Summer 1997

Welfare Reform, Work-Related Child Care, and Tax Policy:

The "Family Values" Double Standard

 

Mary L. Heen [*]


Introduction

 

{1}The welfare reform legislation signed into law last year repeals the entitlement to welfare and imposes strict time limits on the receipt of benefits.[1] Although federal work requirements have been in effect for nearly thirty years, the new law requires the states to meet more stringent work participation levels [2] and makes the work requirements applicable to mothers with younger children.[3] The shift in the welfare paradigm toward mandatory wage work for mothers with young children has not been accompanied, however, by a corresponding policy shift toward universal or affordable child care.

{2}Historically, federal welfare and labor policies have impeded women's access to the wage labor market through the lack of affordable child care.[4] Tax policies have contributed to the problem.[5] Efforts to improve women's access to the wage labor market have clashed with policies aimed at reinforcing traditional family values. The policy conflict between increased labor market participation by women and reinforcement of traditional "family values" reflects race- and class-based double standards in the treatment of work and child care by both the income tax and the income transfer (welfare) systems.

{3}Tax policies evidence a tension between reinforcing traditional family values and improving the access of women to the wage labor market. Congress has articulated various reasons for the tax allowance for work-related child care; it has analogized work-related child care to other business-related costs of producing income and at the same time has treated it as a hardship allowance for families disrupted by the death or disability of the primary breadwinner (usually the husband and father) or the death or disability of the primary caregiver (usually the wife and mother). In the early 1970s Congress linked the child care deduction to welfare- related work programs and expanded the deduction to encourage the employment of welfare recipients in household service positions.[6] Policymakers also have periodically addressed child care issues by providing additional or alternative tax allowances for families with children through increased exemption amounts for dependents or by advocating refundable or nonrefundable per child tax credits.[7] These tax adjustments are sometimes described as promoting traditional family values because they do not tie eligibility for the tax allowance to the parents' work outside of the home. Child tax credit proposals directed at the middle class have recently received renewed political support, and a $500 per child tax credit has been enacted as the centerpiece of the Taxpayer Relief Act of 1997.[8]

{4}In requiring wage work of welfare mothers with young children, policymakers assume that the care welfare mothers provide their own children does not constitute work at least equivalent in value to the wage work available to welfare recipients (including child care they may provide to other people's children);[9] alternatively, they assume that the wage work required of welfare recipients will produce long-term benefits greater than the intervening cost of providing (or not providing) substitute child care for their children.[10] At best those assumptions evidence an underestimation of the cost of quality substitute child care. At worst they reveal an entrenched race- or class-based devaluation of the care provided by welfare recipients to their children. Without the provision of adequate substitute child care, the work requirements represent an attempt to shift welfare mothers into poorly paid service positions while tacitly expecting that their child care responsibilities will be met by neighbors and relatives, including the aunts, siblings, and grandmothers of the children now receiving welfare. In any event the largely unstated assumptions suggest disturbing race, gender, and class stereotyping at work, along with a return to certain pre-entitlement era approaches to poor relief.[11]

{5}On the one hand tax policies favor the in-home provision of child care and household services by mothers in certain "traditional" two parent households and facilitate the employment of child care providers if the single parent or secondary wage earner (usually the wife) can earn enough after taxes in the wage labor market to pay for child care and other household services. On the other hand welfare policies reject the in-home provision of child care for poor mothers. Low-income families are generally unable to afford adequate child care without additional government subsidies or the modification of current tax provisions. The interrelationship of tax and welfare policies thus creates a classic double bind for poor families and suggests an apparent race- and class-based double standard.

{6}The implications of the double standard applied to the poor through the tax system are troubling, and worthy of further examination. This essay begins a preliminary exploration of the interrelationship between tax and welfare double standards. In the welfare context, discussed in Part I, a double standard historically has been applied by making race-based distinctions between the "deserving" and the "undeserving" poor. Although the discriminatory denial of welfare benefits largely ended as a result of major reforms achieved by the welfare rights movement in the 1960s,[12] welfare reform eliminates the structural "entitlement" to benefits on which those reforms were built. In the tax context, discussed in Part II, a family values double standard may be identified through the close correlation between an individual's race or gender and his or her family income or wealth.[13] Traditional family values are reinforced through the tax code; nevertheless, access to the wage labor market has been improved for middle- and upper-income women through offsetting tax allowances for work- related child care. Those adjustments are not generally available, however, to low-income working families. The tax double standard may be eliminated only through offsetting adjustments or more comprehensive changes in the income tax system.

I. Welfare, Work and the Mothers of Young Children

A. The Historical Race-Based Double Standard

{7}Federal work requirements for welfare recipients represent a shift away from the origins in 1935 of the Aid to Dependent Children (ADC) program, which developed from "mothers' pensions" or "mothers' aid" programs.[14] Such programs were intended by social reformers to enable widows and certain other "deserving" mothers with "suitable homes" to care for their young children without being compelled to work outside of the home.[15] Local welfare offices, particularly those in the South, used the "suitable home" and other rules to deny assistance to African-American children and their families.[16] In addition, long before the federal government imposed work requirements, some states used "employable mother" rules to deny welfare assistance to women with children, especially nonwhite women, on the ground that they should work.[17] The first employable mother rule was adopted by Louisiana in 1943, refusing ADC assistance to families during times when the mothers and their older children were needed to work in the cotton fields.[18] Georgia adopted a similar rule in 1952, denying assistance to mothers with children over three years of age if "suitable" employment (at any wage) was deemed to be available.[19]Thus, local welfare policies coincided with local labor market demands by keeping nonwhite women in seasonal agricultural and other labor pools.[20] As late as 1966 New Jersey notified Aid to Families with Dependent Children (AFDC) recipients that their grants would be cut because seasonal farm work was available.[21]

{8}Federal work requirements were first imposed on welfare recipients by the Work Incentive Program (WIN) in 1967. WIN was not very effective, however, due to weak funding and enforcement.[22] Although potentially subject to the original WIN work requirements, mothers with preschool children were determined by many states to be "inappropriate" for job training or work and thus exempt from the work requirement.[23] As amended in 1971, WIN II required participation by mothers with children six years of age or older.[24]

{9}The WIN program was replaced by the Job Opportunity and Basic Skills Program (JOBS), which was established by the Family Support Act of 1988.[25] The Family Support Act mandated improved procedures for child support enforcement and the establishment of paternity; guaranteed federal assistance for child care during participation in education, training, and employment (AFDC-related child care); and provided transitional eligibility for a year of extended child care and medical assistance for former AFDC recipients who became ineligible for AFDC because of increased income from employment (Transitional Child Care).[26] Unless exempted by law, AFDC recipients were required to participate in JOBS.[27] Mothers caring for a child under six years of age were required to participate in JOBS only if child care were guaranteed and participation were limited to twenty hours per week.[28] Those caring for children under three years of age were exempt from participation, unless required to participate under State option.[29] Most states, however, exempted caretakers of children under the age of three from the work requirements.[30]

{10}The shift in welfare policy toward work requirements occurred as the welfare population expanded and as more African-American and other women of color and their families were added to the rolls.[31] The shift also coincided with the trend of increased labor market participation by women with young children.[32] Although AFDC mothers have been reported as participating in the labor market at significantly lower levels than their nonwelfare counterparts,[33] studies conducted by the Institute for Women's Policy Research show more comparable labor-force participation levels.[34] Like many of their counterparts in the labor market, AFDC mothers do not earn enough on their own to support themselves and their children; many need both their welfare benefits and their earnings to survive. Because AFDC was not otherwise structured to encourage work effort, the paid work of welfare recipients has sometimes been driven underground.[35]

B. Child Care Funding Under JOBS and Under the New Welfare Reform Law

{11}As discussed above, the JOBS program was created in the last round of welfare reform during the late 1980s. The JOBS program was funded through a capped entitlement under which states were partially reimbursed (pursuant to a federal matching rate) for each dollar spent on JOBS until they reached the maximum amount allocated to them.[36] Federal funds for guaranteed JOBS-related child care were separately provided as open-ended entitlement matching funds to partially reimburse (at the Medicaid matching rate) state expenditures for AFDC-related child care and Transitional Child Care.[37] As of the end of fiscal year 1993, states had drawn down only about 70% of the allotted $1 billion in federal JOBS funds.[38] State budget constraints as well as the cost of guaranteed child care were among the reasons identified for the less-than-full implementation of the JOBS program. In many states the same amount was spent on JOBS-related child care as on the JOBS program itself.[39]

