Alan Weil and Kenneth Finegold
Just one year after President Bill Clinton signed sweeping welfare reform legislation into law, he declared, "I think it's fair to say the debate is over. We now know that welfare reform works" (Harris and Havemann 1997). Other public officials quickly joined the chorus, and in their platforms for the 2000 presidential campaign, both major parties heralded welfare reform as a success. Congressional debate over federal welfare policy since reforms were enacted in August 1996 has been limited to discussion of the law's immigrant provisions and the level of funding for the program. Other important features of the law have gone largely unquestioned.
With five years of research on welfare to draw upon, it is time to take a dispassionate look at welfare reform. A determination of success implies well-defined goals and strong evidence that policy is furthering those goals. Given the political compromises that generated the legislation, it is not surprising that the goals expressed in welfare reform are somewhat ambiguous. Progress toward some goals is apparent, but to attribute all good news to welfare policy is to deny the importance of external factors, such as the unprecedented strength of the economy during welfare's early years, or to ignore that one of the central features of reform—time limits—has barely begun to have an effect. A review of the evidence provides a basis for both optimism and concern.
The substantial policy shift embodied in welfare reform followed years of accumulated grievances surrounding the old Aid to Families with Dependent Children (AFDC) program, which was established in 1935 as part of the Social Security Act. Yet, lurking behind broad agreement that the old welfare system had failed were highly divergent views about the reasons for the failure. Had welfare failed because it encouraged births outside of marriage, allowed single mothers to stay at home while most other mothers were out working, and created multiple generations of dependence? Because low benefits and withdrawal of those benefits as soon as recipients started working kept families trapped in poverty? Or because too many decisions best made at the state or local level were centralized in Washington?
Despite numerous perspectives on why the old system failed, early claims of reform's success were based primarily on a single factor: the rapid decline in the number of people receiving assistance. Although this simple metric is appealing, it does not capture the many possible dimensions of welfare's success or failure. Declining caseloads may or may not reflect declining need, hardship, or even eligibility for assistance, if those who are eligible are deterred or discouraged from applying. Caseload statistics say nothing about the long-term employment prospects of those who leave welfare, the well-being of children in the families welfare is intended to assist, or changes in family structure.
Since 1996, the Urban Institute's Assessing the New Federalism (ANF) project has gathered and analyzed data on policy choices made by states and on the resulting well-being of children and adults, particularly those in low-income families. The project reaches far beyond welfare policy, but many of the issues it covers speak directly to the future of welfare. This book draws heavily, although not exclusively, upon data sources developed as part of ANF (see appendix).
Congress must reauthorize funding for Temporary Assistance for Needy Families (TANF), child care subsidies, food stamps, and other programs for low-income families before October 1, 2002. Reauthorization offers a focal point for discussing welfare policy. The issues discussed in this book, however, will remain on the national agenda after that debate has come to an end.
This introduction provides an overview of the history of welfare reform, what research has revealed about the welfare system as it exists today, and how families are faring under the new system. It then presents five recommendations for improving welfare policy. Each of the chapters that follows examines an aspect of welfare reform and discusses policy implications in greater detail. Chapter 1 provides an overview of how states and localities have implemented welfare reforms. Chapters 2 and 3, on work and family structure, focus on concerns emphasized in the federal welfare reform legislation; chapters 4 and 5, on income and child well-being, focus on emerging concerns. The next three chapters examine programs to support work. Chapter 6 discusses employment programs, and chapters 7 and 8 discuss child care and health insurance programs. Chapters 9, 10, and 11 look at three particularly vulnerable subgroups: the "hard-to-serve," immigrants, and racial and ethnic minorities. Chapter 12 addresses the fiscal issues in TANF reauthorization and in the larger continuing debates about welfare policy.
A Brief History of Welfare Reform
Throughout the twentieth century, U.S. welfare policy was caught between two competing values: the desire to help those who could not help themselves, and the concern that charity would create dependency. AFDC was an entitlement program—that is, families that met state and federal criteria for cash assistance had a legal right to receive it. Within parameters set by Congress, states administered the program, set eligibility criteria, and shared costs with the federal government.
As more mothers began to work outside the home, AFDC became increasingly unpopular, and was seen as a source of financial support that allowed poor single mothers to stay at home but was not available to middle-class or married mothers. Moreover, some observers suggested that welfare, which was targeted to unmarried mothers, contributed to increases in divorce and in births outside of marriage.
