employer on the day she applied for TANF because, as she informed her intake worker, she had no one to watch her three small children during the job interview. The intake worker deemed this a refusal to comply with job search requirements and denied her benefit application.
After telling her caseworker that she could not find anyone to care for her special-needs child, a Chicago-area TANF applicant never got a referral to a child care resource and referral agency for help in finding child care. (Illinois Welfare News 2000b)
Thus what we might call negative administrative orientation prevents what appears—on the surface, at least—to be adequate child care funding from reaching its intended recipients.20 While it is difficult to know precisely how widespread such practices are, variations in child care availability, coupled with other regulations, the vicissitudes of the job market, and the complexities of the lives of the poor, would suggest that they occur not infrequently (for a vivid example, see Bernstein 2001).
In other states, different problems crop up as the result of local variants in regulations. In North Carolina, for example, state regulations prohibit reimbursement
Denise Urias Levy and Sonya Michel    253
for child care provided by "kith and kin." For rural African-American mothers, however, this is often the only kind of care that is available or readily accessible, given a lack of public transportation. Moreover, given the racial dynamics of their state, it is the only kind they feel they can trust. Many of these women thus try to defer employment until their children reach school age, and then they limit themselves to part-time jobs that will allow them to be home when their children reach school age (Sinclair 2001; Uttal 1998).
Inadequate Compensation
To make matters worse, several features of the new system have actually reduced the amount of care available for children from poor and low-income families. For instance, TANF's work requirements have weakened the informal networks that low-income women, particularly those in urban neighborhoods, typically depended upon for care not only for their children but for elders, the chronically ill, and other dependents needing regular assistance (Oliker 2000).21 Such networks comprised kin, friends, and neighbors who were similarly situated economically and more or less available to provide services for one another.22 Some of these women relied on public assistance, while others held jobs that did not pay enough for them to afford formal services. Before TANF, they could turn to their networks to arrange care for those for whom they were responsible, either on a regular basis, if they took employment outside the home, or occasionally, in order to accomplish other tasks (such as dealing with the welfare bureaucracy). TANF, however, has had an ambiguous impact on such practices. Though federal regulations permit child care payments to be made to kith and kin, TANF's mandatory work regulations have removed many of the women who had previously been available to offer care, leaving tattered support networks in their wake (Waller 1997, 1).
A second feature of TANF that has reduced the capacity of the existing system to care for poor and low-income children is the level and manner in which providers are reimbursed. As noted above, states may set their own rates for reimbursement and these are often below market rates; moreover, many states are slow to pay, forcing providers who accept vouchers or certificates in exchange for services to wait months for payments. This places a particular hardship on small independent and individual family providers, many of whom are already on the margins financially and can ill afford to offer places at a discount or extend credit to the state. Since providers are not mandated by law to reserve slots for poor and low-income children, many have simply refused to or stopped doing so, with the result that in some locales the number of subsidized slots for poor and low-income children has simply declined (Children's Defense Fund 2001).
A third factor feeds into the paradox that, despite an apparent abundance of child care funding, the supply is not keeping pace: this is due, in part, to low salaries and poor benefits typically offered to providers. The Children's Defense Fund has documented that, year after year, child care workers are paid less than amusement park attendants and garbage collectors, while other studies have shown that child
254    More Can Be Less
care providers are less likely than other workers to receive needed benefits such as health care and pensions (Gallagher and Clifford 2000). For example, in 1997 in Champaign County, Illinois, the average annual salary of a lead teacher in a child care center was $ 13,770—only $440 above the federal poverty level for a family of three, and approximately half the average salary for all female workers in the United States with "some college education"—$26,747 (University of Illinois Extension 2001, 2). As a result, in the recent climate of near-full employment, child care centers have found it increasingly difficult to attract and retain qualified employees, and women who might become independent home-based providers choose other occupations. Under CCDBG, states were required to set aside 25 percent of their funding for "activities to improve the quality of child care and to provide . . . early childhood development services"; current law reduces this to 4 percent. Using these funds, however limited, and sometimes augmenting them with state money, at least eleven states have initiated programs to encourage child care workers to obtain more training and then reward them for doing so (see, e.g. T.E.A.C.H. Early Childhood Illinois, n.d. [2001]; for an overview, Blank and Poersch 1999),23 while one state, Rhode Island, subsidizes benefits for child care workers. Such measures are, however, few and far between; they do little to raise the status of child care as an occupation or address the attendant issues of working conditions and quality.