{12}The At-Risk Child Care Program provided federal matching funds for states to provide child care services for low-income families who were "at risk" of becoming welfare recipients if they did not receive work-related child care.[40] Families were required to contribute to the cost of care on a sliding fee schedule based on the family's ability to pay.[41]

{13}Low-income families also received child care assistance through various federal block grant programs, which survive the new welfare reform law in modified form. The Child Care and Development Block Grant (CCDBG) program provides funding for child care services to low-income families, as well as for efforts to improve the quality and availability of child care in general.[42] Federal funds are distributed to states under a formula, and no matching funds are required. In addition to the CCDBG program, some child care funds are available through the Social Services Block Grant Program of Title XX of the Social Security Act.[43] Title XX block grants operate as a capped entitlement, with no state matching requirement, under which states are allocated funds pursuant to a formula based on their relative population. Most states spend some portion of their block grants on child care services, and some but not all states determine eligibility for child care services based on income standards.[44]

{14}Under pre-welfare reform federal child care funding levels,[45] states were unable to meet the need for child care assistance for low-income families.[46] Between 5 and 6% of the AFDC caseload received AFDC child care subsidies,[47] and only about one-third of JOBS participants received JOBS-related child care.[48] About 20% of those eligible received Transitional Child Care assistance for the first year after leaving welfare for work.[49] Families that used up their one year of guaranteed transitional assistance after leaving welfare had to compete with other low-income, non-AFDC families for child care assistance. Surveys conducted in 1993 and 1994 found that most states either had lengthy waiting lists for child care assistance or had stopped accepting new applications.[50] In addition, the competition for slots resulted in the shifting of scarce state child care funds from low-income working families to families receiving AFDC.[51]

{15}Experience with prior work programs indicates that those with low reimbursement rates and retroactive reimbursement tended to steer families toward informal child care.[52] Such informal arrangements are also more likely to be of relatively poor quality. A recent study concluded that children who are in the care of family and relatives are receiving substandard care from providers who are "taking care of children to help out the mothers and not because they want to care for children."[53] The study's authors recommended that low-income families receive a child care subsidy sufficient to pay for higher quality care.[54] In addition, the authors recommended against requiring welfare recipients to become family child care providers and urged states to screen all welfare-to-work recipients for interest, commitment, and aptitude before they become providers.[55]

{16}The new welfare reform law repeals "entitlement" programs such as AFDC and AFDC-related child care and substitutes two separate capped federal block grants to the states, giving the states the freedom to impose their own requirements or restrictions without the necessity of applying for waivers of federal requirements.[56] The new law cuts the projected growth in overall welfare spending by approximately $55 billion over the next six years.[57] It also increases the projected number of children in poverty.[58]

{17}In place of AFDC and JOBS, the new law creates a welfare block grant called Temporary Assistance for Needy Families (TANF) capped at $16.4 billion per year, approximately the level of federal welfare expenditures in 1995. Implementation of TANF is effective July 1, 1997, although states may implement their block grant programs sooner.[59]

{18}The new block grant for child care, effective October 1, 1996, is an expanded and revised version of the Child Care and Development Block Grant Program (CCDBG), and replaces AFDC-related child care, Transitional Child Care, and the At-Risk Child Care programs.[60] The new child care block grant largely consolidates federal child care programs, funded at 1995 levels of just over $2 billion per year (provided through numerous separate programs),[61] into one block grant program of about $3 billion in total funds in fiscal year 1997, increasing to $3.7 billion by fiscal year 2002.[62] Although the new law adds limited amounts over 1995 funding levels for work-related child care, the amounts provided fall far short of what would be needed to move families off the welfare rolls and to keep them off on a long-term basis. Without substantially increased federal or state support of work-related child care, the new work requirements may be programmed for failure, or worse, they may result in the endangerment of children.

{19}If low-income families must pay the full cost of child care themselves, they face a major obstacle in their transition from welfare to work. In general the type of child care purchased and the amount spent on care varies by the family's economic situation and the type of care used. Lower-income families spend on average about 25% of their incomes on child care even though they spend significantly less, in absolute terms, on child care than families with higher incomes.[63] Without subsidized child care, low-income families will likely rely on lower-quality child care or informal arrangements and relative-provided care. Those who pay relatives to care for their children pay the lowest average weekly costs, with increasingly higher weekly average costs for family child care, center care, and in-home care by a nonrelative.

{20}The family and relative care received by children from low-income families and the center-based care for very young children have raised developmental concerns. A recent study of children in family child care and relative care concluded that "regardless of maternal education, the lower the child's family income, the lower the quality of the child care home in which he or she is enrolled."[64] That finding differed from research findings on center-based care, in which low-income children in subsidized care often were in better-quality arrangements than middle-income children.[65] In center-based care, the lowest quality care is received by toddlers and infants, with about 40% of those studied receiving below a minimally adequate level, although little difference in fees was found for centers providing high- or low- quality care.[66]

II. Taxes and Work: Family Values and Access to the Wage Labor Market

A. A History of Congressional Approaches

{21}Congress has combined at least two or three notions in its approach to work-related child care costs for income tax purposes. Although it has treated child care expenses as comparable to an employee business expense, it has also targeted the deduction or credit to hardship situations and used the allowance as part of an overall effort to develop jobs for household workers, including former welfare recipients.[67]

{22}Congress first provided a tax adjustment for employment-related child care costs in 1954,[68] as a type of working expense deduction targeted to those in hardship situations such as widows, widowers, and low-income families.[69] Over the next two decades, Congress increased the statutory dollar amounts and expanded the coverage of the child care provision, but retained the basic structure of a child care deduction.

{23}In the 1970s the rationale for the deduction shifted to include a job development purpose in addition to its continued function as a type of employee business expense in hardship situations.[70] When the deduction was significantly expanded in 1971, the Senate committee report identified it as a "job development deduction for household services and child care," and discussed it immediately following its description of a proposed tax credit for salaries paid welfare recipients under the WIN program.[71] When discussing the proposed WIN tax credit, the committee observed that the WIN program "has not been as successful as had been hoped, largely because persons have been placed in institutional rather than employment-based training."[72] The expanded child care deduction was intended to provide an incentive for the employment of household workers by giving large numbers of welfare recipients "the opportunity to perform socially desirable services in jobs that are vitally needed," while also helping "to remove these individuals from the welfare rolls and reduce the cost of providing public assistance."[73]

{24}In 1976 Congress changed the child care deduction to a nonrefundable tax credit by repealing the deduction provision and adopting the predecessor of the current child care tax credit.[74] The change to a tax credit was adopted as a way to reach taxpayers who elect the standard deduction, and as a simplification measure.[75] Although the family income limitation amounts were eliminated and eligibility requirements were somewhat broadened, the credit otherwise retained the basic design of the earlier provisions with regard to determination of qualified expenses.[76] The credit was changed again in 1981, resulting in the current child care tax credit structure.[77] Although the Senate version of the credit included refundability, the conference agreement rejected making the credit refundable.[78] The credit was redesignated in 1984,[79] and Congress did some fine-tuning to curtail perceived abuses or to make technical adjustments in 1987,[80] and again in 1988.[81] No major changes have been made to the credit since 1981.

B. The Income Tax Work-Related Child Care Provisions

{25}The Internal Revenue Code provisions specifically addressing child care expenses are I.R.C. § 21, the child and dependent care tax credit, and I.R.C. § 129, the exclusion from income for certain employer-provided child care benefits. The child care tax credit and the exclusion for employer-provided child care are estimated to reduce federal revenues by about $2.8 billion and $.8 billion, respectively, in fiscal year 1997.[82] Although not specifically aimed at the child care expenses of working parents, the earned income tax credit, I.R.C. § 32, provides a refundable tax credit for certain low-income working families with children. In addition, the personal exemption deduction for dependents, I.R.C. § 151, and the recently enacted child tax credit, I.R.C. § 24, provide tax adjustments to account for the added household costs of supporting children.[83] I.R.C. § § 21 and 129 provide tax benefits to all working parents, but upper- and middle-income taxpayers utilize them the most, for reasons explained below. The following subpart describes in greater detail how the child care credit and employer-provided child care exclusion provisions work, and how the current design of these provisions makes it difficult for low-income taxpayers to benefit from them.