Bipartisan dissatisfaction with AFDC led to periodic efforts to reform welfare by shifting its emphasis toward putting mothers to work. The Family Support Act of 1988 increased the amount of income welfare recipients could earn without losing benefits, provided transitional Medicaid and child care for those who left welfare for work, permitted states to require that teenage mothers live with their parents or other adults, strengthened child support enforcement and establishment of paternity, and expanded AFDC eligibility for two-parent families (Weaver 2000). The act also established a new work program, Job Opportunities and Basic Skills Training (JOBS), and made child care an entitlement for families participating in JOBS. In 1996, 16 percent of adult AFDC recipients were participating in JOBS (Bell with Douglas 2000).
Between 1986 and 1996, many states received waivers from the Reagan, Bush, and Clinton administrations for welfare reform experiments (Zedlewski, Holcomb, and Duke 1998; Office of the Assistant Secretary for Planning and Evaluation 1997; Teles 1998). Waivers expanded earnings disregards, transitional assistance, and two-parent eligibility, but they also included time limits, tougher work requirements, and family caps on benefits (which prohibited families already on cash assistance from receiving increased grants when additional children were born).
Welfare reform, embodied in the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA), represents a major shift in policy in several areas (see box 1 for key provisions of the law).1 The law abolished open-ended funding of welfare and replaced it with the TANF block grant, which provides each state with a fixed amount of money as long as the state meets certain conditions. It also restricted eligibility for TANF and other means-tested programs, required states to strengthen child support laws, and delinked Medicaid eligibility from welfare.
Box 1. Key Provisions of the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA)
Temporary Assistance for Needy Families (TANF)
- Purposes. Increase state flexibility; keep children in own homes or homes of other relatives; end parental dependence on government benefits by promoting job preparation, work, and marriage; discourage pregnancies outside of marriage; encourage formation and maintenance of two-parent families.
- Block grant. Replaced Aid to Families with Dependent Children (AFDC) entitlement with TANF block grant. Allocation to states based primarily on historic spending levels.
- Time limits. Federal lifetime limit of 60 months on cash assistance. Period may be shorter at state option. States may exempt up to 20 percent of recipients, and may continue benefits beyond federal time limits with own funds.
- Work requirements. Recipients required to work after two years of cash assistance. States must meet targets for percentage of cases in work activities, or face financial penalties.
- Family cap. States may deny additional benefits when children are born to families already receiving cash assistance.
Immigrants
- Restricted eligibility of noncitizen immigrants for Supplemental Security Income, food stamps.
- States may restrict eligibility of noncitizen immigrants for TANF and Medicaid.
Child Care
- Consolidated four existing programs into Child Care and Development Fund block grant.
- Eliminated the entitlements for child care for current and former welfare recipients.
- Allowed states to transfer TANF funds into CCDF, or to spend directly for child care.
- Increased CCDF funding levels.
Medicaid
- Delinked Medicaid eligibility from welfare.
- Froze state eligibility standards as of the date law was enacted.
- States may cover all members of a family using a slingle, higher income eligibility standard.
Child Support
- States required to strengthen child support enforcement.
- Eliminated requirement that states disregard first $50 of monthly child support in calculating TANF eligibility and benefits.
Food Stamps
- Reduced maximum benefit and limited deductions.
- Able-bodied adults without dependents limited to three months of benefits if not working at least 20 hours per week.
Supplemental Security Income
- Narrower standards of disability for children.
- More frequent review of disability status for children and adults.
The Reformed Welfare System
The most prominent features of the new welfare system are the time limit on benefits and the work activity requirement. An adult cannot receive federal welfare funds for more than five years over the course of his or her lifetime. In addition, federal law requires states to cut off assistance after two years of continuous receipt if the adult is not participating in a work activity. Most states require participation in such an activity in even less time.
States are free to impose additional conditions on receiving assistance. Some states have gone beyond the basic requirement of participation in job search or job training activities to require that children remain in school or that parents take classes in parenting or money management. Federal law requires states to reduce benefits for families that fail to follow program rules, and 31 states impose full family sanctions that eliminate benefits entirely in cases of noncompliance. States may refuse to accept or process a TANF application until the applicant has attempted to secure a job or assistance from another source, or they may offer one-time payments to help the family meet a specific need rather than provide ongoing assistance.
