Series: CCF Briefs

Welfare Reform and the Work Support System

Isabel V. Sawhill and Ron Haskins

Overview of Work Support System

The 1996 welfare reform law represents a fundamental shift in how the federal government provides support to destitute families. Under pre-1996 law, low-income families were entitled to a package of welfare benefits that included cash, food stamps, and Medicaid. The American public came to believe that this system of entitlement benefits contributed to a decline in work by poor parents and an even more striking decline in the number of poor children being reared in two-parent families. Among other provisions, the 1996 reforms required work of almost every adult that joined the welfare rolls. In addition, with some exceptions, a limit of five years was placed on the receipt of cash welfare by individual families.

Far less visible than the widely debated welfare reform revolution was a second set of reforms in public policy that may be even more important in the long run. Beginning roughly in the mid-1970s with the enactment of the Earned Income Tax Credit (EITC), the federal government originated or expanded a series of programs that provide benefits to working families. Unlike welfare benefits, which are intended primarily for the destitute, these work support benefits are designed to provide cash and other benefits to working adults and their families. In addition to the EITC, the major benefits in the system include the child tax credit, the minimum wage, state income supplement programs, food stamps, health insurance, and child care. In 1999, low- and moderate-income families were eligible for $52 billion in assistance from these programs, compared to the $6 billion they would have been eligible for if these programs had not been expanded by Congress after the mid-1980s. As a result, the typical one-parent family with children was far better off working than on welfare, and employment rates among this group increased dramatically, due to the strong economy of the 1990s, welfare reform, and the availability of these expanded work supports

This evolution toward a work-based system of support progressed further as a result of state responses to the 1996 welfare law. The sharp drop in caseloads after 1994 freed up funds that states have devoted primarily to supporting work. By fiscal year 2000, only half of total federal and state spending under Temporary Assistance to Needy Families (TANF) was devoted to cash assistance, compared to 70 percent in fiscal 1995, according to Gene Falk of the Congressional Research Service. However, if a recession-induced increase in caseloads requires states to reallocate these funds to pay basic benefits, these investments will almost certainly decline.

The value of these new work support programs at both the federal and state level cannot be overemphasized. The EITC alone provides roughly $4,000 a year in extra benefits to a low-wage worker with two or more children, and the children remain eligible for Medicaid. The average woman leaving welfare earns about $7 an hour, or $13,000 in after-tax income. The combined value of food stamps and the EITC, then, brings her total income up to about $19,000—enough to boost a single parent family with three or fewer children above the federal poverty line (the poverty threshold for a family of four was about $18,000 in 2001).

Notwithstanding the expansion of work supports in recent years, advocates for the poor point to the low wages earned by many adults and believe that the next phase of welfare reform should be devoted to ensuring that jobs are available and work more adequately rewarded. Polls show that the public is willing to do more for those who work. Two-thirds of the electorate, including 71 percent of Democrats and 63 percent of Republicans, say it is very important for President Bush and the Congress to do more to help those trying to work their way off welfare, according to a poll conducted last spring by Peter D. Hart Associates. Voters rank this goal just below other major concerns such as providing prescription drug coverage for seniors and improving education.

The ability to make ends meet is especially serious for low-income families who must pay for child care and other work-related expenses, and who have no access to subsidized health insurance through an employer. Child care costs average $2,000 a year for the 40 percent of working poor families that pay for care, according to a Brookings analysis of Census Bureau data. Because health insurance can easily cost $6,000 a year or more, most adults in low-income families without employer-based health coverage remain uninsured, although most of the children are covered by federal programs. Current measures of poverty fail to incorporate these realities. Thus, many are advocating for an expanded definition of poverty and a more generous set of supports for low-income working Americans. These supports could include a higher minimum wage, additional income supplements, greater access to subsidized child care, more health care and job training, and a stronger safety net of community service jobs for those unable to find work in the private sector.

The purpose of this policy brief is to provide basic information about the current work support system and to discuss ways in which it might be expanded. A commonly advanced set of policy proposals that would help low-income working families, along with their advantages and disadvantages, is summarized in Table 1 in the last pages of this brief. Many of these policies respond to complaints that the 1996 welfare law placed too much emphasis on reducing caseloads and not enough on reducing poverty.

Goals of Work Support System

The work support system serves three primary goals. First, it provides incentives for work. Under the pre-1996 welfare system, able-bodied adults who did not work were given benefits, but these benefits were often reduced dollar-for-dollar as earnings increased, leaving adults no better off financially after they went to work. Research now shows that increasing the incentive to work through programs such as the EITC contributes to large increases in employment among less skilled workers. Still other research shows that programs that combine work requirements with financial incentives can improve educational and other outcomes for children because these incentives raise income beyond what is available from either welfare or work alone.