1. How the Child Care Tax Credit Works

{26}I.R.C. § 21 provides a nonrefundable tax credit, the amount of which is equal to an "applicable percentage"[84] of the eligible employment-related child care expenses[85] paid by the taxpayer during the year.[86] The applicable percentage, which ranges on a sliding scale of 20 to 30%, varies with adjusted gross income. The amount of child care expenses that may be taken into account depends upon the number of children[87] included in the household maintained by the taxpayer. Eligible expenses are limited to $2,400 per year for one child, and $4,800 per year for two or more children.[88] A taxpayer with adjusted gross income of $10,000 or less receives a credit of 30% of employment-related expenses. The credit percentage declines by one percentage point for each $2,000 (or fraction thereof) in adjusted gross income above $10,000, but in no case is the applicable percentage reduced below 20%.[89] For taxpayers with adjusted gross incomes greater than $28,000, therefore, the applicable percentage is 20%. For taxpayers with adjusted gross income of $10,000 or less, and thus qualifying for the highest applicable percentage of 30%, the maximum credit is $720 for one child, and $1,440 for two or more children. For taxpayers with incomes in excess of $28,000, and thus qualifying for the lowest applicable percentage of 20%, the maximum credit is $480 for one child, and $960 for two or more children.

{27}The amount of the dependent care credit and the applicable percentage income phase-down schedule have not changed since 1981. Income tax thresholds, however, have substantially increased since then.[90] Thus, although § 21 appears to target low income taxpayers, the relationship between the credit percentage income phase-down and current income tax thresholds makes it unlikely that poor taxpayers receive any benefit from the credit. The Tax Reform Act of 1986[91] removed about six million poverty level families from the income tax rolls[92] by increasing standard deduction[93] and personal exemption amounts,[94] and adjusting those amounts on a yearly basis for inflation.[95] In 1997, for example, a family of four (two parents and two children) would owe no taxes on up to $17,500 of adjusted gross income,[96] which is above the federal poverty threshold for a family of four.[97] A single head of household with one child would owe no taxes on up to $11,350 of income,[98] which is above the poverty level for a family of two.[99] Although both families could be entitled to a child care tax credit, they would have no income tax liability to offset through use of the credit.[100] The current thresholds for tax liability,[101] combined with the nonrefundability of the credit, thus make it unlikely for poor families to benefit from the child and dependent care tax credit.[102]

2. The Exclusion for Employer-Provided Dependent Care Assistance Programs

{28}I.R.C. § 129 provides an exclusion from the gross income of employees of amounts up to $5,000 paid by the employer under a dependent care assistance program.[103] The dependent care assistance program must be a separate written plan of the employer for the exclusive benefit of employees[104] and must meet certain other requirements.[105] The amount of the exclusion may not exceed the lesser of the earned income of the employee or the earned income of the employee's spouse.[106] Payments for child care made to the employee's spouse or certain other related individuals (another child of the employee, for example) are ineligible for exclusion.[107]

{29}Employers most frequently provide the dependent care assistance benefit through reimbursement accounts, sometimes referred to as flexible spending accounts, which may also cover other types of expenses, such as out-of-pocket health care expenses.[108] Up to $5,000 may be paid into an dependent care assistance account (through a salary reduction plan) from which child care expenses of the employee are reimbursed. The effect of such a program is that the employee may pay child care expenses (or out-of-pocket health care expenses) with pre-tax dollars. Thus, the I.R.C. § 129 exclusion operates as a complete adjustment, offsetting the tax costs of up to $5,000 of child care expenses, regardless of the taxpayer's marginal tax rate. About one-third of full-time employees at large and medium-sized private firms were eligible for such accounts in 1991, compared to nearly one-tenth of such workers who were eligible for child care benefits provided by the employer in the form of child care facilities provided at or near the workplace or through direct reimbursement of employee expenses.[109]

{30}Generally, taxpayers choose whether eligible child care expenses will be claimed under the § 21 credit or the § 129 exclusion. Double dipping is not permitted.[110] For most middle- or upper-income taxpayers, the § 129 exclusion will provide the most benefit.[111] For example, for taxpayers subject to the highest marginal tax rate of 39.6%, the § 129 exclusion is worth $1,980[112] compared to the maximum § 21 credit of $480 for one child or $960 for two or more children.

C. Tax Theory: The Implications of Viewing a Child Care Allowance as a Subsidy or as a Cost of Producing Income

{31}As discussed above, Congress has never really decided whether to conceptualize the child care credit as a cost of earning income, as a hardship allowance for child care, or as a job development program for household workers. Tax theorists have also disagreed regarding the proper treatment of such expenses.[113] If child care expenses are a legitimate cost of producing income, they should be deductible regardless of the amount or the taxpayer's income level.[114] But if child care costs are personal consumption expenditures, they should not be deductible -- just as expenditures for the costs of food or shelter are nondeductible.[115] Any special tax allowance for personal consumption expenditures may be viewed as a tax expenditure,[116] and thus equivalent to a direct subsidy for child care. The conclusion one reaches with regard to these theoretical issues may have as much to do with one's view of the family, and the role of women within the family, as with one's understanding of tax policy.[117]

{32}The following subparts consider whether an adjustment to income to reflect child care costs can be justified under the tax norms of ability to pay and neutrality. I conclude that an income tax adjustment for child care costs should not be viewed as a subsidy because it reflects a taxpayer's ability to pay taxes. Even if it were viewed as a subsidy or equivalent to a direct expenditure, however, there are arguments in favor of retaining or expanding an income tax adjustment for child care expenses. Such an adjustment should be tolerated as a "second best" solution because it offsets other tax nonneutralities between wage work and household labor. Eliminating the "subsidy" would exacerbate already serious allocative inefficiencies involving participation by women in the labor market.

1. The Ability-to-Pay Norm

{33}The ability-to-pay norm derives from the idea that taxpayers should contribute to the government according to the relative amount of material resources they control, above subsistence amounts. Ability to pay may be understood in both horizontal and vertical equity terms. Questions concerning the normative underpinnings of the traditional tax policy equity analysis have recently provoked much commentary, and several theorists have emphasized that if two taxpayers pay different amounts in tax, the difference must be consistent with an appropriate theory of distributive justice.[118] Analysis under the traditional tax norm of horizontal equity, under which similarly situated taxpayers should be similarly taxed, tends to be conclusory because of the lack of a tax-determined method of identifying similarly situated taxpayers. For example, the conclusion reached with regard to horizontal equity may depend upon whether one begins with a worker with or without children, and how one views one- earner versus two-earner working families.[119] Thus, the prior question of how taxpayers with equal incomes are identified determines the outcome of the horizontal equity analysis.

{34}The application of the ability-to-pay norm generally does not favor deductions unless they relate to minimum subsistence amounts, certain nondiscretionary expenditures,[120] or legitimate costs of producing income. Child care costs reduce the taxpayer's ability to pay taxes only if one concludes that child care fits within such a category of expenditures. If so, a deduction or a credit would be justified based on the taxpayer's reduced capacity to pay taxes.

{35}Debate about the tax treatment of child care costs generally centers on whether such expenses are personal or business expenses, that is, whether to treat such expenses as a cost of producing income or as a personal consumption expenditure. Although the business/personal boundary is difficult to delineate when the expenses involve additional costs of being employed, child care has been analogized to nondeductible personal expenses, such as commuting costs and higher clothing expense, where the person "already at work" marks the boundary between business and personal expenses.[121] Some have argued that child care costs may contain elements of either personal or business expenditures or a mixture of both,[122] and at least one commentator has suggested that a limit in the amount of deductible expenses may be appropriate as a means of restricting the personal consumption element for middle- or upper-income taxpayers.[123]

{36}Arguably, however, child care costs (up to some generally recognized standard amount for quality care) are legitimate costs of producing income, and a child care deduction properly reflects a working parent's ability to pay taxes. A caretaker is required if single or dual parents work outside of the home, and the tax code should recognize child care as a deductible work-related expense. But those who view child care costs as a personal expense would conclude that an income tax adjustment constitutes a subsidy. Under such a view the neutrality norm becomes more important because an adjustment to income cannot be justified on the basis of a working parent's relative ability to pay taxes.

2. The Neutrality Norm

{37}The neutrality norm derives from the notion that taxes should influence allocation of resources in the economy as little as possible; otherwise, economic inefficiencies may result. Under the neutrality norm subsidies are suspect and should be discouraged. For those who view child care costs as a personal consumption expense, an income adjustment for such costs would constitute an income subsidy violating the neutrality norm. The neutrality norm, however, is tempered by several other theories.