Welfare policy now focuses on work. Whereas states had a mixture of approaches to encouraging or assisting recipients to move to work before PRWORA was enacted, most of them have now adopted some variant of a work-first approach (chapter 1). This move was spurred by increasingly stringent federal standards for the share of a state's welfare recipients that must participate in federally defined work activities. More states are supplementing their work-first requirements with employment and training programs (chapter 6) or with programs to reduce barriers to work among the hard-to-employ (chapter 9). Still, the primary emphasis of welfare policy is moving recipients into paid work as quickly as possible.
Consistent with the focus on work are efforts to make work pay. Employed current and former welfare recipients are typically paid low wages and receive few benefits from their employers. New state and federal policies supplement low wages to such an extent that even a minimum-wage job, worked full-time, can form the foundation for lifting a family of three slightly above the poverty level—if the family receives all the government benefits for which it is eligible. Chief among state efforts to make work pay is the trend toward higher earnings disregards, which allow recipients to keep more of their benefits as they enter the paid labor force. A few states use TANF funds to support state-level earned income tax credits. By far the largest contribution to making work pay is the federal earned income tax credit (EITC). In addition, the federal minimum hourly wage rose from $4.25 to $5.15 in September 1997.
Recent policy changes have recast low-income programs including child care subsidies (chapter 7), health insurance programs (chapter 8), and food stamps as work supports to move and keep people off welfare. Direct federal funding for child care has increased, and states are designating some of their TANF funds for child care, as well as spending their own money on it. Similarly, the new State Children's Health Insurance Program (SCHIP) has expanded federal and state funding for publicly provided health insurance. These newly expanded work support programs combine with older work supports, such as unemployment insurance and family and medical leave.
In conjunction with supporting work, new policies deny assistance to people who do not work. The work requirements in TANF are stricter than those in AFDC, fewer recipients are exempted from them, and failure to comply with them can lead to financial sanctions. In addition, PRWORA limits nonworking able-bodied adults without dependents to three months of food stamps (Bell and Gallagher 2001), and state General Assistance programs have declined (Gallagher et al. 1999; Gallagher 1999).
States have done less to achieve welfare reform's family structure goals—encouraging marriage, reducing the number of births outside of marriage, and keeping children in their own homes or the homes of relatives—than they have to promote work (chapter 3). Many states make TANF available on a fairly equal basis to families with single and with married parents, which reverses a longstanding bias in the welfare system against serving two-parent families and thereby possibly discouraging marriage. Twenty states have adopted family caps that deny additional benefits to adults who have children while they are on welfare. Another PRWORA provision strengthens the child support system, with the federal government developing a data registry to facilitate collections from working noncustodial parents, states being required to adopt new child support enforcement tools, and individual welfare recipients facing sanctions if they fail to cooperate with the child support system.
A New Federalism
Welfare reform changed more than welfare policy; it changed the relationships among national, state, and local governments and among public and private organizations at each of these levels. State flexibility was deliberately expanded under welfare reform. While the size of the cash grant was left to state discretion under AFDC, states may now choose whom they will assist, what requirements they will impose upon those who receive aid, and what non-cash supports those families will receive.
With the expansion of state flexibility to set welfare policy has come greater diversity in state approaches to welfare. States have made different choices about how much to focus on getting welfare recipients into the labor force quickly, helping people overcome barriers to getting a job, or providing postemployment services so people can keep their jobs; who should get child care subsidies and how much they should receive; how to encourage marriage and reduce childbearing outside of marriage; whether to use state funds to replace benefits for noncitizen immigrants rendered ineligible for welfare by PRWORA; and how much of their TANF grants to spend, save, or transfer. States have also made different decisions about which of these choices should be made at the local level and thus about the extent to which flexibility and diversity should be increased within as well as among states.
Despite its stated purpose of increasing state flexibility, PRWORA also imposed new rules on states that reduce flexibility. Most significant are the federal time limits and work participation requirements. These provisions make it more difficult for states to adopt strategies that blend welfare receipt with work, training, or other activities. States also face new requirements with respect to child support enforcement and treatment of noncitizen immigrants.
Encouraging a variety of approaches to welfare has several benefits. It can lead states to experiment with welfare policies to determine what works best. Flexibility also enables states to tailor their policies to the specific needs of their populations. And state-specific welfare policies can better reflect local values and preferences (Weil 2002).
This diversity creates problems as well. Two children in identical situations in different states now live with very different realities. One may have household resources above the poverty level, stable and high-quality child care, and health insurance, while the other may have none of these. Differences in state policies existed before PRWORA, but the potential for extreme variation is now greater.
