The effect of welfare reform on SSA's disability programs: design of policy evaluation and early evidence

During the past several years, the U.S. social safety net has gone through substantial changes involving an emphasis on personal responsibility and incentives, the shift of certain responsibilities to the states, and new limits on entitlements for benefits. Two pieces of recent legislation affected the Social Security Administration's (SSA's) disability programs. Section 105 of Public Law 104-121, enacted on March 29, 1996, mandated the removal of persons from the Supplemental Security Income (SSI) and Social Security Disability Insurance (DI) rolls for whom drug addiction and alcoholism (DA&A) were material to the determination of disability. It eliminated allowances on the basis of DA&A immediately and required the termination of benefits to all persons receiving benefits at the time of enactment. The other major piece of legislation was the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) of 1996, which was later amended by the Balanced Budget Act (BBA) of 1997. PRWORA converted the Aid to Families with Dependent Children (AFDC) program from an open-ended entitlement program into a block grant, Temporary Assistance for Needy Families (TANF), incorporating time limits on the receipt of benefits as well as strict work requirements. PRWORA also tightened child eligibility for SSI, narrowed eligibility for noncitizens, and reduced funding for food stamps. The BBA restored SSI eligibility for noncitizens receiving SSI prior to August 1996 and for legal noncitizens residing in the United States prior to August 1996 who become disabled in the future. SSA designed three studies to assess the effects of this legislation. Two of the studies focused on direct effects on SSA's disabled beneficiary population, targeting drug addicts and alcoholics and SSI children. The third study focused on the indirect effects of PRWORA, particularly the replacement of AFDC with TANF, on SSA's programs. The three studies were tied together by a concern of the overall effects--direct or indirect--of the legislative changes on SSA's beneficiary populations and a host of interrelated evaluation issues. The key methodological challenge of these evaluations is the nonexperimental nature of the evidence. The legislative pieces mandating the changes designed to affect SSA's target populations were implemented nationally, without prior demonstration projects. Nonexperimental strategies, such as comparison group designs, must therefore be used to measure the effects of interest. Other challenges relate to the time frames. Since implementing the changes requires a certain amount of time, and outcomes are realized over a period of time after that, the information that can be made available to interested policymakers in the short run is inherently limited to descriptive data on the populations affected and to impressionistic evidence from case studies and process study analyses. The timing problem is particularly acute with respect to measuring the indirect effects of replacing AFDC with TANF, because the most important likely effects will occur over a period of several years, and this time frame may substantially vary across states as a result of the decentralized nature of TANF programs. Finally, the analyses that can be conducted are constrained by the lack of relevant data from existing surveys. Administrative record data alleviate the need for survey information for some purposes, but the lack of survey data still seriously constrains the analyses that can be done in the short term. Because of these methodological challenges, SSA designed an evaluation strategy that uses several methods and data sources, including quantitative analyses of data from surveys and administrative records (particularly data from the Survey of Income and Program Participation, or SIPP, matched to data from administrative records) and qualitative analyses through case studies. (ABSTRACT TRUNCATED)