{38}First, under the theory of optimal taxation, nonneutrality does not result in economic distortions when taxes do not affect consumer or other allocative choices (that is, where there is a low degree of elasticity or substitutability of behaviors).[124] Accordingly, because economic distortions are a function of elasticity, higher taxes may be imposed on inelastic commodities without creating allocative inefficiencies. Second, under the Pigouvian theory of taxation, departures from the neutrality norm may be justified to correct market failures; when free markets do not work, through the presence of externalities or information failures, taxation may legitimately correct the failure.[125] For example, an observed market failure of parents or society to invest adequately in children's human capital[126] could provide theoretical justification for a tax incentive to correct such market failure. Third, the theory of the "second best" suggests that tax nonneutralities should not necessarily be corrected if allocative inefficiencies would be aggravated because of the existence of other nonneutralities.[127]

{39}The theory of the second best provides a justification for a tax allowance for work- related child care because of the existence of the tax nonneutrality between wage work and household labor.[128] The tax system generally favors nonmarket production by failing to tax imputed income from services taxpayers perform for themselves or their households.[129] For example, no income tax is imposed on the value of services such as vegetable gardening, meal preparation, or hair cutting provided by taxpayers to members of their own households. However, for those taxpayers who hire others to perform such services, no deductions from income are generally allowed for the cost of the services. Unless each taxpayer earns more than the value of the services plus taxes, the tax system encourages taxpayers to provide the services on an in-kind basis (assuming that they have or can develop the skill to perform the services). A deduction for market-purchased services would eliminate the tax incentive for home production. Alternatively, nonneutrality could be eliminated by including the imputed income from the in-kind family-provided services in gross income. Inclusion of imputed income would be quite problematic, however, due to the administrative difficulties of valuation and enforcement.

{40}Child care expenses arguably should be treated differently from other types of nondeductible household expenses because of the necessary relationship between child care and access to the labor market.[130] The need for a child care deduction or credit to offset the current tax incentive for a parent to provide child care at home is typically advanced in the context of a constellation of other social and economic factors discouraging women from full labor force participation.[131] Studies suggest that labor force participation of secondary workers responds to changes in tax rates,[132] and thus tax nonneutrality between wage work and household labor may result in allocative inefficiencies. The second best solution of a tax adjustment for work-related child care achieves special force in a setting otherwise discouraging women from entering or staying in the labor market.

D. The Double Standard Applied to Low-Income Taxpayers

{41}As argued above, the tax treatment of child care expenses cannot be evaluated in isolation from the Code's taxation of the family in general. Even if one concludes that the child care credit cannot be justified under the ability-to-pay norm, analysis under the neutrality norm suggests that a tax allowance for child care costs could be justified as a second best solution to currently existing nonneutralities between wage work and household labor. I.R.C. § § 21 and 129 offset, at least partially, the effects of certain policy trade-offs made in connection with taxation of the family. Thus, the child care allowances should be understood as serving important structural functions within the tax system. As I have argued elsewhere, because of the structural role played by these provisions as an offset to other nonneutralities involving taxation of the family, the child care tax credit should not be phased out for middle- and upper-income taxpayers as a means of redirecting benefits to low-income families. Instead, revenues should be reallocated from other sources to extend the benefits of such offsets to low-income families.[133]

1. The Policy Trade-Offs Creating Nonneutralities in Taxation of the Family

{42}Conflicts among the competing tax policy goals of marriage neutrality, progressivity, and the policy of taxing equal-income married couples equally[134] have forced inescapable trade-offs in the taxation of the family. As many analysts have pointed out, it is mathematically impossible to accomplish all three goals at the same time, and given a progressive rate structure, nonneutralities may result.[135] The tax system has shifted the balance among these goals over time as Congress has responded to changes in social patterns, distributive goals, and prevailing perceptions of the role of the family in society.

{43}The I.R.C. § 21 child care tax credit and the § 129 exclusion for employer-provided child care serve an important function as an offset to current nonneutralities between married and unmarried earners, given the following features of the current tax structure: (1) the phase-out percentages of the earned income tax credit for low-income workers,[136] (2) a progressive rate structure, and (3) a joint filing regime for married taxpayers. Unless these features of the tax structure are altered, the child care tax provisions should be retained or expanded as a second best solution.

2. How the Nonneutralities Affect Low-Income Taxpayers

{44}At low income levels, tax costs make working to cover child care expenses an inherently losing proposition. Although tax costs of working in the wage labor market are somewhat offset by I.R.C. § § 21 and 129 for middle- and upper-income taxpayers, low-income taxpayers receive little or no benefit from those provisions. Thus, the low-income mother generally is better off staying at home to care for the children unless she earns more than it costs to purchase adequate child care, or can rely on unpaid relatives or low-cost providers for child care.

{45}As has been pointed out by Professor Edward McCaffery, the tax costs result from a combination of the 15% marginal income tax rate on earned income above the tax threshold amounts, the 7.65% employee portion of social security taxes, and the phase-out percentage of the earned income credit.[137] The earned income tax credit is structured to benefit low- income working families: The amount of the credit initially increases with earnings, then remains constant as earnings increase, and then decreases with earnings until it is fully phased out. In 1997, the maximum benefit for a family with two or more qualifying children is $3,656 (as adjusted for inflation, equal to 40% of the earned income amount of $9,140).[138] The maximum benefit applies at incomes between $9,140 and $11,930, and declines thereafter. A phase-out percentage (21.06%) applies to adjusted gross income (or, if greater, the earned income) in excess of $11,930. Thus, the benefit will be fully phased out at $29,290 of adjusted gross income for a taxpayer with two or more qualifying children.[139] In 1997, for example, the marginal income tax rate (15%), the employee portion of social security tax rates (7.65%), and the earned income credit phase out rate (21.06%) equal a combined tax rate of 43.71%, without taking into account state taxes and the incidence of the employer portion of social security taxes.

{46}The earned income tax credit phase-out percentages have the effect of increasing the marriage penalty for families at low income levels; in addition, they make the marginal tax rate very high for low-income families earning at levels within the phase-out range.[140] A possible offsetting adjustment to these nonneutralities would be to make the child care tax credit refundable, and to increase the applicable percentage to at least 50% of an increased level of eligible child care expenses. Alternatively, § 129 programs could be made available to all employees.

III. Conclusion

{47}The tax double standard as applied to low-income families operates at cross purposes to current welfare reform proposals, and creates a double bind for welfare mothers. The problem could be partially addressed by delivering increased child care benefits through the tax system or through direct assistance programs, or some combination of the two. More comprehensive tax policy reforms would require revisiting the policy trade-offs regarding the use of the joint return, the goal of taxing equal earning families equally, and the nontaxation of imputed income from household labor. In the absence of more comprehensive tax reforms, the policy clash between welfare reform goals and the tax code will require increased attention by policy-makers to the effects of the family values double standard as applied to low-income families.

References

*Copyright © 1997 Mary L. Heen. Associate Professor of Law, University of Richmond. B.A., Yale University; J.D., University of California at Berkeley; LL.M. (Taxation), New York University. An earlier version of this essay appeared in Taxing America, a collection of essays on the social and economic implications of tax reform, edited by Karen B. Brown and Mary Louise Fellows, published by New York University Press in 1996. The essay has been revised to reflect major developments since the work for the book chapter was completed. Special thanks to Sandra Boscia and Karen Duncan for their research assistance, and to colleagues who commented on these ideas at a faculty colloquy held at the University of Richmond School of Law in January 1997. Some related portions of the essay were adapted from Mary L. Heen, Welfare Reform, Child Care Costs, and Taxes: Delivering Increased Work-Related Child Care Benefits to Low-Income Families, 13 YALE L. & POLICY REV. 173 (1995) (examining federal work-related child care programs, describing the need for increased assistance, and addressing the issue of whether additional federal child care assistance to low-income families should be provided through a transfer payment system, through the tax system, or through some combination of the two systems).