The end of the AFDC entitlement had important fiscal implications for states. An entitlement program expands or contracts with the number of people on welfare, but a block grant gives each state a fixed annual amount. The abolition of the entitlement may have also curtailed the federal courts' ability to expand benefits and eligibility, as they had done under AFDC and other social programs (U.S. Congress 1996, 267; Melnick 1994).
Evolution, Not Revolution
Some aspects of welfare policy shifted dramatically when PRWORA was enacted, but the underlying structure and administration of the welfare system have changed relatively little since 1996. States and localities quickly implemented the new program rules, especially time limits and work requirements. With their new funds, states expanded existing programs such as welfare-to-work and child care. Some states forged new links to community-based service providers, created "one-stop shops," or changed caseworker roles to emphasize job counseling over application processing. However, every state continues to operate a system of cash assistance with rule-based determination of eligibility as its administrative core. No state has replaced cash assistance with vouchers, a negative income tax, or a program of orphanage construction. Every state continues to operate TANF as if it were an entitlement, without the waiting lists that characterize other assistance programs such as housing and child care. For that matter, no state has declined to accept its federal TANF grant.
Once states implemented time limits and work requirements, they could make as few or as many changes in their welfare policies as they chose, so long as they maintained their own spending at the required levels. States had little difficulty meeting federal work participation targets as caseloads declined and the economy was strong. While Wisconsin and a few other states and localities have attracted media attention for making substantial changes in their welfare systems, most states have taken a decidedly incremental approach.
Most state and local governments have addressed PRWORA's family structure goals only at the margin, making adjustments to reduce marriage penalties or using relatively small, experimental programs to promote marriage, encourage responsible fatherhood, promote sexual abstinence, or discourage childbearing outside of marriage. Child support enforcement has become much tougher, but in this area PRWORA continued a trend well under way before 1996.
How Low-Income Families Are Faring under Welfare Reform
The most widely reported fact of welfare reform is the rapid and large decline in the number of families receiving cash assistance. Figure 1 shows that the number of recipients in 2000 was less than half of what it had been in 1996. Researchers continue to debate how much of this drop should be attributed to welfare reform and how much to low unemployment or to other policies such as the EITC or minimum wage (Bell 2001).
Less widely reported is the fact that child-only cases now account for a larger share of TANF recipients, approaching 30 percent in 1999. Figure 1 shows that while the number of child-only cases has dropped under welfare reform, it has done so less steeply and consistently than the overall caseload; hence the increased percentage of these cases. Child-only cases include children whose parents are ineligible because of sanctions, receipt of Supplemental Security Income (SSI), or immigration status, and children living with nonparent caregivers (Farrell et al. 2000). The time limits and work requirements that are central components of welfare reform do not apply to child-only cases.
Chapter 2 shows that most former welfare recipients, and almost a third of current recipients, are now employed. Both groups typically have jobs that offer low wages and few benefits. Recent data, however, show gains in wages among single mothers, especially high-school dropouts (Lerman 2001a, 2001b). Predictions that employers would be unwilling to hire welfare recipients or that there were not enough jobs in metropolitan labor markets to absorb recipients without displacing other low-skilled workers do not appear to have been borne out (see chapter 6).
Chapter 4 presents evidence that the combination of earnings and work supports has made low-income children and single mothers financially better off, in the aggregate, than they were before welfare reform. This finding stands in contrast to predictions that child poverty would increase if welfare reform were enacted. More troubling findings in chapter 4 indicate that many low-income families do not receive the benefits for which they are eligible and that a higher proportion of persons in independent single-parent families are living in extreme poverty under welfare reform.
Some trends in childbearing among teenagers have also been positive in the early years following welfare reform (chapter 3). The decline in births to teenage mothers that began in the early 1990s has continued under PRWORA, and the birthrate among unmarried teens has declined slightly.
Looking beyond the Averages
Aggregate measures of work and earnings paint a positive picture of the former welfare population. Yet the low-income population has always been heterogeneous, and portions of it are not thriving under welfare reform. Some people face substantial barriers to employment (chapter 2) and remain on welfare or cycle off and back on the rolls. In a work-based system, those who do not make it into employment do not share in the benefits provided through the EITC and the more generous earnings disregards. Cash grants remain insufficient to lift these families out of poverty. In addition, many who do find jobs lose other supports designed to help them, such as food stamps and health insurance, leaving them no better off—and sometimes worse off—than when they were not working.
