Research Evidence on the Impact of Work Requirements in Need-Tested Programs

Summary

Congress is debating work requirements for recipients in programs providing need-tested assistance to low-income families and individuals. Legislation before the 115th Congress—the House-passed version of H.R. 2—would expand work requirements in the Supplemental Nutrition Assistance Program (SNAP). H.R. 5861, reported to the House from the Ways and Means Committee, would alter some of the Temporary Assistance for Needy Families (TANF) program’s rules regarding work. In addition, the Trump Administration is currently granting demonstration waivers for states to implement work requirements for certain recipients of Medicaid, and it has proposed expanded work requirements for tenants in assisted housing.

Work requirements for recipients of government assistance seek to achieve a variety of policy goals, including promoting work, reducing assistance caseloads, and improving the economic status of individuals and families. What does the research evidence indicate about the impact of work requirements? The impact of a policy is the difference it makes. Most of the research that addresses work requirements and need-tested assistance comes from a set of experiments, conducted prior to the 1996 welfare reform law, on alternative approaches to the work and education provisions in TANF’s predecessor program, Aid to Families with Dependent Children (AFDC).

The pre-1996 welfare-to-work experiments provided evidence that, for mostly nonworking single mothers, work requirements combined with employment and education services could increase employment, increase wages, and reduce assistance payments. Such programs—without supplementing earnings—did not raise incomes. A subset of experiments on programs that combined work requirements, employment and education services, and continued earnings supplements found that some programs increased incomes, but when they did, they did not reduce the amount of government assistance provided to the family.

However, the findings from the pre-1996 experiments cannot be automatically applied to TANF. TANF was implemented differently than the pre-1996 experiments.

As Congress debates work requirements in SNAP, Medicaid, and housing assistance, there is no large accumulated research base to draw from that applies to these three programs. Given the differences in populations, presence of those in the programs who are already working, goals, and funding structures for employment and education services, the findings of the pre-1996 welfare-to-work experiments cannot be directly applied to the current debate.

Different populations. These three programs serve a broader population (including men, single persons, and childless couples) than did the evaluated pre-1996 experiments (mostly nonworking single mothers), with potentially different challenges.

Different goals. SNAP, Medicaid, and housing assistance include populations that are already working. Applying work requirements to recipients who are working implies different policy goals than that of moving nonworking recipients into work. The goal of the Medicaid waivers that include work requirements is to improve the health status of recipients, a different purpose, with different outcomes, than the goals tested in the pre-1996 experiments.

Different approaches to employment and education services. The pre-1996 experiments were of mandatory welfare–to-work programs, which combined a participation requirement with a program that funded employment or education services. Among the SNAP, Medicaid, and housing programs, only SNAP has an employment and training program.