A second goal of the work support system is to help ensure that parents working at low-wage jobs have enough total income to provide an adequate standard of living. In the past, many low-skilled workers, especially men, were able to find reasonably well-paid jobs in manufacturing. However, in the post-industrial economy, many jobs require high levels of education and far fewer jobs provide good wages for workers with limited education and training. In the long run, the solution to this problem is to improve the nation’s education system to equip young people with the job skills needed in the new economy. Another long-run strategy is to increase the proportion of children growing up in families where there are two parents who can share bread-winning and child care responsibilities. But in the short run, and especially for those single parents who have already completed their education and need to support a family, supplementing the low earnings of the least skilled may be the only feasible response—and is a better and more popular approach than expanding welfare.

The third goal of the work support system is to insure that those who lose their jobs or cannot find work will not be destitute. Although this was not a major issue in the late 1990s when the demand for workers was high, it could be a bigger problem during a recession or a prolonged slowdown in the economy. A number of current programs address one or more of these three objectives.

Reforming Work Support Programs

Minimum Wage

The current minimum wage of $5.15 an hour has not been raised since 1997 and leaves a family of three with one full-time worker below the poverty line. This has led to proposals in Congress to raise the minimum wage by $1.00 or $1.50 and to index it for inflation. These proposals spark heated debate, with liberals generally arguing that a higher minimum would put a floor under the incomes of low-wage workers and conservatives often arguing that it would be too costly for business and might reduce employment opportunities for the least skilled.

The minimum wage is not very well-targeted. Only one quarter of minimum-wage earners live in poor families. Many teenagers or others in higher income families earn the minimum. At the same time, as shown in research by Isabel Sawhill and Adam Thomas of The Brookings Institution, over 60 percent of wage earners in poor families would benefit from a $1.00 increase in the minimum wage because they are currently earning less than $6.15 an hour. The same study also suggests that, even if one makes a relatively strong assumption about the number of jobs that would be lost as a result of a minimum-wage increase, a $1.00 boost would still lift almost one million people out of poverty.

Earned Income Tax Credit (EITC)

Enacted in 1975 primarily as a way to offset the payroll taxes paid by low-wage workers, the EITC now provides a 40 percent cash supplement for every dollar of earnings up to about $10,000 for families with two or more children. Unlike some other tax credits, the EITC is refundable—meaning that families with little or no income tax liability get a check from the Treasury. The maximum benefit of $4,000 remains flat up to earnings of a little more than $13,000 and then phases out at the rate of around 20 cents for every dollar of earnings above $13,000. The supplement is completely gone when earnings reach about $32,000. By 2000, the federal EITC was providing over $30 billion in cash supplements to working families, making it the biggest program other than Medicaid and Supplemental Security Income that provides benefits to low-income families. And unlike nearly every other program for low-income families, it provides benefits only to families that work. It is, in short, the quintessential work support program.


Child Tax Credit

Prior to 2001, the child tax credit provided few benefits to lower-income families because it was not refundable. But the 2001 tax bill not only expanded the credit from $500 to $1,000 per child but also made it partially refundable for families with modest amounts of earned income and little or no income tax liability.

The credit provides important assistance to low-income working families but is also very complicated. It could be both simplified and better integrated with the EITC. One option would be to eliminate the child tax credit and create instead a second, and more generous, benefit tier in the EITC available to families that work full-time (as proxied by their having earnings above $10,000 a year). A two-child family with full-time earnings of less than $20,000 a year might qualify for a $6,000 EITC, phasing down to a flat $1,200 ($600 per child) at an income of $44,000 a year. Research done at the Manpower Demonstration Research Corporation suggests that conditioning benefits on full-time work creates a potent incentive for families to work and earn more with few net costs to the government. Moreover, this type of two-tiered working family tax credit has operated quite successfully in Great Britain. However, unless offset by savings from the expansion of the child tax credit to higher income families (not yet phased in), this proposal would be very expensive.

State Income Supplements

Not all the improvements in the work system have come at the federal level. States have taken two major approaches to improving work incentives. First, since enactment of the 1996 reforms, nearly every state has allowed parents who find jobs to retain more of their welfare benefit. This policy enables many families to work and continue receiving earnings supplements from welfare. These “earned income disregards” vary in duration and generosity. In California, for example, families that go to work can keep $225 per month plus 50 percent of earnings over $225 before their welfare benefit is reduced. The disadvantage of generous rules like this is that they discriminate against low-income families that have never been on welfare. Also, under current federal rules, working families can exhaust their five-year limit on welfare while receiving just a small supplement to their earnings. For this reason, time limits may actually discourage work, and have led to proposals to “stop the clock” on the five year time limit for those who are working a certain number of hours but still receiving some welfare.