Some welfare reformers have suggested that child care would be ideal occupation forTANF recipients seeking employment, and indeed, PRWORA singles it out for special mention (Greenberg 1996). Accordingly, TANF-related jobs programs also promote this type of employment (Little 1999), but such training does not adequately prepare recipients for working in centers (many of which require at least an associate's degree in early childhood education from a community college), or give them the wherewithal to overcome the many practical obstacles to starting a home-based service, such as lack of suitable housing, insurance and equipment costs, and so on. Moreover, child care analysts warn that not all women are, simply by gender, qualified to become child care providers. As one put it, "No public policies at the federal or state level should push or require people to care for children if they do not want to be providers" (Galinsky et al. 1994, 6). Thus the welfare-to-work program cannot, in itself, produce enough child care workers or providers to meet the increased needTANF itself has generated.
Quality and Coordination
The supply of child care is, of course, closely linked to its quality, and vice versa. When the supply is low and/or the variety of settings limited, parents may be compelled to settle for provisions with which they do not feel comfortable, that do not suit their child rearing preferences, or leave them feeling uneasy for any number of reasons. When, in addition, parents lack purchasing power, as doTANF recipients restricted by low rates of state reimbursement, their choices become even more limited. As noted above, prior to 1996, the quality of child care varied widely, with poor and low-income children more likely to be placed in low-quality settings.
Denise Urias Levy and Sonya Michel    255
More recent studies confirm that this is still the case, particularly for former welfare recipients who no longer receive child care subsidies (Fuller and Kagan 2000; Schumacher and Greenberg 1999).
Again, the paucity of federal funds for improving quality is partly responsible for this situation, but other factors feed into it as well. For example, child care facilities receiving TANF funds may have difficulty tapping into other funding streams or coordinating their services with those offered through other federal or state programs. The most obvious example of this problem is Head Start. Many low-income children participate in Head Start, the federally funded program of "compensatory" preschool education that had been in place since the mid-1960s. Beginning in 1990, appropriations for this program more than tripled, from $1.235 million in 1989 to $3.981 million in 1997, and enrollments nearly doubled, from 451,000 in 1989 to 794,000 in 1997 (U.S. House of Representatives 1999, 395, table 637). Explicitly designed to emphasize cognitive, social, and emotional development, Head Start programs are generally high in quality according to prevailing early childhood educational standards. But they usually run for only part of the day and thus cannot fully meet the child care needs of parents who work full-time or on irregular or night shifts. Moreover, income ceilings for Head Start families are extremely low—below what a parent might earn working full-time in a minimum-wage job. Thus, although both funding and enrollment in Head Start have increased since the implementation of TANF, the program is not ideally situated to address the new child care needs created by this policy.
Child care centers seeking to access Head Start funding or expose their charges to the benefits Head Start has to offer must not only meet cumbersome requirements but may also find that some of their families exceed income eligibility limits. In an effort to take advantage of Head Start for those of their children who do qualify, some centers transport them to Head Start classrooms for part of the day and offer "wrap-around services" during the remaining hours. Such plans are, however, awkward for staff and disruptive for the children involved. Centers attempting to coordinate their services with state-initiated prekindergarten programs for children of low-income families encounter similar problems. As of 1998—99, forty-two states had such initiatives, with a total of $ 1.7 billion in funding, but these served only about 725,000 children—approximately one-third of those eligible (Blank, Schulman, and Ewen 1999).
A final quality issue has to do with the lack of federal standards for federally funded child care. Long a bone of contention among child care advocates and their opponents (see Michel 1999, chap. 7), this issue predictably raised the hackles of proponents of devolution and never made it into the final legislation. Moreover, because policymakers' primary goal was moving welfare recipients into employment, they were not concerned when quality issues fell by the wayside. While licensed child care providers and centers must comply with state regulations, these vary widely, and weak enforcement often renders them virtually meaningless (Blank and Poersch 1999).