Notes

  1. Personal Responsibility and Work Opportunity Reconciliation Act of 1996, Pub. L. No. 104-193, § 103, 110 Stat. 2105, 2112-13, 2137 (repealing individual entitlement to welfare and imposing a twenty-four month limit on welfare benefits without work and a sixty month lifetime cap on benefits) [hereinafter Personal Responsibility Act of 1996].
  2. Id. § 103, 110 Stat. at 2129 (requiring states to meet a minimum work participation rate of 50% by fiscal year 2002).
  3. Id. § 103, 110 Stat. at 2131 (permitting States to exempt parents of a child under age one from the work requirement); Id. § 103, 110 Stat. at 2132 (providing that for purposes of determining monthly minimum work participation rates, a single parent of a child under the age of six is deemed to be meeting work participation requirements if the parent is engaged in work for 20 hours per week); Id. § 103, 110 Stat. at 2133 (exempting from failure-to-work penalties single custodial parents caring for a child under the age of six if the parent "proves that [she] has a demonstrated inability (as determined by the State) to obtain needed child care," including the unavailability or unsuitability of "informal child care" by a relative or under other arrangements). For a discussion of the work requirements imposed under prior law, see discussion infra Part IA.
  4. E.g., Sylvia A. Law, Women, Work, Welfare, and the Preservation of Patriarchy, 131 U. PA. L. REV. 1249, 1310-18 (1983).
  5. E.g., Edward J. McCaffery, Taxation and the Family: A Fresh Look at Behavioral Gender Biases in the Code, 40 UCLA L. REV. 983, 986-87 (1993).
  6. See discussion infra, Part II.A.
  7. See, e.g., C. EUGENE STEUERLE & JASON JUFFRAS, A $1,000 TAX CREDIT FOR EVERY CHILD: A BASIS OF REFORM FOR THE NATION'S TAX, WELFARE, AND HEALTH SYSTEMS 3-4 (Urban Institute Changing Domestic Priorities Series 1991); see also Jonathan B. Forman, Beyond President Bush's Child Tax Credit Proposal: Towards a Comprehensive System of Tax Credits to Help Low-Income Families with Children, 38 EMORY L.J. 661, 693-96 (1989). A tax credit reduces a taxpayer's tax liability dollar for dollar in the amount of the credit. If the taxpayer owes no tax, a tax credit is of no benefit unless it is refundable. See generally Lawrence Zelenak, Children and the Income Tax, 49 TAX L. REV. 349 (1994) (discussing various proposals for change).
  8. During the last session of Congress, although House and Senate conferees agreed to a nonrefundable $500 per child tax credit, phasing out the credit at adjusted gross incomes above $110,000 for joint returns, and $75,000 for unmarried individuals, the provision did not become law due to a standoff over the budget. H.R. CONF. REP. NO. 350, 104th Cong., 1st Sess. 473, 1292-94 (1995). President Clinton's recent budget proposal included a nonrefundable $500 per child tax credit for families with children under age 13. Under the President's proposal, the credit was $300 per child for tax years 1997-1999, increased to $500 per child in 2000, and adjusted for inflation after 2000. The credit would be applied before the refundable earned income credit and would be phased out for taxpayers with adjusted gross income between $60,000 and $75,000. See Excerpts from President Clinton's FY 1998 Budget Submitted to Congress Feb. 6, 1997, reprinted in Daily Tax Rep. Special Supp. (BNA) No. 26, at S-7, S-8 (Feb. 7, 1997) (estimated revenue loss of $9.9 billion in 1998, $6.8 billion in 1999, and $8.6 billion in 2000); Joint Committee on Taxation, Estimated Budget Effects (JCX-8-97) of Revenue Provisions Contained in President's FY 1998 Budget Proposal, Issued Feb. 27, 1998 [sic], reprinted in Daily Tax Rep. (BNA) No. 40, at L-3, L-6 (Feb. 28, 1997) (estimated revenue loss of $8.87 billion in fiscal year 1998, $7.67 billion in fiscal year 1999, $8.06 billion in fiscal year 2000). The Republican leadership's tax proposals included a nonrefundable $500 per child tax credit that would have resulted in larger annual revenue losses over that period. See S. 2, 105th Cong., § 101 (1997) (proposing a $500 per child tax credit for families with children under age 18, with phaseout provisions similar to those vetoed in H.R. 2491).
  9. The Taxpayer Relief Act of 1997, signed into law on August 5th, 1997, contains elements of both the Clinton and Republican proposals. The new law adds I.R.C. § 24, which provides a child tax credit of $400 beginning in 1998, and $500 thereafter, for each qualifying child under the age of 17. The credit phases out beginning at modified adjusted gross incomes of $110,000 in the case of joint returns, $75,000 for unmarried individuals, and $55,000 for married individuals filing separately. Although generally a nonrefundable credit, it is partially refundable for certain low-income taxpayers with three or more qualifying children. Taxpayer Relief Act of 1997, Pub. L. No. 105-34, § 101, 111 Stat. 788. The new child tax credit is estimated to cost $183.4 billion over 10 years. Staff of Joint Committee on Taxation, Estimated Budget Effects of the Conference Agreement on the Revenue Provisions of H.R. 2014, the "Taxpayer Relief Act of 1997," (JCX-39- 97), reprinted in 76 Tax Notes 592 (Aug. 4, 1997).

  10. See, e.g., Gwendolyn Mink, Welfare Reform in Historical Perspective, 26 CONN. L. REV. 879, 881-83 (1994); Dorothy E. Roberts, The Value of Black Mothers' Work, 26 CONN. L. REV. 871, 873-75 (1994).
  11. See Lance Liebman, Evaluating Child Care Legislation: Program Structures and Political Consequences, 26 HARV. J. ON LEGIS. 357, 360-61 (1989) (questioning whether it is important that single mothers work even if child care costs more than their short-term earnings); see also Martha L. Fineman, Images of Mothers in Poverty Discourses, 1991 DUKE L. J. 274.
  12. E.g., MIMI ABRAMOVITZ, REGULATING THE LIVES OF WOMEN: SOCIAL WELFARE POLICY FROM COLONIAL TIMES TO THE PRESENT 341 (1988); LINDA GORDON, PITIED BUT NOT ENTITLED: SINGLE MOTHERS AND THE HISTORY OF WELFARE 1890-1935 (1994); see also THEODORE R. MARMOR ET AL., AMERICA'S MISUNDERSTOOD WELFARE STATE: PERSISTENT MYTHS, ENDURING REALITIES 23-24 (1990) (describing the behaviorist vision of social welfare policy in which the poor are induced to behave "in a more socially acceptable manner"); cf. GERTRUDE HIMMELFARB, THE DEMORALIZATION OF SOCIETY 242-63 (1995) (defending the "moralization" of welfare policy).
  13. E.g., FRANCES FOX PIVEN & RICHARD A. CLOWARD, REGULATING THE POOR: THE FUNCTIONS OF PUBLIC WELFARE 248-340 (1st ed. 1971).
  14. For a discussion of this correlation, see Dorothy A. Brown, The Marriage Bonus/Penalty in Black and White, in TAXING AMERICA 45 (Karen B. Brown and Mary Louise Fellows eds., 1996) [hereinafter TAXING AMERICA]; John A. Powell, How Government Tax and Housing Policies Have Racially Segregated America, id. at 80.
  15. E.g., ABRAMOVITZ, supra note 11, at 181-206, 315-19; WINIFRED BELL, AID TO DEPENDENT CHILDREN 3-75 (1965); GORDON, supra note 11, at 37-64, 253-85.
  16. Historically, poor women, including mothers, have had to work for wages out of necessity. See, e.g., ALICE KESSLER-HARRIS, OUT TO WORK: A HISTORY OF WAGE-EARNING WOMEN IN THE UNITED STATES, viii, 16-19, 119-27 (1982); PIVEN & CLOWARD, supra note 12, at 3-41, 123-45.
  17. BELL, supra note 14, at 174-94.
  18. PIVEN & CLOWARD, supra note 12, at 138.
  19. See id. at 134; Bell, supra note 14, at 46 (noting that "[i]n one parish, the policy extended to children as young as 7 years of age").
  20. PIVEN & CLOWARD, supra note 12, at 134-35.
  21. BELL, supra note 14, at 46, 107, 141.
  22. ABRAMOVITZ, supra note 11, at 333.
  23. JOEL F. HANDLER & YEHESKEL HASENFELD, THE MORAL CONSTRUCTION OF POVERTY: WELFARE REFORM IN AMERICA 141-42 (WIN I), 156-58 (WIN II) (1991).
  24. ABRAMOVITZ, supra note 11, at 341 (attributing the exemption to limited funding, a lack of child care, and an excess of welfare recipients over WIN slots).
  25. HANDLER & HASENFELD, supra note 22, at 154.
  26. Family Support Act of 1998, Pub. L. No. 100-485, 102 Stat. 2343 (codified as amended at 42 U.S.C. § 602 (1988)).
  27. H.R. CONF. REP. NO. 998, 100th Cong., 2d Sess. 1 (1988). Under the Family Support Act, families were eligible for Transitional Child Care "for a period of 12 months after the last month for which the family received [AFDC]," 42 U.S.C. § 602(g)(1)(A)(iii)(1988), if they "received [AFDC] in at least 3 of 6 months immediately preceding the month in which the family became ineligible for [AFDC]." 42 U.S.C. § 602(g)(1)(A)(iv)(1988). Families contributed to the cost of child care under a sliding scale formula established by the state. 42 U.S.C. § 602(g)(1)(A)(vii)(1988). The program otherwise generally operated under the same rules as the child care program for AFDC recipients.
  28. 42 U.S.C. § 602(a)(19)(B)(i) (1988).
  29. 42 U.S.C. § 602(a)(19)(C)(iii)(II) (1988).
  30. 42 U.S.C. § 602(a)(19)(C)(iii)(I) (1988).
  31. Compare HOUSE COMM. ON WAYS AND MEANS, 104th Cong., 2d Sess., OVERVIEW OF ENTITLEMENT PROGRAMS, 1996 GREENBOOK, 404-08, tbl. 8-4 (Comm. Print 1996) [hereinafter 1996 GREENBOOK] (listing ten states exempting only those caretakers with children under age one, and three states and the Virgin Islands exempting those with children under age two) with HOUSE COMM. ON WAYS AND MEANS, 103d Cong., 2d Sess., OVERVIEW OF ENTITLEMENT PROGRAMS, 1994 GREENBOOK, 344-48, tbl. 10-4 (Comm. Print 1994) [hereinafter 1994 GREENBOOK] (listing eight states exempting only those caretakers with children under age one, and four states and the Virgin Islands exempting those with children under age two).
  32. PIVEN & CLOWARD, supra note 12, at 341, app. source tbls. 1, 4.
  33. See generally, Martha Minow, The Welfare of Single Mothers and Their Children, 26 CONN. L. REV. 817, 826-31 (1994) (rejecting the argument that "work requirements for mothers on welfare simply reflect the changing social expectations of all women").
  34. 1996 GREENBOOK, supra note 30, at 474 (reporting that in 1994, 3.3% of AFDC mothers were employed in a full-time job and 4.6% were employed in a part-time job); 1994 Greenbook, supra note 30, at 404 n.2 (reporting that in 1992, 16.1% of AFDC mothers or other caretakers were at school or training, 2.2% worked more than 30 hours per week, and 4.2% worked fewer than 30 hours per week); see also Ann L. Alstott, The Earned Income Tax Credit and the Limitations of Tax-Based Welfare Reform, 108 HARV. L. REV. 533, 546-47 n.52 (1995) (stating that studies typically show that few AFDC recipients work and citing other studies indicating that a majority of women work at some point while receiving welfare). In comparison, about 56.8% of married women whose youngest child is under six participate in the labor force. See Minow, supra note 32, at 827 n.53 (citing figures based on census data from 1987); see also Lucy A. Williams, The Ideology of Division: Behavior Modification Welfare Reform Proposals, 102 YALE L. J. 719, 745 n.173 (1992) ("Of women with children under the age of six, 64% worked at some time during the year, although only 25% worked full-time year-round.").
  35. Heidi Hartmann & Roberta Spalter-Roth, Reducing Welfare's Stigma: Policies That Build upon Commonalities Among Women, 26 CONN. L. REV. 901, 908 (1994). They found as follows:
  36. On average, mothers work in paid employment about half time, devoting the other half of the "normal" work week as well as the "second shift" to child and family care. Our research shows that about forty percent of poor mothers receiving AFDC are also working in paid employment, and they work approximately half time, about as much as all mothers. Id.

