Last month was the eleventh anniversary of the federal welfare reform law, the Personal Responsibility and Work Opportunity Reconciliation Act of 1996, which marked a turning point in the political debate about shifting public assistance toward a system of temporary support with a concerted focus on moving recipients into the labor force. What have we learned about the effectiveness of welfare-to-work programs, and what challenges remain for those who have left welfare, as well as for those who are still on the rolls?

Americans consistently want government to lend a hand to people who cannot provide for themselves — particularly the elderly, the disabled, and children — but only if it can do so without discouraging work and promoting dependence among those who can work. The evolution of U.S. welfare policy over the twentieth century reflects this balancing act. Aside from small state and local programs, there is no cash assistance for able-bodied, working-age adults who do not have children. The main cash welfare program, Temporary Assistance for Needy Families (previously Aid to Families with Dependent Children), only serves families with children, mostly those headed by single mothers.

The Evolution of Welfare: From Supporting Widows to Requiring Work

When AFDC was created in 1935, its goal was to enable widows to raise their children at home. At that time, single mothers, along with the aged and the disabled, were not expected to work. As the population of single parents grew and became dominated by divorced and never-married mothers, and as other mothers increasingly entered the labor market, norms changed and support for the original aim of welfare waned.

Since the 1970s, welfare policy has been characterized by a tug of war between those trying to protect children from deprivation and those who believe that the best way to help poor children is to reduce their parents’ welfare dependency. Several times, notably in 1988, Congress passed laws prodding states to require welfare recipients to work or prepare for work; most states responded with incremental changes.

In the early 1990s, when the national welfare rolls rose to an all-time high of five million families, fears that welfare was discouraging both work and marriage resurfaced with new urgency. Disillusioned with previous reforms and alarmed by the surging rolls, candidate Bill Clinton promised to “end welfare as we know it.” A few years later, the Republicans who took control of Congress vowed to end welfare entirely.

The landmark welfare law that passed in 1996 had two personalities. On the one hand, it imposed time limits on benefit receipt, toughened the requirements for states to move recipients into work, and urged states to discourage out-of-wedlock childbearing and to encourage marriage. On the other hand, it gave states primary responsibility for designing and implementing welfare systems. While states had always had great discretion in running welfare programs, the new law afforded them new and ample flexibility and went as far as to end the legal entitlement to benefits for individual families. Indeed, both the rules and administration of welfare are now entirely different from state to state.

Two unprecedented trends occurred in the 1990s:

• The number of families on welfare nationwide fell by more than half.
• Labor force participation among single mothers increased dramatically.

In early 2006, Congress reauthorized the 1996 law and took steps to strengthen its work requirements — chiefly by narrowing the definitions of what “counts” as work by recipients and by increasing the proportion of each state’s welfare caseload that must be involved in work activities. States are now focused on trying to meet these new requirements.

Welfare as a “Safety Net”

Cash welfare, together with food stamps, provides a critical safety net for many families who have few other means of support. However, many families who might qualify for benefits don’t receive them, and most families who do receive benefits remain very poor, with incomes well below the federal poverty line. Moreover, welfare benefits vary widely from state to state, from less than $200 a month for a family of three in Mississippi or Tennessee, to more than $700 a month for the same size family in California or Vermont.

Welfare receipt is highly dynamic. For many recipients, welfare is a temporary support following a job loss, marital break-up, or the birth of a child. A smaller group cycles between low-wage jobs and welfare. A final group are long-term recipients suffering from a range of more complex problems; for some of them, welfare serves as a partial or temporary disability support system. Because long-term recipients cumulate on the rolls, they make up a majority of those on welfare at any point in time — even though, over time, they constitute a small fraction of those ever on welfare. Understanding how public policies affect these welfare dynamics is a key challenge facing administrators.

The Effects of Welfare-to-Work Programs

There is some evidence that welfare receipt may discourage work and marriage, but the size of these effects is probably quite small. The primary policy and research focus has been to help welfare recipients make the transition from welfare to work. Over the past three decades, as states have gradually expanded the scope and intensity of work-related requirements for welfare recipients, a series of large-scale evaluations — most using the random assignment research design familiar from trials of new medicines — have reliably assessed the impacts, costs, and benefits of these mandatory welfare-to-work programs.

Mandatory welfare-to-work programs increase employment but not necessarily income. Evaluations in the 1980s and early 1990s found that programs requiring welfare recipients to participate in work-focused activities (job search classes, adult education, skills training, and, occasionally, work-for-benefits) succeeded in increasing employment and reducing welfare receipt. The most effective programs required some recipients to look for a job immediately and placed others in brief education or training classes before searching for work (programs with a stronger emphasis on adult education were no more successful and cost more to operate). Although the increases in employment were not dramatic — usually 5 to 10 percentage points — some of the welfare-to-work programs actually saved government budgets more in reduced welfare spending than they cost to operate, a rare accomplishment for any public program. This unusually strong evidence strengthened support for continued public investments in welfare-to-work programming and shaped the 1988 and 1996 federal welfare reform laws.