256    More Can Be Less
Conclusion
Although more public funding is available than ever before, the gap between the supply of child care and demand for it is, if anything, widening as the number of TANF applicants and recipients continues to grow, and rates of employment among moderate to high-income mothers remain steady or even climb (Children's Defense Fund 2001).24The obvious conclusion to be drawn is that while more funding is necessary, it will not be sufficient to carry out the task of meeting these new needs (though less funding would obviously make it even more difficult). States must be willing to loosen regulations, rationalize the organization of child care and facilitate access to it, and spend more of their own money in order to draw down maximum federal funds. While this last decision depends specifically on states' own fiscal capacity, all of these moves require a generally favorable and constructive attitude toward poor and low-income citizens. State-level decision making often founders on the paradox that the very political forces that are intent on reducing welfare rolls (primarily conservative Republicans25) are also opposed to both expanding government services and encouraging maternal employment. It is thus difficult to mobilize political support and funding for the very programs that are needed to make work-fare succeed even in its own terms.
At worst, the supply of child care will continue to fall short of demand, causing TANF recipients to renege on their employment and training responsibilities. Under pressure from state bureaucracies and politicians to meet quotas, caseworkers will deny or terminate benefits rather than place blame where it belongs—on the child care crisis—and join clients in calling for appropriate responses. Only slightly better will be a kind of "gray" solution in which parents and children seeking child care circulate in a maelstrom of irregular, ad-hoc provisions, settling for arrangements that may be inadequate or shifting children around in an attempt to find better provisions and maximize child care dollars, while providers themselves cycle in and out of business, quitting when they can no longer afford to subsidize the state with their own labor and capital and/or when they find more lucrative jobs. Both scenarios are exacerbated by the fact that states are more or less on their own in setting criteria and regulations for most aspects of public child care, and that parents' employment status, rather than children's needs for developmental and educational services, have become the paramount factors in setting policy.
To move beyond either scenario, the federal government must itself be willing (or must give states incentives) to improve the quality and supply of child care and the working conditions of those who provide it. Obvious measures might include offering or earmarking funds for start-up costs for new child care centers; setting minimum salaries and funding mandated benefits for workers; and creating packages of liability insurance and benefits for independent providers, coupled with much more stringent regulations and oversight for such provisions.26 States might make it more feasible for TANF recipients and other low-income individuals who
Denise Urias Levy and Sonya Michel    257
want to go into child care (and this option should by no means be imposed on them) to receive training by providing tuition money, adequate living stipends, and benefits (including child care, of course) while they are preparing themselves.27 In addition to sufficient funding, all of these measures require political predisposition and administrative capacity. While administrative capacity is probably best (and perhaps only) built up at the subnational (state) and local level, federal incentives, standards and regulations will undoubtedly be required to overcome local impediments and provide the necessary momentum when political predisposition is lacking.28
There is, of course, always the danger that such measures will tranform something that has usually been considered a social good—a boon to both parents and children—into an instrument of punitive control over women. In the feminist view, child care is an essential element of social citizenship; along with paid parental leaves and other policies allowing "time to care," it should be part of any social structure that allows women to choose freely whether to work or pursue education or care for family members. But feminists also believe that child care should never be used to compel women to work.29 One is reminded of Jane Addams's comment, made in 190S, that the day nursery (as child care centers were called then) "is a 'double-edged implement' for doing good, which may also do a little harm. . . ." (quoted in Michel 1999, 72). Indeed, today's public child care system may be used to do more than a little harm to poor mothers, but in the current political climate, there seems to be little hope of turning back policymakers' commitment to work-fare, at least not for the time being. Perhaps the best that feminists can hope for— and certainly what they should work toward—is a greatly improved system of public child care, one that provides poor and low-income families with convenient, affordable, high-quality care of the types they prefer. Only with such a system in place will TANF applicants and other women who must work outside the home be able to avoid making a Solomonic decision between leaving their children in questionable circumstances in order to comply with work requirements or staying at home to care for their children and sacrificing the wages and benefits they need to support them. And perhaps with the addition of federal standards and an extended sliding scale, such a system could come to constitute the basis for a universal system of provision in the United States. But this may be too optimistic.