A second approach states have followed is to create their own EITC programs. These programs, now available in sixteen states, typically supplement the federal EITC by adding a fixed percentage to whatever is due the family under federal rules. The amount of state supplementation varies from 4 to 25 percent of the federal benefit. However, not all of the state EITCs are refundable, and nineteen states still tax the incomes of families below the poverty line. One way to provide more assistance to low-income working families would be to provide a federal incentive for states to expand their EITCs. The incentive would be a federal matching rate for state EITC payments similar to that in the Medicaid program; states that have high per capita incomes (and hence a bigger tax base) would get a smaller match than states with lower per capita incomes.

Still another approach the federal government could take to encourage work would be to replace the current caseload reduction credit with an employment credit. Under the caseload reduction credit, states are allowed to fulfill their TANF mandatory work requirement by reducing their TANF caseload rather than by placing adults on welfare in actual jobs or in work programs. The employment credit would be designed to encourage states to move people into jobs and not just off the rolls. Such a credit, however, would be administratively complex.

Food Stamps

Although not a program well targeted to the working poor, the rules governing food stamp eligibility ensure that families of three earning up to around $19,000 remain eligible for some benefits. Thus, nearly all the families leaving welfare are eligible for food stamps. In a typical situation, with a mother of two earning $14,000 per year, the family would be eligible for about $1,000 in food stamps, a major income boost.

Unfortunately, the food stamps program has a number of serious deficiencies in the way it is administered. Research conducted by Sheila Zedlewski and her colleagues at the Urban Institute in Washington, D.C. shows that less than half the families leaving welfare receive the food stamp benefits to which they are entitled. If the administrative problems that contribute to such low participation rates can be reduced, food stamps could take their place alongside the EITC as a benefit of considerable value to working families. Possible reforms include less emphasis on error rates, less frequent redeterminations of eligibility for working families, and presumptive eligibility for some period of time for those leaving welfare for work.

Medicaid and State Child Health Insurance Program (SCHIP)

A major flaw in the original Medicaid program, enacted in 1965, was that the only way families could qualify for coverage was to join either the Aid to Families with Dependent Children program or the Supplemental Security Income program. Confining Medicaid coverage to welfare beneficiaries was a classic case of building perverse incentives into the nation’s welfare system. Thus in 1984 Congress embarked on a series of reforms that broadened Medicaid coverage for children, including those not on welfare. Health insurance for children was expanded still further through enactment of the SCHIP program in 1997. States are now required to cover all poor children under the age of 19, and most states are providing coverage to children in families with incomes under 200 percent of poverty ($29,000 for a family of three in 2001). Even so, according to the Urban Institute, 23 percent of children in families below 200 percent of poverty remained uninsured in 1999. In addition, state laws vary enormously and families, faced with major hurdles in establishing and maintaining eligibility, often drop out of the system. Although mothers are covered for up to a year after leaving welfare in most states, government health insurance coverage for adults is much narrower than that for children. And only about one quarter of those leaving welfare for work have health coverage through an employer.

There are several ways in which this system could be improved. One would be to cover the parents of eligible children. An Urban Institute study reports that 37 percent of low-income children with public coverage in 1999 had a parent who was uninsured. Another option would be to extend coverage to still more children through either Medicaid or SCHIP. Most low-income working families with incomes above the poverty line but below, say, 200 percent of poverty, find it difficult to afford health insurance. The result is that unless they have coverage through an employer, too many become part of the uninsured population. Reluctance to extend health insurance to this group has foundered on the high cost and disagreements about the best way to do so.

Child Care

Especially for mothers with young children, child care is a vital work support. As a result, the federal government has a long history of enacting legislation to support child care. The basic outlines of current federal child care policy are as follows. First, the federal government provides states with major funding (almost $4.6 billion) in the form of a block grant to help low- and moderate-income working families pay for child care. States also use around $4 billion in TANF dollars to subsidize child care. Although they must ensure that parents have choices in their selection of child care types and facilities, states have tremendous flexibility in the use of federal child care dollars. Second, the federal government does not regulate child care. Rather, responsibility for the quality of care is left to parents and to state and local government. Third, the federal government provides child care subsidies to low-income working and middle-class families through the tax code. However, because these child care tax credits are not refundable, families with no or little income tax liability lose all or part of the credit and most of the benefits accrue to relatively well-off families. Fourth, Head Start and a few other programs provide early education and developmental services to many of the children whose mothers are likely to be on welfare. However, because these programs are not usually full-day or full-year, they do not fully meet these mothers’ need for child care while they work.