Earnings disregards (or earnings supplements) increase employment and income. While acknowledging these successes, critics pointed out that participants in welfare-to-work programs generally traded a welfare check for a low-wage job and ended up no better off financially. Some lost their jobs and returned to welfare. This finding led to another series of experiments, in the mid-1990s, with programs that supplemented the earnings of recipients who took low-wage jobs. Some programs did this by changing the welfare rules to allow recipients to keep some of their benefits after going to work, while others delivered special payments to working families outside the welfare system. While they can cost more, these programs managed to increase both employment and family income simultaneously and even led to improvements in the school performance of participants’ young children. This strategy of using earnings supplements has become a mainstay of public policies designed to support low-income families: The largest antipoverty program, the federal EITC, now offers a refundable credit worth up to $4,700 per year to parents working in low-wage jobs who have two or more children.

Welfare time limits have been less significant than many predicted. A third strand of research examined some of the more radical welfare reforms of the 1990s, such as the imposition of time limits on benefit receipt. Although welfare was always time-limited, in the sense that adults could no longer receive benefits once the youngest child in the household turned 18 years old, in 1996 Congress limited most adults to 60 months of federally funded assistance in their lifetime. But there were exceptions in the federal law, and, more importantly, welfare is also funded by the states, and states have tremendous flexibility in the use of their own funds. Thus, states can set time limits that are longer or shorter than 60 months and, in fact, several of the largest states, in effect, do not have any time limits. The few rigorous studies of programs with time limits found little evidence that they led to widespread destitution, but these studies were conducted during the booming economy of the 1990s and in states that applied time limits fairly cautiously. In general, while time limits created great controversy when they were first implemented, they have turned out to be less central to the welfare reform story than many predicted at the time. On the other hand, the increased use of “full-family sanctions” (policies that close a family’s entire welfare case when the parent does not comply with work requirements) has arguably affected a much larger number of families.

Current Challenges: The Working Poor and the Hard-to-Employ

The dramatic decline in the welfare caseload during the late 1990s focused attention on two groups — former recipients who are now struggling in the labor market and those who have been left behind, some clinging to the shrunken welfare system and others who are neither working nor on welfare. These are key areas of program experimentation and research today.

Promoting Employment Retention and Advancement Among the Working Poor

For those who have left the welfare rolls, the key issue is how to help them remain employed and advance in their jobs — and how to keep them connected to the work supports for which they remain eligible. Even with the growth in publicly funded work supports (including the EITC, subsidies to help pay for child care, and health care coverage for children), surveys have shown that many former recipients struggle to maintain steady employment, fail to advance to significantly higher-paying jobs over time, and often remain poor.

Job retention. A number of recent studies have tested strategies for promoting employment stability among former welfare recipients or other low-wage working adults. Most results have not been encouraging, but there have been a few bright spots: the earnings supplementation strategy described above has been shown to promote steady work, and there is some evidence that programs run by nongovernmental, community-based organizations and those that have a direct connection to individual employers may be more successful. In the coming years, several long-term studies in the U.S and the United Kingdom will shed more light on strategies to promote job retention.

Job advancement. Given the strong link between skills and wages, strategies to help adults obtain more education or training would seem to hold the greatest promise for promoting wage advancement, but this has been difficult to accomplish — in part, because current welfare rules do not encourage long-term education and training. In addition, when welfare-to-work programs have tried to mandate participation in remedial adult education in the past, many found that recipients dropped out quickly. Many parents working in low-wage jobs lack the academic preparation needed to succeed in skills training programs, and many also have difficulty balancing the demands of full-time work, child-rearing, and training (this is particularly difficult for single-parent families, who make up a substantial fraction of the working poor). One advancement program that was not focused on training, run by a for-profit company with strong links to local employers, was modestly successful in moving employed welfare recipients from very low-paying jobs, often outside the formal labor market, to slightly higher-paying jobs. In addition, community colleges play an important role in providing education and training to low-wage workers. One recent study of a performance-based scholarship, targeted mostly to low-income, single women, showed positive effects on academic performance and retention.

The role of child care. In addition, maintaining access to work supports, especially child care, is crucial to job retention and advancement for low-wage workers. And evidence suggests that center-based care may have some benefits for young children’s later academic achievement. Although federal and state funding for child care subsidies more than doubled between 1996 and 2002, which helped to support the rapidly increasing number of single mothers entering the workforce, spending on child care has since flattened and even declined. In recent years, many states have had to restrict access to child care for low-income working families. While the 2006 welfare reauthorization added some additional funds for child care, demand still outstrips supply and the value of funding has been eroded by inflation.

Providing Services to the “Hard-to-Employ”

Although many assume that those who remain on welfare today are much more disadvantaged than their counterparts 10 or 15 years ago, studies have questioned this conclusion. Nonetheless, it is certainly true that welfare recipients with multiple, serious problems, such as depression, substance abuse, learning disabilities, and domestic violence — individuals who were typically exempted from mandatory programs in the past — are now increasingly expected to take steps toward work and self-sufficiency. Many of these individuals may be seriously impaired but cannot qualify for disability assistance programs, which require evidence of a permanent disability that leaves them unable to work. The most promising strategies for these “hard-to-employ” recipients may come from other fields, such as vocational rehabilitation for individuals with disabilities. States are experimenting with a range of strategies to assist recipients with substance abuse problems, depression and other forms of mental illness, and a range of disabilities. Results from rigorous research currently being conducted on these interventions should begin to lay an evidence base for future programming. However, it may be challenging to implement these models in the context of current welfare rules, which do not encourage lengthy activities or treatment to remove barriers to employment.

Additional Resources

For more information about MDRC’s research on welfare, programs for low-wage workers, and interventions for the hard-to-employ, visit the following sections of our site:

Welfare and Barriers to Employment
Low-Wage Workers and Communities