  37. Id. See also 1996 GREENBOOK, supra note 30, at 472 (citing studies).
  38. 1994 GREENBOOK, supra note 30, at 342, 789.
  39. 42 U.S.C. § 602(g)(3) (1988). See 1994 GREENBOOK, supra note 30, at 342.
  40. 1994 GREENBOOK, supra note 30, at 349, tbl. 10-5.
  41. Hearing on the Job Opportunities and Basic Skills Program: Views from Participants and State Administrators Before the Subcomm. on Human Resources of the House Comm. on Education and Labor, 103d Cong., 2d Sess. 75-76 (1994) (statement of Raymond C. Scheppach, Executive Director, National Governor's Association); see 1994 Greenbook, supra note 30, at 349, tbl. 10-5 ($646.6 million total federal funds expended on JOBS in 1993, compared to $582.5 million in JOBS-related child care). In fiscal year 1995, $855 million was expended on AFDC child care and for Transitional Child Care. 1996 GREENBOOK, supra note 30, at 658, tbl. 10-13.
  42. The program was authorized as a capped entitlement at $300 million annually. 42 U.S.C. § 603(n)(2)(B) (Supp. IV 1992).
  43. 42 U.S.C. § 602(i)(3)(A) (1988).
  44. 42 U.S.C. § § 9858, 9858b-9858n (Supp. 1997) (amending 42 U.S.C. § § 9858, 9858a-9858q (1995)).
  45. 42 U.S.C. § § 1397 (1992), 1397a-1397f (1992 & Supp. 1997) (Social Services Block Grant Program of Title XX of the Social Security Act).
  46. For a discussion of these programs, see Mary L. Heen, Welfare Reform, Child Care Costs, and Taxes: Delivering Increased Work-Related Child Care Benefits to Low-Income Families, 13 YALE L. & POL'Y REV. 173 (1995).
  47. Id. at 181-82 (stating that federal child care assistance totaled roughly $2.05 billion for fiscal year 1993, not including amounts expended on programs such as Head Start and the Child and Adult Care Food Program). See 1996 GREENBOOK, supra note 30, at 643, tbl. 10-12, 651 (showing 1995 fiscal year federal outlays of $633 million in AFDC-related child care, $192 million for Transitional Child Care, $279 million for At-Risk Child Care, and $933 million for Child Care and Development Block Grants, totaling $2.037 billion; if an estimated portion of Title XX social services block grants supporting child care were included, the total of $2.037 billion would be increased an additional $71 million, or approximately 16% of $448 million, to about $2.1 billion).
  48. See Ann Collins & Barbara Reisman, Child Care under the Family Support Act: Guarantee, Quasi-Entitlement, or Paper Promise?, 11 YALE L. & POL'Y REV. 203 (1993).
  49. U.S. General Accounting Office, Child Care: Working Poor and Welfare Recipients Face Service Gaps 4-5 (GAO/HEHS-94-87, May 1994) [hereinafter GAO Child Care Report] (based on preliminary fiscal year 1992 data reported to the Department of Health and Human Services by the states).
  50. See Impact of Welfare Reform on the Child Care System: Hearings Before the Senate Committee on Labor and Human Resources, 104th Cong., 1st Sess. 211 (1995) (statement of Sandra L. Hofferth, University of Michigan Institute for Social Research) (citing reports published in 1991 and 1992) [hereinafter Hearings]. But see 1996 GREENBOOK, supra note 30, at 647 (citing 1995 GAO report that "about three-fourths of State JOBS Programs have been able to provide child care subsidies or help arrange child care for all or most of their participants who needed such assistance," and noting that "GAO attributed this success to the relatively small number of AFDC recipients actually participating in JOBS--about 13 percent of the adult caseload in a given month").
  51. Hearings, supra note 48 (basing testimony on transitional care data from 20 states).
  52. CHILDREN'S DEFENSE FUND, THE STATE OF AMERICA'S CHILDREN YEARBOOK 42 (1995) [hereinafter 1995 YEARBOOK] (finding that eight states had at least 10,000 children on the child care assistance waiting lists); CHILDREN'S DEFENSE FUND, THE STATE OF AMERICA'S CHILDREN YEARBOOK 32 (1994) [hereinafter 1994 YEARBOOK] (finding that 31 states and the District of Columbia had waiting lists).
  53. GAO Child Care Report, supra note 47, at 15; 1994 Yearbook, supra note 50, at 34; 1995 Yearbook, supra note 50, at 41-42.
  54. 1994 Yearbook, supra note 50, at 35.
  55. ELLEN GALINSKY ET AL., THE STUDY OF CHILDREN IN FAMILY CHILD CARE AND RELATIVE CARE: HIGHLIGHTS OF FINDINGS 5, 97 (1994).
  56. Id. at 97.
  57. Id.
  58. See Personal Responsibility Act of 1996, supra note 1; see also 1996 Greenbook, supra note 30, at 434 (stating that by mid-February 1996, all but 10 states had received waivers from AFDC provisions and listing examples of waivers approved). See generally Daniel Patrick Moynihan, The Devolution Revolution, N.Y. TIMES, Aug. 6, 1995, at D15 (describing the "devolution" of federal authority to the states under proposed welfare reform legislation).
  59. 1996 GREENBOOK, supra note 30, at 1331.
  60. Alison Mitchell, Greater Poverty Toll Is Seen in Welfare Bill, But White House Says It May Be Forced to Accept Senate Measure, N.Y. TIMES, Nov. 10, 1995, at A27 (referencing studies).
  61. See Personal Responsibility Act of 1996, supra note 1 (authorizing states to use up to 30% of their TANF block grants for other purposes, including for Title XX and Child Care and Development Block Grant programs); 1996 GREENBOOK, supra note 30, at 1333.
  62. See supra text accompanying notes 26, 40, and 42.
  63. See supra note 45 (explaining computation).
  64. 1996 GREENBOOK, supra note 30, app. L at 1362-63 (explaining that discretionary funds are provided through reauthorization of CCDBG through fiscal year 2002 at an annual authorization level of $1 billion and that entitlement funds for child care are authorized at $2 billion in fiscal year 1997, $2.1 billion in 1998, $2.2 billion in 1999, $2.4 billion in 2000, $2.6 billion in 2001, and $2.7 billion in 2002; of the entitlement funds, no state match is required for about $1.2 billion each year, which is the amount provided to the states in 1995 for AFDC-related child care, Transitional Child Care, and At-Risk Child Care; and the remainder of the entitlement funds are subject to historic maintenance-of-effort and matching requirements).
  65. See 1996 GREENBOOK, supra note 30, at 636-37 (stating that "lower income families devoted 25% of their income to child care, while the higher income families spent less than 6% of their income for child care"); SANDRA L. HOFFERTH, ET AL., NATIONAL CHILD CARE SURVEY, 1990, at 119-96, 198-99 (1991) (defining child care as care provided while the mother is at work, and including care provided by fathers, mothers, and children themselves); Hearings, supra note 48, at 202 (in 1990, only 27% of the working poor paid for child care; the working poor who paid for child care spent about 33% of their incomes on child care, compared with 13% for working-class and 6% for middle-class families).
  66. GALINSKY ET AL., supra note 53, at 90.
  67. Id. at 91.
  68. COST, QUALITY & CHILD OUTCOMES STUDY TEAM, ECONOMICS DEPT., U. OF COLO. AT DENVER, ET AL., COST, QUALITY, AND CHILD OUTCOMES IN CHILD CARE CENTERS at Executive Summary 2, 5 (1995) (finding also that the average center in the study expended $95 per week per child for full-time care).