Chapter 10 describes the provisions of PRWORA that explicitly restrict the access of noncitizen immigrants. These restrictions have contributed to the racial and ethnic patterns of program participation and overall well-being explored in chapter 11. Although some of the findings in this book apply across racial and ethnic groups, there are important ways in which black and Hispanic low-income families have fared differently under welfare reform than white low-income families or the low-income population as a whole.
The Well-Being of Children
Welfare is intended to benefit children, but most welfare policy is directed at the behavior of parents. Welfare reform is based on the premise that working parents provide a better environment for raising children than do parents who depend upon public assistance, yet early experimental and survey evidence suggests that the dynamics of welfare, work, and child well-being are more complicated than that (chapter 5). Parental work appears to yield better outcomes for children only when it results in additional financial resources for the family—and then only for some subgroups. Early data show that children in welfare families and in families that have left welfare are at similar risk for poor developmental outcomes, and that there have been no major shifts in well-being for either group. More time is needed to determine whether significant positive or negative changes will occur.
A Virtuous Cycle—So Far
Welfare policy has been in a virtuous cycle in the years since its enactment. The combination of work-oriented welfare policies, an exceptionally strong economy, healthy state budgets, the requirement that states maintain their spending levels, and the block grant structure that has kept the federal contribution constant even as the number of people on welfare has fallen has created a self-reinforcing system. All of the pieces work together, and they all benefit from the same set of dynamics.
Each state's TANF grant was set to cover the costs of the largest caseload that the state had had in the last few years of AFDC. Falling caseloads have resulted in more resources per person remaining on welfare and more resources to fund work supports for low-income families at risk of requiring cash assistance. Chapter 12 shows that combined state and federal spending on welfare-related programs has actually been higher under welfare reform than it would have been under AFDC, and that the federal government is picking up a larger share of the tab than it did before. This combination of circumstances has yielded large new pots of flexible funds. It would be an overstatement to say that welfare reform has accomplished what it has because policymakers threw money at the problem, but it would be naive to deny the role that extra funds have played in promoting the welfare reform agenda.
Yet this virtuous cycle could just as easily become vicious. Low unemployment and sustained economic growth have contributed to the recent decline in welfare caseloads. A recession will reverse these trends while straining state budgets. If the number of people on welfare rises, funds that would have been used to help people find work, support work, or make work pay will be needed to provide cash assistance to the neediest welfare families. Federal work participation requirements could become harder to meet, further reducing state resources. Many of the families most affected by a recession will not be eligible for unemployment insurance, and some will have already used up their lifetime limit of five years on TANF cash assistance. Federal and state contingency mechanisms for funding TANF programs during recessions or other emergencies, chapter 12 suggests, are inadequate.
The new structure of welfare may make the highs and lows of policy more extreme than they were in the past. The United States has been living through the highs; it has yet to experience the lows.
Policy Implications
Each chapter in this book describes policy implications stemming from findings in the area it covers. As the editors, we have identified five overarching recommendations that emerge from the research presented here.2
Keep Federal Policy Stable
One of the most encouraging aspects of the current welfare system is that states and localities have begun to move from relatively narrow work-first policies to more complex ways of identifying and overcoming barriers to work, providing financial and in-kind support for work, and promoting job retention and advancement. This evolution has taken time, and it has depended upon the existence of sufficient federal resources and a stable federal funding commitment.
At least as important as these state policies are local developments. In some localities, public agencies have developed new relationships with each other and with the private for-profit and nonprofit sectors. These relationships enable welfare recipients and former recipients to obtain a constellation of services and supports to assist them in their economic progress. As with evolving state policy, these relationships take time to build, and arise only when local leaders have resources, commitment, and focus.
These positive aspects of welfare reform took time to develop, even in the most forward-looking states. They will evolve more slowly in other locations. Moreover, they are sufficiently fragile that they must be nurtured everywhere if they are to become a permanent part of social welfare policy.
Major shifts in the federal welfare environment, whether in the form of budget cuts or substantial new statutory or regulatory provisions, threaten the best aspects of existing welfare policy. If states are forced to shift their attention to new topics, the creative aspects of welfare policy are the ones most likely to suffer. Therefore, the overall structure and design of federal welfare policy should remain stable for at least the next few years to enable states and localities to continue developing systems that support welfare families making the transition to work.
Retain Core Funding and Expand Funding for Work Supports
Most TANF funds are being spent in three areas: cash support for the neediest families, transfers to existing or new job search and work support programs that often serve a broader low-income population, and new services to help those who face barriers to work. Most states waited a couple of years to be sure that the number of welfare recipients would continue to decline before they began shifting funds into the second category, and spending in the third remains relatively low.
