The Effect of the 1981 Welfare Reforms on AFDC Participation and Labor Supply

From 1992 to 1995, forty states applied for federal waivers in order to test new welfare reforms. About 80 percent of these waiver applications included expansions of earnings disregards and asset limits for welfare recipients. These changes would effectively reverse the changes imposed by the 1981 Omnibus Budget Reconciliation Act (OBRA81), which significantly restricted eligibility and reduced earnings disregards for working recipients. Hence an understanding of the effects of OBRA81 can be helpful in predicting the effects of new welfare reform proposals. This paper presents empirical estimates of the labor supply and AFDC participation effects of the individual components of OBRA81. Estimates are obtained from a discrete-choice maximum likelihood model in which female heads of household choose among six welfare/work combinations: on or off welfare together with zero, half-time, or full-time work. The paper focuses on estimation of parameters that define the utility of leisure and of welfare participation. Estimates are obtained from a sample of 2462 female heads of household from the Panel Study of Income Dynamics (PSID), covering the years 1978 to 1984. The changes imposed by OBRA81 are explicitly accounted for in the budget set, as are the decline in real benefits, changes in the federal tax system, and the interaction of AFDC and Food Stamps. Estimated utility parameters are used to decompose the individual effects of the 1981 reforms. Descriptive evidence shows that the overall effect of the legislation was to reduce participation by about 8 percent and cut the incidence of working recipiency by more than 40 percent. Simulations based on structural parameters suggest that lower real needs and payment standards reduced AFDC eligibility and participation by more than the OBRA81 changes combined. An important result is that for many recipients, the share of Food Stamps in total income increased as real AFDC benefits declined. Hence the Food Stamps program played an important role over this period in preventing the well-being of welfare recipients from eroding more than it did. Estimated utility parameters are also used to predict the effects of hypothetical policy changes. Lower payment standards cause some recipients to leave welfare and others to increase their work effort. Working recipiency is significantly encouraged by lower benefit-reduction rates, but this effect is offset by lower labor supply among women drawn on to AFDC. Finally, the AFDC participation choice is quite responsive to wage levels, but increasing wages would have only a small effect on working recipiency in the absence of higher disregards.