  69. For a more detailed discussion of the legislative history of the child care tax credit and its predecessors, see, for example, Alan L. Feld, Deductibility of Expenses for Child Care and Household Services: New Section 214, 27 TAX L. REV. 415 (1972); Heen, supra note 44, at 211-14; John B. Keane, Federal Income Tax Treatment of Child Care Expenses, 10 HARV J. LEGIS. 2-7 (1972); William A. Klein, Tax Deductions for Family Care Expenses, 14 BOSTON C. IND. & COM. L. REV. 917, 919-32, 936-37 (1973). See also Wendy Gerzog Shaller, Limit Deductions for Mixed Personal/Business Expenses: Curb Current Abuses and Restore Some Progressivity Into the Tax Code, 41 CATH. L. REV. 581, 606-609 (1992) (describing the § 129 exclusion for dependent care assistance).
  70. Internal Revenue Code of 1954, Pub. L. No. 591-736, 68A Stat. 70 (codified at I.R.C. § 214) (repealed 1976).
  71. See H.R. REP. NO. 1337, 83d Cong., 2d Sess. (1954), reprinted in 1954 U.S.C.C.A.N. 4019, 4055 (a detailed discussion of the technical provisions of the bill is printed in 1954 U.S.C.C.A.N. 4137, 4197-98).
  72. See S. REP. NO. 437, 92d Cong., 1st Sess. 13-14, 59-62 (1971), reprinted in 1971 U.S.C.C.A.N. 1918, 1929, 1966-68.
  73. Id. at 1928-29.
  74. Id. at 1928.
  75. Id. at 1929.
  76. I.R.C. § 44A (1976).
  77. H.R. REP. NO. 658, 94th Cong., 2d Sess., 147-48 (1976), reprinted in 1976 U.S.C.C.A.N. 2897, 3040-41; S. REP. NO. 938, 94th Cong., 2d Sess., pt. I, 132-33 (1976), reprinted in 1976 U.S.C.C.A.N. 3439, 3565-66.
  78. I.R.C. § 44A(c) (1976 & 1978).
  79. Economic Recovery Tax Act of 1981, Pub. L. No. 97-34, § 124, 95 Stat. 172, 197-201 (codified at I.R.C. § § 44A, 129).
  80. H.R. CONF. REP. NO. 215, 97th Cong., 1st Sess. 195, 200-201 (1981), reprinted in 1981 U.S.C.C.A.N. 285, 290-91.
  81. Deficit Reduction Act of 1984, Pub. L. No. 98-369, § 471(c)(1), 98 Stat. 494, 826.
  82. Omnibus Reconciliation Act of 1987, Pub. L. No. 100-203, § 10101, 101 Stat. 1330-384 (codified at I.R.C. § 21(b)(2)(A)).
  83. Family Support Act of 1988, Pub. L. No. 100-485, § 703, 102 Stat. 2343, 2426- 27; Technical and Miscellaneous Revenue Act of 1988, Pub. L. No. 100-647, § 2004, 102 Stat. 3342, 3598.
  84. Joint Committee on Taxation, Staff Estimates of Federal Tax Expenditures for Fiscal Years 1997-2001 (JCS-11-96) 19, 26 (Nov. 26, 1996), reprinted in Daily Tax Rep. (BNA) No. 229, at L-1, L-7, L-10, (Nov. 27, 1996) [hereinafter 1996 JCT Tax Expenditure Estimates].
  85. See I.R.C. § 151(d)(3) (providing for the phaseout of personal exemptions for taxpayers with adjusted gross incomes above certain threshold amounts); see, e.g., Deborah H. Schenk, Simplification for Individual Taxpayers: Problems and Proposals, 45 TAX L. REV. 121, 127-49 (1989); Zelenak, supra note 8. For a description of the child tax credit, see supra note 8.
  86. I.R.C. § 21(a)(2).
  87. I.R.C. § 21(b)(2).
  88. I.R.C. § 21(a)(1).
  89. Section 21(a) refers to "qualifying individuals" rather than children. Section 21(b)(1) defines "qualifying individuals" as including three categories of individuals: 1) a dependent under the age of 13; 2) a dependent who is physically or mentally incapable of self-care; or 3) the spouse of the taxpayer, if physically or mentally incapable of self-care. I.R.C. § § 21(a), 21(b)(1).
  90. I.R.C. § 21(c). In addition, the amount of the employment-related expenses may not exceed the lower of earned income of the taxpayer or that of the taxpayer's spouse. I.R.C. § 21(d)(1). However, if the taxpayer's spouse is a full-time student or incapable of self- care, a monthly amount of income is deemed to be earned by the spouse in the amount of $200 (if the $2,400 limit applies), or $400 (if the $4,800 limit applies). I.R.C. § 21(d)(2).
  91. I.R.C. § 21(a)(2).
  92. See Forman, supra note 7, at 686.
  93. Pub. L. No. 99-514, 100 Stat. 2085.
  94. MICHAEL J. GRAETZ & DEBORAH H. SCHENK, FEDERAL INCOME TAXATION PRINCIPLES AND POLICIES 428 (3d ed. 1995).
  95. I.R.C. § 63(c)(2).
  96. I.R.C. § 151(b), (c).
  97. I.R.C. § § 63(c)(4) (requiring inflation adjustments to the standard deduction amounts beginning after 1988), 151(d)(4) (requiring inflation adjustments to the $2,000 personal exemption amount for tax years beginning after 1989).
  98. For 1997, the inflation-adjusted standard deduction amount for a married taxpayer filing a joint return is $6,900. The inflation adjusted personal exemption amount is $2,650. Rev. Proc. 96-59, 1996-53 I.R.B. 1. Thus, a family of four claiming a standard deduction ($6,900) and four exemptions (4 x $2,650 = $10,600) would pay no tax on up to $17,500 of income.
  99. The federal poverty guideline for a family of four is $16,050 for 1997. Notice, Annual Update of the HHS Poverty Guidelines, 62 Fed. Reg. 10,856 (1997).
  100. For 1997, the inflation-adjusted standard deduction for a single head of household is $6,050. The inflation adjusted personal exemption amount is $2,650. Rev. Proc. 96-59, supra note 96, at § § 3.05, 3.09. The standard deduction ($6,050) plus two personal exemptions ($5,300) equals $11,350.
  101. The federal poverty guideline for a family of two is $10,610 for 1997. See Notice, supra note 97.
  102. Thus, even families below the poverty line would be subject to an applicable percentage of less than 30%. For example, a family with $14,500 of income would be entitled to a credit of only 27% of its eligible child care expenses.
  103. See Jonathan B. Forman, Simplification for Low-Income Taxpayers: Some Options, in TAXING AMERICA, supra note 13, at 277, 281, tbl.12.1.
  104. The Joint Committee on Taxation staff prepares estimates by income class for the child and dependent care credit. The estimates illustrate the concentration of benefits in the middle and upper income ranges. 1996 JCT Tax Expenditure Estimates, supra note 82, at 26, tbl. 3, at L-10.
  105. I.R.C. § 129(a).
  106. I.R.C. § 129(d)(1).
  107. I.R.C. § 129(d)(1)-(8) (including requirements that contributions to the plan not discriminate in favor of highly compensated employees, id. at (d)(2), and that employees be notified of the terms and availability of the program, id. at (d)(6)).
  108. I.R.C. § 129(b)(1). The same rules as are applicable to the child and dependent care tax credit apply for determining a deemed amount of earned income for a student spouse or a spouse incapable of self-care. I.R.C. § 129(b)(2) (incorporating by reference the provisions of § 21(d)(2)).
  109. I.R.C. § 129(c).
  110. 1994 GREENBOOK, supra note 30, at 708 (citing BUREAU OF LABOR STATISTICS, U.S. DEPARTMENT OF LABOR, EMPLOYEE BENEFITS IN MEDIUM AND LARGE FIRMS, 1991 (May 1993)).
  111. See id.
  112. See I.R.C. § 21(c) (providing that the amount of employment-related expenses claimed for purposes of the credit shall be reduced by the amount excludable from gross income under § 129 for the taxable year).