It is difficult to determine what it will cost to continue to meet needs in these areas. The number of people receiving cash assistance is no longer declining rapidly, suggesting that future needs will be equal to or higher than today's. States are continuing to identify new needs for their work support programs, reaching beyond child care to areas such as transportation. The need for intensive services designed to overcome substantial barriers such as illiteracy, substance abuse, or poor mental health is even harder to determine, as most of these programs are relatively new and their effectiveness and long-term costs and benefits are not known. Given that states are now generally spending their entire block grant to meet these varied needs, even in the context of record low welfare caseloads, the federal government should maintain existing overall funding for the TANF block grant—both the federal and the state shares—adjusted upward to account for the effects of inflation.
Combining spending on these three areas of need into a single block grant gives states the flexibility to allocate resources to their highest priority. Unfortunately, the relative costs of these endeavors vary substantially, and they vary across states because they depend upon the composition of the welfare caseload and the characteristics and strength of local labor markets. Placing funding for all three areas into a single grant forces them to compete against each other, with results that may impede the overall goals of welfare reform.
Funding for child care is simply too low to meet the needs of low-income families in a work-oriented welfare system. Almost all states have elected to spend some of their TANF money on child care, whether directly, indirectly through transfers to the Child Care and Development Fund (CCDF), or both. Yet waiting lists and other methods of rationing are widespread, suggesting that many families who meet federal eligibility standards and want child care subsidies are not getting them. In addition, funding child care out of TANF is inequitable because the block grant is based upon factors that have little to do with the demand for or cost of child care assistance. While it is appropriate that states retain the option of shifting TANF funds into the CCDF, the federal government should increase direct funding for child care.
Increase States' Flexibility to Support Working Families
Despite welfare reform's stated goal of increasing state and local flexibility, several aspects of the law constrain states' flexibility. Two prime examples are the five-year lifetime limit on benefits and the work participation requirements states must meet in order to obtain their block grant without penalty. Both provisions support the federal goals of promoting work and ensuring that cash assistance is temporary.
Unfortunately, these constraints may prevent states from adopting policies that support working families. Specifically, states that use expansive income disregards, thereby allowing welfare recipients who work to keep a larger share of their cash assistance, may find a larger share of their recipients hitting the time limit. The work participation requirements negatively affect states that impose substantial nonwork requirements on some of their welfare recipients, whether to expand recipients' earning capacity through education, to provide them with intensive services such as language skills, or to meet other requirements, such as those imposed by the child welfare system.
Research demonstrating that increased family income is good for children suggests that states wishing to provide work supports that also increase income should be permitted to do so. Therefore, federal policies with respect to time limits and work participation requirements should be modified so they do not discourage states from adopting policies that enable families to combine cash assistance with work or other activities designed to improve the likelihood the family will become economically self-sufficient.
Respond to a Weaker Economy
All of the good news about welfare reform came during an unprecedented period of sustained economic growth. It is clear now that the economic sands have shifted. When setting public policies that affect the most vulnerable families, federal and state policymakers cannot risk waiting too long to respond to an economic downturn.
In most means-tested programs, the obvious response to economic weakness is to expand appropriations to meet the needs of a larger population. For TANF, this response may be necessary, but it is not sufficient. Existing program rules, including the "Temporary" in TANF, are based on the premise that needy parents can and should move into jobs. If the pace of this movement falls, the effects will ripple throughout the program. Program costs and administrative burdens will increase, families will become more likely to exhaust their lifetime benefits, and states will find it harder to meet federal welfare participation and maintenance of effort requirements.
As an entitlement program, AFDC ensured that appropriations would automatically increase during economic downturns and that families would receive benefits as long as they needed them. TANF's block grant funding does not have these features. States and the federal government should ensure that program rules and funding respond to the many different challenges TANF will face in worse economic times. These provisions include a more effective mechanism for increasing funding when needs increase and an approach that ties individual work requirements and state performance measures to labor market conditions.
Solidify the New Federal Role
Under PRWORA, states take the lead in setting welfare policy, leaving the federal government fewer options for affecting the policies it is funding. Yet federal dominance in financing TANF, and the important national goals stated in the welfare reform legislation, suggest that a federal role remains. The government must adjust to its new role in a manner that encourages continuous improvement in welfare policy.