Notes
1. In 1999, the U.S. rate was 60 percent, compared to 59 percent in Canada, 48.9 percent in Japan, and 58.S percent in Sweden (van der Lippe and van Dijk 2001).
2. In theory, consumers of both public and private provision have "free choice" as to the type of child care they access for their children, but in fact, in both sectors, parents are constrained by what is available and what they can afford; poor and low-income parents generally have the fewest "choices."
3. As Jennifer Mittelstadt (2001) has recently shown, the idea of using public policy to
258     More Can Be Less
help poor women become self-sufficient actually had its roots in legislation passed in 19S6, but implementation did not really get underway until the 1960s.
4. For a cogent critique of this term, see Young (forthcoming).
S.These included the Children's Caucus, led by Senator Christopher Dodd (D-Con-necticut) and the Select Committee on Children, Youth and Families, led by Representative George A. Miller (D-California).
6. Another controversy arose around the issue of funding child care that included religious instruction. Republicans opposed any provision that would not fund child care sponsored by religious institutions, arguing that since most of the existing nonprofit centers fell into that category, excluding them from federal funding limited the available supply of child care and restricted parental choice.
7.  Seventy-five percent of the funds was to be used for this purpose, while the remaining 25 percent was to be spent on quality improvement activities (5 percent), development of early childhood education (18.75 percent), and other quality-related activities (1.25 percent).
8. An alternative, less optimistic, interpretation is that devolution facilitates resolution by significantly reducing the power of diffuse interest groups such as poor families (Winston 1999).
9. The maximum time allowable is five years, but states have the option to reduce this, and many have.
10.  Had PRWORA not been passed, maximum allocations for 1997 would have been $ 1.4 billion for theTitle-IVA programs and $935 million under CCDBG (Long and Clark 1997,2).
11. These include, for each state, the proportion of children who are under five, how many children receive free or reduced-rate school lunches, and the average per capita income. Thus, both overall wealth and the gap between rich and poor are taken into consideration.
12.  For example, in 1994, Massachusetts, Connecticut, the District of Columbia, and Washington state all drew down 100 percent of their federal funds, while Louisiana drew down none, Mississippi 5 percent, andTennessee 27 percent. Alabama, however, received 95 percent of its maximum.
13.  For example, by failing to appropriate child care funds for 1997, only Arkansas, Mississippi, and New Mexico would end up with more money for child care than they had in 1995, while in most states, the amounts would range from 23 to 50 percent less (Long and Clark 1997,4).
14.  In 1997, the average SMI was $38,370, but this varied from just under $25,000 in West Virginia to just over $52,000 in Connecticut.
15.  In 1997, the federal poverty level for a family of three was $13,330 ($1110.80 per month) in all states except Alaska and Hawaii, where it was $16,670 and $15,330 respectively.
16.  Consensus has formed around the idea that "affordable" child care should take up no more than 10 percent of a family's total annual income; this percentage is, of course, far more onerous for a family earning $12,000 per annum than one earning $50,000 or more.
Denise Urias Levy and Sonya Michel    259
17.  Many states already attempt to discourage additional births through the use of "family caps" that deny benefits to children born to parents while they are receiving assistance.
18. It should be noted that in 2000, Ohio finally decided to reallocate funds, transferring more than $77 million, or 11 percent of itsTANF funds, to CCDF.
19. These projected increases in need were paralled by equivalent, if not greater, decreases in theTANF caseload; for fiscal year 2001, for example, Governor George Ryan requested an additional $66 million to cover a projected 12 percent increase in child care need, while calculating that the state would save $76 million from a projected 16 percent decrease in caseloads, a savings of $10 million. Other indicators, however, contradict the aforementioned signs of Illinois' generosity; according to Douglas and Flores (1998), the state is 33rd in child poverty, 8th in per capita income, and only 35th in fiscal effort (8, table 2). Indeed, for 2002, Ryan has drastically reduced his budget request for child care (Day Care Action Council 2001).