  113. See generally 1994 GREENBOOK, supra note 30, at 708 ("[T]he credit generally is less valuable than the exclusion for taxpayers who are above the 15 percent tax bracket.").
  114. The dollar value of the exclusion is equal to the value of the child care provided (up to a maximum of $5,000) times the taxpayer's marginal tax rate ($5,000 x 39.6% = $1,980).
  115. Under the Haig-Simons concept of income, which is frequently used in tax policy analysis, income is defined as the market value of rights exercised in personal consumption plus the net change in wealth during the taxable period. HENRY C. SIMONS, PERSONAL INCOME TAXATION 50, 140 (1938); see also ROBERT M. HAIG, The Concept of Income-- Economic and Legal Aspects, in THE FEDERAL INCOME TAX 1, 7 (Robert M. Haig ed., 1921), reprinted in READINGS IN THE ECONOMICS OF TAXATION 54 (Richard A. Musgrave & Carl S. Shoup eds., 1959).
  116. See Grace Blumberg, Sexism in the Code: A Comparative Study of Income Taxation of Working Wives and Mothers, 21 BUFF. L. REV. 49, 64-66 (1971).
  117. Personal consumption expenditures are included in the tax base by not allowing a deduction for personal living expenses. I.R.C. § 262; see Smith v. Commissioner, 40 B.T.A. 1038 (1939) (denying a business expense deduction for child care costs on the basis that child care was one of the basic functions of family living and thus was a "personal" concern), aff'd without opinion, 113 F.2d 114 (2d Cir. 1940).
  118. Under tax expenditure theory, if child care expenses represent personal consumption expenditures, they ought to be included in the tax base; accordingly, permitting a tax deduction or credit for child care expenses would constitute a tax preference. STANLEY S. SURREY & PAUL R. MCDANIEL, TAX EXPENDITURES 3 (1985) (explaining that departures from the normal tax structure are tax expenditures or special preferences and are viewed as equivalent to direct government outlays).
  119. See, e.g., Boris I. Bittker, Federal Income Taxation and the Family, 27 STAN. L. REV. 1389, 1463 (1975).
  120. E.g., Thomas D. Griffith, Theories of Personal Deductions in the Income Tax, 40 HASTINGS L. J. 343, 385-94 (1989); see also Paul R. McDaniel & James R. Repetti, Horizontal and Vertical Equity: The Musgrave/Kaplow Exchange, 1 FLA. TAX REV. 607 (1993).
  121. See Klein, supra note 67, at 937-40; Brian Wolfman, Child Care, Work, and the Federal Income Tax, 3 AM. J. TAX POL'Y 153, 167-74 (1984).
  122. See Klein, supra note 67, at 941. See generally Boris I. Bittker, Income Tax Reform in Canada: The Report of the Royal Commission on Taxation, 35 U. CHI. L. REV. 637, 638-45 (1968)(discussing the Commission's conclusion that a taxpaying unit's ability to pay taxes is measured by its discretionary economic power).
  123. MARVIN A. CHIRELSTEIN, FEDERAL INCOME TAXATION: THE LEADING CASES AND CONCEPTS 6.01(a) (7th ed. 1994); Keane, supra note 67, at 30-35.
  124. E.g., Feld, supra note 67, at 429; McCaffery, supra note 5, at 1005-10; see also Daniel C. Shaffer & Donald A. Berman, Two Cheers for the Child Care Deduction, 28 TAX L. REV. 535, 535-36 (1973).
  125. Wolfman, supra note 119, at 190-93.
  126. For a more detailed discussion of optimal income taxation, see, e.g., Joseph Bankman & Thomas Griffith, Social Welfare and the Rate Structure: A New Look at Progressive Taxation, 75 CAL. L. REV. 1905, 1919-29 (1987); McCaffery, supra note 5, at 1035-46.
  127. See A.C. PIGOU, A STUDY IN PUBLIC FINANCE (3d ed. 1947). See generally McCaffery, supra note 5, at 1046-53 (1993).
  128. Child care costs have been viewed in economic terms as an investment by parents in their children's human capital. See Lynn A. Stout, Some Thoughts on Poverty and Failure in the Market for Children's Human Capital, 81 GEO. L. J. 1945 (1993); see also Liebman, supra note 10, at 359. The view of child care costs as human capital investment has not been extensively discussed by tax theorists. Such an analysis would be problematic. The personal nature of human capital raises serious questions about whether child care may be viewed as an investment in an "asset" with any measurable payoff for parents in the form of future support from their children or any other type of income. See Bittker, supra note 117, at 1447-48 (noting that the concept of children as the "poor man's capital," or as an informal social security system, has made little headway in the analysis of American society); Klein, supra note 67, at 940 n.118. Even if such questions were satisfactorily resolved and such expenses treated as investments by parents in their children, child development expenses would not be deductible under the Code's current approach to the taxation of human capital investment. See generally Joseph M. Dodge, Taxing Human Capital Acquisition Costs--Or Why Costs of Higher Education Should Not Be Deducted or Amortized, 54 OHIO ST. L. J. 927, 948-61 (1993).
  129. R.G. Lipsey & Kelvin Lancaster, The General Theory of the Second Best, 24 REV. ECON. STUD. 11 (1956). The term "second best solution" has been used more generally to refer to solutions to problems that take into account existing imperfections and tolerate compensating imperfections. See generally Boris I. Bittker, A "Comprehensive Tax Base" As a Goal of Income Tax Reform, 80 HARV. L. REV. 925, 983-84 (1967); Dodge, supra note 126, at 941-943.
  130. See generally, e.g., Schaffer & Berman, supra note 122, at 537-43 (discussing the nonneutrality between wage work and house work).
  131. Wolfman, supra note 119, 175-81.
  132. See Schaffer & Berman, supra note 122, at 543-45 (distinguishing child care from other self-provided services because of the necessary, although not sufficient, relationship between the expense and the income earned).
  133. See Rachel Connelly, The Effect of Child Care Costs on Married Women's Labor Force Participation, 74 REV. ECON. & STAT. 83, 90 (1992).
  134. Bankman & Griffith, supra note 124, at 1925-27 (discussing studies).
  135. See Heen, supra note 44, at 210-17.
  136. For a discussion of the "fiction" of marital unity underlying the policy of taxing equal earning families equally, see Lily Kahng, Fiction in Tax, in TAXING AMERICA, supra note 13, at 25.
  137. E.g., Bittker, supra note 117, at 1395-97.
  138. See discussion infra part II.D.2.
  139. McCaffery, supra note 5, at 1015-16.
  140. Omnibus Budget Reconciliation Act of 1993, Pub. L. No. 103-66, § 13131, 107 Stat. 312, 433. The benefit amounts set forth in the text example have been adjusted for inflation as set forth in Rev. Proc. 96-59, supra note 96, at § 3.03.
  141. See I.R.C. § 32(b); see also, e.g., Alstott, supra note 33, at 541-44; Regina T. Jefferson, The Earned Income Tax Credit: Thou Goest Whither? A Critique of Existing Proposals to Reform the Earned Income Tax Credit, 68 TEMP. L. REV. 143 (1995); George K. Yin et al., Improving the Delivery of Benefits to the Working Poor: Proposals to Reform the Earned Income Tax Credit Program, 11 AM. J. TAX POL'Y 225 (1994).
  142. See Alstott, supra note 33, at 549-50, 559-64.