The 1996 welfare reform legislation boosted funding for the child care block grant by around $4.5 billion over 6 years. In addition, states were given authority to spend an unlimited amount of money for child care from their annual share of the $16.5 billion TANF block grant. Largely as a result of these provisions, total federal spending on child care, Head Start, and other child development programs has increased from $9 billion in 1993 to over $20 billion in 2001. State spending on child care has probably increased as well.

Nonetheless, a widely cited Department of Health and Human Services study shows that only 12 percent of children potentially eligible under federal guidelines are receiving subsidies through the child care block grant. These guidelines permit families with incomes of up to 85 percent of a state’s median (median family income in the United States was $51,000 in 2000) to receive state child care subsidies. However, it was not Congress’ intent to make all of these families eligible and most states have established somewhat lower income eligibility limits.

Other studies suggest that current funding is adequate to provide subsidized care for all families leaving welfare who need it, but many families have difficulty accessing the benefits for which they are eligible and only about a third of mothers leaving welfare receive subsidized care. Equally important, research suggests that there is not enough funding to serve all of the working poor, especially those who have never been on welfare. Some states, such as Illinois, have sought to extend child care assistance to this group. Waiting lists exist in some states and child development experts are concerned about the quality of available care. If every state were to provide as much assistance to the working poor as Illinois now does, funding for child care would need to increase by about 50 percent, according to a study by Jean Layzer and Ann Collins conducted at Abt Associates in Cambridge. But even this level of funding would provide little room for quality improvements. For these and other reasons, proposals to expand funding for the child care block grant are likely to be considered during the reauthorization debate.

Child Support Enforcement

Child support enforcement is a federal-state program that attempts to collect money from parents who do not live with their children. There are now over 50,000 child support caseworkers in the U.S. who, thanks to sweeping reforms enacted as part of the 1996 welfare reform law, have numerous collection mechanisms and information systems at their finger tips. In the last decade, child support collections nationwide have nearly doubled to about $18 billion.

Child support payments are potentially a major support for struggling single mothers and their children. If a mother of two earning $10,000 received even the modest sum of $2,000 in child support, her total income including EITC, food stamps, and child support would be $18,000. Unfortunately, data from the Census Bureau show that only about one quarter of single mothers with total incomes below $23,000 received child support in 2000 and the average amount they received was only $620. On the other hand, the mothers who actually received child support in 2000 got almost $2,600, a considerable sum to these families. It does appear that both the percentage of families receiving child support and the amount of money they receive are creeping up, although the pace of improvement is slow.

Even so, a realistic assessment of the role of child support in supporting low-income single mothers requires us to have modest expectations. The program is improving and the help provided to mothers who actually receive payments is substantial. But future improvement is constrained by the fact that many of the fathers of poor mothers have limited income, especially when they are young. Even so, the nation should continue its current course of aggressive improvement in the child support program. The frequency of paternity establishment, which more than doubled between 1994 and 2000, is one of the great successes of social policy in recent decades and implies that the program can expect to continue its current path of modest improvement. One policy that would lead to instant improvement in the financial status of single mothers is reversing the current practice of government retention of some child support payments to mothers who spent time on welfare. Approximately half the money collected on overdue child support owed to mothers who have left welfare is retained by states as an offset for welfare payments. If Congress provided financial incentives for states to give all this money to mothers, the income of these mothers could be increased by as much as $1 billion per year.

Education and Training

Greater access to education and training would seem to be an obvious solution to the low wages earned by less skilled workers. For this reason, the pre-1996 welfare system stressed the importance of helping recipients acquire skills before taking a job. In contrast, the new law stresses “work first” and limits access to skill-building programs among those still on welfare.

This new emphasis is based on research, such as a recent comprehensive study by the Manpower Demonstration Research Corporation, suggesting that “work first” is a more cost effective approach to increasing employment and earnings. In addition, welfare leavers have the same opportunities to access community colleges, tuition assistance through Pell grants, and other forms of training as the rest of the low-income population. However, some liberalization of the amount of education and training that can be counted toward a state’s work participation requirement might enable more mothers on welfare to gain the skills they need to get better jobs with higher pay. This approach might be especially appropriate for mothers returning to the welfare rolls because they have been laid off from their jobs during a recession. This group is likely to have a better understanding of the world of work, to be highly motivated to find work in another sector of the economy, and to have “earned” the right to upgrade their skills.