20.  Negative administrative orientation has been flagrant in other areas of Illinois' welfare bureaucracy, as IllinoisWelfare News (2000c) has demonstrated consistently sinceTANF went into effect. Its March 2000 column "Let's Get It Right" documented in detail one case of application delay and one of improper processing of appeals. IllinoisWelfare News is published by the National Center on Poverty Law, which works with the Legal Assistance Foundation of Chicago and Prairie State Legal Services to advise TANF clients and assist them in making appeals.
21. The classic study of such arrangements is Stack (1974).
22. Assuming this was their preference; many women in these neighborhoods preferred placing their children in the care of coethnics rather than in formal centers that they perceived as sterile or culturally alien (Uttal 1998). For a historical perspective on this, see Lemke-Santangelo (1996).
23.The Illinois program requires the sponsoring child care program as well as the recipient to pay part of the cost; the sponsoring program must also provide release time and "agree to award the recipient either a bonus or raise at the end of the scholarship year." Students pay up to 20 percent of expenses; in return, they must "agree to remain at the sponsoring center, or continue to operate their center or home, for at least one year after the term of the scholarship and remain in the early childhood field in Illinois for an additional year" (T.E.A.C.H. n.d., 3-4).
24.  It is not yet clear what impact, if any, the economic downturn of 2001 will have on these trends.
25.  Indeed, many conservative women, with the support of their menfolk, have mobilized a back-to-the-home or mothers-at-home movement which is, among other things, adamantly opposed to child care (Michel 2000).
26. Americans might look to the French system for an excellent model of family child care, in which providers are linked into a local or neighborhood center that they visit regularly for advice and assistance with "problem" children, bringing their charges along with them to be supervised while they are attending sessions. Centers also loan toys, equipment, and other supplies (Richardson and Marx 1989).
27. As of March 2001, Democrats on the Senate Finance Committee, which oversees
260    More Can Be Less
welfare and child care legislation in that body, were considering proposing legislation to accomplish just these goals (Democratic Staff 2001).
28.  Indeed, the history of federal programs like certain parts of Social Security should have taught us that certain states cannot be relied upon to make fair and equitable provisions for all of their citizens (Mettler 1998, 6—7 and passim).
29.  Consider, for example, calls for nighttime child care to enable women to work the "swing" or "graveyard" shift. (This often becomes necessary because office cleaning, one of the few occupations open to unskilled entrants to the job market, is usually done at night.) On the one hand, such services enable women to take jobs, but on the other, they also disrupt children's routines and probably lead to exhaustion for mothers who have no time to rest during the day. See also Arlie Hochschild's (199S) comments about "hot" and "cold" modern relationships between work and family.
References
Adams, Gina, and Nicole Oxendine Poersch. 1996. Who Cares? State Commitment to Child Care and Early Childhood Education. Washington, D.C.: Children's Defense Fund.
Bell, Winifred. 196S. Aid to Dependent Children. New York: Cambridge University Press.
Bernstein, Nina. 2001. The Lost Children ofWilder:The Epic Struggle to Change Foster Care. New York: Pantheon.
Blank, Helen, and Nicole Oxendine Poersch. 1999. State Child Care and Early Education Developments .Washington, D.C.: Children's Defense Fund.
Blank, Helen, Karen Schulman, and Danielle Ewen. 1999. "State Prekindergarten Initiatives, 1998—1999."Executive Summary, Seeds of Success. Washington, D.C.: Children's Defense Fund.
Children's Defense Fund. 1996. America's Children. Washington, D.C.: Children's Defense Fund.
__________.   2001. "New  Investments  for  Child  Care  Needed." Available  online  at
http://www.cdfactioncouncil.org/New%20Investments.htm.
Day CareAction Council of Illinois. 2001. Press release, March 8.
Democratic Staff, U.S. Senate. 2001. Personal communication with the author. March 20.
