Amer. J. Agr. Econ. 82 (August 2000): 642–648 Copyright 2000 American Agricultural Economics Association Pre-1997TrendsinWelfareandFood AssistanceinaNationalSampleofFamilies ParkeE.Wilde,SandraHofferth,StephenStanhope,MaryNoonan, andNancyCollins Both the Food Stamp Program (FSP) and Aid to Families with Dependent Children (AFDC) saw unprecedented caseload growth from the late 1980s until 1994, followed by caseload declines in the following years. These rapid caseload declines are attributed to changing macroeconomic conditions and to substantial changes in program design. First, in the early 1990s, states applied for waivers from federal welfare program regu- lations, which allowed them more freedom to experiment with welfare reforms. Then, in 1996, Congress redesigned the federal safety net for low-income Americans through the Personal Responsibility and Work Opportu- nity Reconciliation Act (PRWORA). These real-world developments suggest some research questions. Do macroeconomic conditions affect FSP caseloads and cash wel- fare program caseloads in the same way? Do policy changes in cash programs affect par- ticipation in the FSP, which serves a partly overlapping population? Because caseload changes depend directly on the rates at which people enter and exit each program, it is use- ful in answering these questions to investi- gate participation transitions at the individual level. This paper uses data from the Panel Study of Income Dynamics (PSID) to inves- tigate participation transitions from year to year during the “waiver period,” from 1989 to 1996. The distinctive feature of the approach here is that it considers all possi- ble transitions at the individual level among three participation states—AFDC (with or without FSP), FSP Only, and Neither— rather than treating each program separately. Parke E. Wilde is with the Food and Rural Economics Division, Economic Research Service, USDA. Sandra Hofferth is with the Institute for Social Research, University of Michigan. Stephen Stanhope is with the Program in Financial Engineering and the Program in Applied Economics, University of Michigan. Mary Noonan and Nancy Collins are with the Population Studies Cen- ter, University of Michigan. This investigation provides a baseline for future work using PSID data to study the impact of the 1996 welfare reforms. Changes to the Food Stamp Program and Cash Welfare Programs In many cases, the 1996 welfare reform law implemented on a broader scale what states had already begun to implement through waiver policies. Starting in the early 1990s, states were given the option of requesting waivers to federal AFDC rules regarding eligibility and benefits.The types of cash pro- gram waivers that are central to the following analysis include Time Limits,which either ter- minated or reduced a recipient’s benefit level after a specified period, Work Requirements, which required non-exempt recipients to par- ticipate in a work activity after a specified period of time as a condition of continued benefit receipt, Expanded Earnings Disre- gard, which increased the amount of earn- ings disregarded in determining the level of the AFDC grant, Family Caps, which reduced or eliminated additional benefits for children who were conceived while the mother was receiving assistance, JOBS Exemptions,which provided either the traditional exemption for mothers of children under age six, an exemp- tion for mothers of children from six months to three years, an exemption for mothers of children from birth to six months of age, or no exemption, and Work Sanctions, which provided either the traditional (mild) sanc- tions for violations of work requirements or more severe sanctions.We consider three cat- egories of more severe sanctions, in increas- ing order of severity: partial sanctions for the first violation and for recurrent viola- tions (“partial/partial”), partial sanctions for the first violation and full sanctions for later