Not all education and training programs are effective. But programs that are closely aligned to the needs of employers, that use existing institutions such as community colleges, and that train for jobs in high growth sectors such as health care could probably help families move up the occupational ladder. Calls for more state flexibility in the use of TANF funds for such purposes, and especially for demonstration programs, are likely to be an important part of the reauthorization debate.

Streamlining the Process

There are a variety of other support programs that low-income working families can access, including housing assistance, transportation assistance, and several child nutrition programs. Indeed, one problem for families is that there are a multitude of programs, all with somewhat different eligibility rules and administrative systems. Finding the time to apply, or reapply, for all of these different forms of assistance can be an exercise in frustration for an employed parent trying to balance work and care of children, especially if the benefits are uncertain or small. The result is that many families simply give up and fail to receive benefits for which they are eligible.

A possible solution is to establish a single application process for as many of these benefits as possible, to allow families to apply at times and places consistent with their work obligations, and to extend eligibility certification periods for those in regular jobs. If a single application for the EITC, the child tax credit, food stamps, Medicaid, and a child care voucher or tax credit could be established, it would go a long way toward solving the problems these families experience with bureaucratic hurdles. It would also make more visible a troubling feature of the entire system: as earnings increase these benefits disappear at a rapid rate, thereby undermining one of the goals of a system that is supposed to reward work. Unfortunately, there are no easy solutions to this problem, since making benefit reduction rates less steep would be very costly to the federal budget.

When Work Disappears: Unemployment Insurance, a Contingency Fund, and Community Service Jobs

Before welfare was reformed in 1996, the prevailing assumption was that low rates of employment among less educated mothers reflected, to a large degree, a dearth of jobs for which they qualified. But the experience of the late 1990s proved that even low-skilled individuals can, if pushed by the welfare system, pulled by the work support system, and buoyed by a strong economy, find work and increase their earnings. Employment rates among women with less than a high school degree, for example, increased from 33 percent to 53 percent between 1994 and 2001, according to the Urban Institute.

But there will always be some adults for whom finding a private-sector job is difficult and the number of such people invariably increases substantially during an economic downturn. Adults with an adequate work history who have been laid off (rather than quit their job) and who want to work full-time qualify for unemployment insurance. Research by Harry Holzer of Georgetown University suggests that 30 to 40 percent of welfare leavers qualify and might be eligible for benefits of around $400 a month. Proposals have been made to broaden coverage by including the most recent quarter of work in the base period earnings calculation; to include those seeking part-time as well as full-time work; to make the weekly benefit more generous; and to extend benefits from the normal 26 weeks to 39 weeks. If enacted, these reforms would increase the proportion of newly employed welfare mothers eligible for unemployment insurance. Even so, many mothers would remain ineligible, mainly because they often voluntarily leave rather than lose their jobs. In addition, the vast majority of adults who have left welfare since 1996 have not exhausted their five-year time limit and thus would be eligible to return to the welfare rolls.

Also worrisome is the possibility that fiscally-strapped states will not have sufficient funds during a recession to pay for both rising caseloads and continued work supports. Without some encouragement and assistance from the federal government, states are likely to cut back on existing work support services, such as child care, and channel the funds into paying for cash assistance. The progress that has been made over the past five years in linking many of the welfare poor to jobs could be threatened. To avoid this outcome, the federal government needs, at a minimum, to maintain existing TANF funding and may want to provide a cyclically based contingency fund to the states. A contingency fund was provided in the 1996 law but it expired at the end of fiscal 2001. Some states have been able to save a portion of their TANF block grant and can draw down these rainy day funds to pay for rising caseloads. But others have exhausted these surpluses, responding in part to congressional prompting that they should use them or lose them.

Still another possibility is that the economy will remain somewhat depressed for a lengthy period and fail to replicate the very low unemployment conditions of the late 1990s. In this case, states may want to provide community service jobs for those unable to find work in the private sector. In the absence of such programs, it will be hard for states to enforce existing work requirements and time limits on welfare. The availability of community service jobs is not only the ultimate safety net but helps to discriminate between those who really want to work and those who use the perceived lack of jobs as a reason to stay home. So far only a few states and communities have felt the need to provide jobs of last resort for those unable to find jobs in the private sector.

Summing Up

The reform of the welfare system in 1996 has tended to overshadow equally important reforms in the work support system over the past decade and a half. Not only has the federal government expanded its support—especially for the EITC, Medicaid, and child care—but the states have used the funds freed up by the decline in their welfare caseloads to in