Douglas, Toby, and Kimura Flores. 1998. "Federal and State Funding of Children's Programs."Washington, D.C.: Urban Institute. Available online at http://newfederalism. urban. org/html/occaS.html#appe.
Esping-Andersen, Gosta. 1990. The Three Worlds of Welfare Capitalism. Princeton, N.J.: Princeton University Press.
Fuller, Bruce, and Sharon Kagan. 2000. Remember the Children: Mothers BalanceWork and Child Care UnderWelfare Reform. Berkeley and New Haven: Growing Up In Poverty Project, University of California-Berkeley andYale University.
Galinsky, Ellen, Carolle Howes, Susan Kontos, and Marybeth Shinn. 1994. The Study of Children in Family Child Care and Relative Care: Highlights of Findings. New York: Families and Work Institute.
Denise Urias Levy and Sonya Michel    261
Gallagher, J., and R. Clifford. 2000. "The Missing Support Infrastructue in Early Childhood." Early Childhood Research and Practice 2, no. 1. Available online at www.ecrp.uiuc.edu/ v2nl /gallagher.html.
Gordon, Linda. 1994. Pitied But Not Entitled: Single Mothers and the History of Welfare, 1890—1935. Cambridge, Mass.: Harvard University Press.
Greenberg, Mark. 1996. "A Summary of Key Child Care Provisions of H.R. 3734, The Personal Responsibility and Work Opportunity Reconciliation Act of 1996. Washington, D.C.: Center for Law and Social Policy (CLASP). Available online at http://www.clasp.org/pubs/childcare/clkccp.html.
__________. 1998. "The Child Care Protection Under TAN F." Washington, D.C.: CLASP.
Available online at http://www. clasp, org/pubs /childcare / childcareprotect. htm.
Greenberg, Mark, Joan Lombardi, and Rachel Schumacher. 2000. "The Child Care and Development Fund: An Overview."Washington, D.C.: Center for Law and Social Policy. Available online at http://www.clasp.org.
Harrington Meyer, Madonna, ed. 2000. Care Work: Gender, Labor and theWelfare State. New York: Routledge.
Hayes, Cheryl D., John L. Palmer, and Martha J. Zaslow. 1990. Who Cares for America's Children? Child Care Policy for the 1990s. Washington, D.C.: National Academy Press.
Hochschild, A. 199S. "The Culture of Politics: Traditional, Postmodern, Cold-Modern, and Warm-Modern Ideals of Care." Social Politics 2: no. 3, 331—46.
__________. 1997. The Time Bind: When Work Becomes Home and Home Becomes Work. New York:
Metropolitan Books.
Hofferth, Sandra. 1993. "The 101st Congress: An Emerging Agenda for Children in Poverty." In Child Poverty and Public Policy, ed. Judith A. Chafel. Washington: Urban Institute.
Illinois Welfare News. 2000a. "Child Care Increases As Income Assistance Declines in Governor's Revised 2000 Spending and 2001 Budget." March: 1,6.
__________. 2000b. "TANF Applicants and Recipients Not Getting the Child Care Assistance
They Need and Are Being Penalized." March: 6—7.
__________. 2000c. "Let's Get It Right! Undue Delays in Processing Benefits and Appeals."
March: 11.
Kornbluh, Felicia. 2000. "A Right to Welfare? Poor Women, Professionals, and Poverty Programs." Ph.D. dissertation, Princeton University.
Lemke-Santangelo, Gretchen. 1996. Abiding Courage: African American MigrantWomen and the East Bay Community. Chapel Hill: University of North Carolina Press.
Levy, Denise Urias. 2000. "Child Care Policies in America: Inter-State Differences and the Progress of Devolution." Ph.D. diss., University of Illinois at Chicago.
Lewis, Jane. 1992. "Gender and the Development of Welfare Regimes. "Journal of European Social Policy 2, no. 3: 37—SS.
Lippe,Tanja van der, and Liset van Dijk, eds. 2001. Women's Employment in a Coroprate Perspective. New York: Aldinede Gruyter.