The federal government exercises complete or primary control over many aspects of the safety net other than TANF, through programs such as the EITC, the Food Stamp program, SSI, and Medicaid. In order for these programs to meet the needs of families, they must work together, despite the variability in TANF across states. Therefore, the federal government must ensure that the safety net programs it controls operate effectively, reach eligible populations, and are coordinated with TANF as it exists in the states.
Experimentation—as distinct from simple variation in welfare policies—can benefit the federal government as well as the states. Experimentation entails the use of analytic tools to evaluate the effects of different policies. Under AFDC, when a state wanted to try an innovative policy, one of the conditions for receiving a waiver was an evaluation of the initiative. TANF has eliminated the need for waivers, and with it the requirement for evaluation. While states conduct some evaluations on their own, their incentives to do so under PRWORA are weak, since they bear the full costs. The federal government should explicitly fund more welfare policy experiments to determine the effects of different program designs and interventions.
With better knowledge, states can make better policy—and so can the federal government. If the evidence shows that objectives such as reducing poverty or improving the well-being of children can be achieved through specific policies, the federal government may want to create incentives for states to adopt programs that further these goals. If most states use their block grants for a particular purpose, such as child care, the federal government may determine that this action reflects a need that should be protected from competition within the TANF grant and therefore funded directly. Block grants are easy to cut in tight fiscal times because they do not have a well-defined constituency. They are particularly vulnerable if spending does not conform closely with federal goals. Therefore, the federal government should monitor state policy choices and research evidence so that devolution does not result in abandonment of the federal role in achieving the national objectives of welfare policy.
The Next Act
There are reasons to be optimistic about the future of welfare, even if no federal policy changes are made. First, PRWORA has tapped into a tremendous reservoir of creativity. The combination of state and local control, new policy objectives, and, in some areas, new funding has brought new actors and ideas into the arena of welfare policy. The sense of possibility is palpable, suggesting that the entire welfare system is better situated to respond to change than the more rigid system of the past.
Second, welfare reform has transformed the central question of American welfare policy, from how much to give single mothers who do not work, to how to support work among low-income families with children. The new frame has different implications than the old one for the politics of welfare and for resource allocation decisions.
Third, dramatic increases in work among current and former welfare recipients may provide lasting benefits that are not yet visible. One premise of welfare reform was that working parents are inherently good for families, providing children with role models and parents with routines and links to the outside world, greater self-esteem, a more optimistic sense of their future, and the possibility of higher income. To the extent that this is true, welfare reform will have substantial long-term benefits for children. In addition, a period of successful employment may provide parents with a sense of hope and potential that will benefit them even in more difficult economic times.
Yet we refuse to join the bandwagon declaring welfare reform an unqualified success. The strong economy has played an indispensable role in the encouraging data on work, earnings, and poverty. Most data showing improvements in the circumstances of low-income families represent averages, while some subgroups fare worse. Welfare recipients are only starting to reach the five-year lifetime limit on benefits, which is one of the central provisions of the federal reform law. Some of welfare's most important outcomes are long-term ones that may not be observable for decades or even generations. In many respects, welfare reform simply has not yet been tested.
The first act of welfare reform demonstrated recipients' capacity for work and the value of supporting all low-income working families, regardless of whether they are receiving welfare. The next act should make this new, work-based welfare system adequate to meet the needs of a diverse group of vulnerable families through all phases of the economic cycle.
Notes
The editors thank Matthew Stagner and Sheila Zedlewski for their review of an earlier draft of this introduction; Sarah Staveteig for her help with research; and Gina Adams, Martha Burt, Kathleen Courrier, Frederick Custer, Pamela Holcomb, Harold Leibovitz, Robert Lerman, Robert Reischauer, and Martha Zaslow for their suggestions.
1. See Urban Institute (1996) for a more detailed summary of PRWORA and a comparison with the laws in effect at the time of its passage.
2. We do not purport to speak for all book authors when presenting these recommendations.
References
Administration for Children and Families. See U.S. Department of Health and Human Services, Administration for Children and Families.
Bell, Stephen H. 2001. Why Are Welfare Caseloads Falling? Washington, D.C.: Urban Institute. Assessing the New Federalism Discussion Paper 01-02.
Bell, Stephen H., with Toby Douglas. 2000. Making Sure of Where We Started: State Employment and Training Systems for Welfare Recipients on the Eve of Federal Reform. Washington, D.C.: Urban Institute. Assessing the New Federalism Occasional Paper 37.
