Welfare Reform Research: What Studies Show About Effects on Families

The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 was one of the most significant changes to the American social safety net in the past half century. The law replaced the entitlement structure of Aid to Families with Dependent Children with Temporary Assistance for Needy Families, a block grant program that imposed work requirements, time limits, and federal lifetime benefit caps while giving states substantial flexibility in program design. The stated goals were to reduce dependence on welfare, promote work, and strengthen families. Research on whether the law achieved these goals and what effects it had on poor families documents a complex picture of employment gains alongside persistent poverty and increasing deep poverty for some groups. Employment among single mothers increased substantially in the years following welfare reform, driven by the combination of welfare reform's work requirements and time limits, the expansion of the Earned Income Tax Credit in the 1990s, a strong economy through the late 1990s, and changes in childcare subsidies. Research using statistical methods to disentangle these concurrent changes finds that each factor contributed to the employment increase, with the EITC expansion and strong economy contributing at least as much as welfare reform itself. The employment gains were concentrated among women with higher education levels and more work experience; those with the most significant barriers to employment saw smaller gains. Income among families who left welfare for work increased modestly compared to their welfare-dependent income, reflecting the combination of earned income and EITC benefits. Research tracking families who left welfare in the late 1990s and early 2000s found that most were employed initially but many experienced subsequent employment instability, and that a significant share returned to cash assistance or lived with incomes below the poverty threshold while not receiving assistance, a situation sometimes called disconnected from work and welfare. Deep poverty, defined as living below half the federal poverty threshold, increased for some families following welfare reform, particularly for single mothers without employment. Research by Kathryn Edin and Luke Shaefer documented growing cash poverty among families with children who were neither working nor receiving cash assistance, the population that TANF was specifically designed to serve. Their book on families living on two dollars per person per day brought research on deep poverty to a wider audience and prompted renewed examination of whether TANF was reaching the families in greatest need. TANF caseloads have declined dramatically since welfare reform, from approximately 4.5 million families receiving benefits at the program's creation in 1997 to approximately one million in recent years. This decline has sometimes been cited as evidence that the program succeeded in reducing welfare dependence. Research on caseload trends finds that the decline reflects both increased employment in strong economic periods and reduced access to benefits for families who remain poor, driven by state administrative barriers, time limits, and work requirements that exclude families who are not employed. Child wellbeing effects of welfare reform have been studied along multiple dimensions. Research on effects during the initial period of reform finds mixed results, with improvements in some outcomes including school attendance and some health measures for children in families where mothers gained stable employment, alongside concerns about the effects on young children when mothers entered low-wage employment with limited access to quality childcare. Research on longer-term effects on children who were young during the reform period is still developing. State variation in TANF program design has produced a natural experiment in welfare policy. States that implemented less restrictive work requirements, provided more generous income disregards that allowed benefit recipients to retain benefits as they began working, and invested block grant funds in comprehensive services for low-income families produced better employment and income outcomes than states that implemented more punitive approaches. Research on state TANF policies finds that program design choices matter, not just the overall structure of the reform. The Great Recession tested welfare reform's design by creating conditions under which large numbers of families who wanted to work could not find employment. TANF provided less of an automatic stabilizer than AFDC had, because the block grant structure did not automatically expand in response to increased need the way an entitlement did. Research on TANF during the recession finds that caseloads increased much less than poverty did, confirming concerns that the program could not fulfill an income support function in severe economic downturns. The research on welfare reform resists a simple verdict. Employment outcomes improved for many single mothers, and some of those improvements have persisted. But deep poverty has increased, cash assistance has become less available to the poorest families, and the program has proven poorly designed for economic downturns. Lessons from this research suggest that income support for families with children should maintain accessibility during recessions, that work incentives should complement rather than substitute for income support, and that families with the most significant barriers to employment require intensive services that time-limited assistance blocks cannot provide.