Little, Deborah. 1999. "IndependentWorkers, Dependable Mothers: Discourse, Resistance, and AFDC Workfare Programs." Social Politics 6, no. 2: 161—202.
262    More Can Be Less
Long, Sharon K., and Clark, Sandra J. 1997. "The New Child Care Block Grant: State Funding Choices and Their Implications," no. A-12, "New Federalism: Issues and Options for States."Washington, D.C.: Urban Institute. Available online at http://newfederalism. urban.org/html/ anf_al2.htm.
Mettler, Suzanne. 1998. Dividing Citizens: Gender and Federalism in New Deal Public Policy. Ithaca, N.Y.: Cornell University Press.
Michel, Sonya. 1999. Children's Interests / Mothers' Rights:The Shaping of America's Child Care Policy. New Haven, Conn.: Yale University Press.
______________. 2000. "Having the Right to Care—and Not to Care." In Harrington Meyer,
ed., Care Work.
Mink, Gwendolyn. 1998. Welfare's End. Ithaca, N.Y.: Cornell University Press.
Mittelstadt, Jennifer. 2001. "'Dependency as a Problem to Be Solved': Rehabilitation and the Liberal Consensus on Welfare in the 1950s." Social Politics 8, no. 3: 228—57.
Oliker, Stacie. 2000. "Examining Care at Welfare's End." In Harrington Meyer, ed., Care Work.
Richardson, Gail, and Elisabeth Marx. 1989. AWelcomefor Every Child. New York: French-American Foundation.
Schumacher, Rachel, and Mark Greenberg. 1999. "Child Care After Leaving Welfare: Early Evidence from State Studies." Washington, D.C.: CLASP.
Schumacher, Rachel, Mark Greenberg, and Janellen Duffy. 2001. "The Impact of TANF Funding on State Child Care Subsidy Programs." Washington, D.C.: CLASP.
Sinclair, Charlene. 2001. Personal communication with the author.
Stack, Carol B. 1974. All Our Kin: Strategies for Survival in a Black Community. New York: Harper and Row.
T.E.A.C.H. Early Childhood Illinois, n.d. [2001]. "Scholarship and Compensation Project." GlenEllyn,Ill.:T.E.A.C.H.
U.S. Bureau of the Census. 1998. "Primary Child Care Arrangements Used for Preschoolers by Families with Employed Mothers, Selected Years, 1977 to 1994." Available online at http://www.census.gov/population/socdemo/child/p70-62/tableA.txt.
U.S. General Accounting Office. 1995a. Welfare toWork: Child Care Assistance Limited: Welfare Reform May Expand Needs. GAO/HEHS-95-220, September. Washington, D.C: Government Printing Office.
__________. 1995b. Child Care: Recipients Face Service Gaps and Supple Shortages. GAO/T-
HEHS-95-96, March 1. Washington, D.C: Government Printing Office.
U.S. House of Representatives, Committee on Ways and Means. 1999. 1999 Green Book. Washington, D.C: Government Printing Office.
University of Illinois Extension. 2001. "Child Care in Champaign County." Available online at http://www.aces.uic.edu/-CCRSCare/champstats.htm.
Uttal, Lynet. 1998. "Racial Safety and Cultural Maintenance:The Childcare Concerns of Employed Mothers of Color." In Families, Kinship and Domestic Politics in the U.S..: Critical Feminist Perspectives, ed. Anita Garey and Karen Hansen. Philadelphia: Temple University Press.
Waller, Margy. 1997. "Welfare-to-Work and Child Care: A Survey of the Ten Big States."
Denise Urias Levy and Sonya Michel    263
September 26. Washington, D.C: Progressive Policy Institute. Available online at http://www.dlcppi.org/texts/social/ccare.htm.
Winston, Pamela. 1999. "The Devil in Devolution: Welfare, the Nation, and the States." Ph.D. diss., Johns Hopkins University.
Young, Iris. Forthcoming.'"Autonomy, Welfare Reform, and Meaningful Work." In Philosophical Approaches to Dependency, ed. Eva Feder Kittay and Ellen Feder. Lanham, Md.: Row-man and Littlefield.