Bell, Stephen H., and L. Jerome Gallagher. 2001. "Prime-Age Adults without Children or Disabilities: The 'Least Deserving of the Poor'—or Are They?" Washington, D.C.: Urban Institute. Assessing the New Federalism Brief B-26.
Farrell, Mary, Michael Fishman, Stephanie Laud, and Vincena Allen. 2000. Understanding the AFDC/TANF Child-Only Caseload: Policies, Composition, and Characteristics in Three States. Report prepared for U.S. Department of Health and Human Services, Assistant Secretary for Planning and Evaluation. http://aspe.hhs.gov/hsp/child-only-caseload00/index.htm.
Gallagher, L. Jerome. 1999. "A Shrinking Portion of the Safety Net: General Assistance from 1989 to 1998." Washington, D.C.: Urban Institute. Assessing the New Federalism Brief A-36.
Gallagher, L. Jerome, Cori E. Uccello, Alicia B. Pierce, and Erin B. Reidy. 1999. State General Assistance Programs 1998. Washington, D.C.: Urban Institute. Assessing the New Federalism Discussion Paper 99-01.
Harris, John F., and Judith Havemann. 1997. "Welfare Rolls Continue Sharp Decline; Percentage on Assistance Is Lowest Since 1970; Many Factors are Cited." Washington Post, August 13. A-1.
Lerman, Robert I. 2001a. "Jobs and Wages Up Sharply for Single Moms, Gains Especially High after Welfare Reform." Single Parents' Earnings Monitor. July 25.
———. 2001b. "Less Educated Single Mothers Achieved High Wage and Employment Gains in the Mid-1990s." Single Parents' Earnings Monitor. October 26.
Melnick, R. Shep. 1994. Between the Lines: Interpreting Welfare Rights. Washington, D.C.: Brookings Institution.
Office of the Assistant Secretary for Planning and Evaluation. 1997. Setting the Baseline: A Report on State Welfare Waivers. June. http://aspe.hhs.gov/hsp/isp/waiver2/title.htm.
Office of Planning, Research, and Evaluation. 2000. Characteristics and Financial Circumstances of TANF Recipients, FY 1999. http://www.acf.dhhs.gov (accessed November 15, 2001).
Teles, Steven M. 1998. Whose Welfare? AFDC and Elite Politics. Lawrence, Kans.: University Press of Kansas.
Urban Institute. 1996. A Comparison of Selected Key Provisions of the Welfare Reform Reconciliation Act of 1996 with Current Law. Washington, D.C.: Urban Institute.
U.S. Congress. 1996. House of Representatives. Personal Responsibility and Work Opportunity Reconciliation Act of 1996: Conference Report to Accompany H.R. 3734. Report 104-725. 104th cong., 2d sess., July 30. frwebgate.access.gpo.gov/cgi-bin/ getdoc.cgi?dbname=104_cong_reports&docid=f:hr725.104.pdf.
U.S. Department of Health and Human Services, Administration for Children and Families. 1988–1997. Characteristics and Financial Circumstances of AFDC Recipients. 10 vols. Washington, D.C.: U.S. Government Printing Office.
———. 2000a. TANF 1936–1999 Statistics. www.acf.dhhs.gov/news/stats/3697.htm (accessed November 15, 2001).
———. 2000b. TANF Total Number of Families, January–June 2000. www.acf.dhhs.gov/news/stats/families.htm (accessed November 15, 2001).
———. 2001a. TANF Total Number of Families and Recipients, July–September 2000. www.acf.dhhs.gov/news/stats/welfare.htm (accessed November 15, 2001).
———. 2001b. TANF Total Number of Families and Recipients, September 2000–March 2001. www.acf.dhhs.gov/news/tables.htm (accessed November 15, 2001).
Weaver, R. Kent. 2000. Ending Welfare as We Know It. Washington, D.C.: Brookings Institution Press.
Weil, Alan. 2002. "Program Redesign by States in the Wake of Welfare Reform: Making Sense of the Effects of Devolution." In For Better and For Worse: Welfare Reform and the Well-Being of Children and Families, edited by Greg Duncan and P. Lindsay Chase-Lansdale (63–80). New York: Russell Sage Foundation Press.
Zedlewski, Sheila R., Pamela A. Holcomb, and Amy-Ellen Duke. 1998. Cash Assistance in Transition: The Story of 13 States. Washington, D.C.: Urban Institute. Assessing the New Federalism Occasional Paper No. 16.
