Optimizing JOBS: Evaluation versus Administration

Abstract

The nation has been reforming welfare for over 30 years. Welfare here means Aid to Families with Dependent Children (AFDC), the principal federal-state family assistance program. AFDC has been controversial since the late 1960s and early 1970s, when its caseload suddenly mushroomed from five million to 11 million persons. After changing little for 15 years, the rolls jumped again, from 11 to 14 million, between 1989 and 1993. That made welfare reform an issue in presidential campaigns. The proportion of adults on welfare who work is much lower than the proportion of people off welfare who work. The public, according to opinion studies, wants to demand work of the welfare adults while protecting vulnerable children. Promoting work within the welfare system, rather than just cutting back aid, meets both of those objectives. Accordingly, the leading approach to welfare reform has been to change welfare benefits or requirements so that more adult recipients will go to work. Several times, Congress has created employment programs intended to move employable recipients into jobs. The latest and most ambitious of these is the Jobs Opportunities and Basic Skills Training Program (JOBS) created by the Family Support Act of 1988, the last federal welfare legislation. JOBS can require adults on welfare to participate in work or training as a condition of eligibility for aid. However, neither the Family Support Act nor its regulations give localities much guidance about how to run JOBS. Primarily, the act mandated that states involve increasing numbers of the employable recipients in the program (roughly, welfare parents with children at least three years old and otherwise not required in the home), or face reductions in federal funding. Twenty percent of the employable recipients had to be active in JOBS on a monthly basis by 1995, a level that sounds low but was much higher than the level under previous law.(1) Some states strained to implement JOBS this fully. Because of their own limited funding, states have been unable to claim all the federal matching funds available for JOBS (Hagen and Lurie, 1994). The Family Support Act, however, took no position on the two key policy issues that have emerged from research on welfare employment programs. First, should programs be voluntary, or should they mandate participation by those eligible? Second, should participation stress the education or training of clients for good jobs in the future, or should it emphasize immediate employment in available jobs, even if they are low-wage jobs? The act required participation only implicitly in chat it allowed states to sanction recipients, chat is, reduce their grants for nonparticipation. In practice, most states found they had to enforce involvement to meet the new participation thresholds. The act also equivocated about immediate work. It permitted states to assign clients to unpaid government jobs (workfare), but it also favored education over job search for young mothers without high school diplomas. Normally, the goals of employment programs are based on performance measures defined by Washington to hold local programs accountable. The Family Support Act left the drafting of performance measures for JOBS to the U.S. Department of Health and Human Services (HHS), but no measures have emerged. The department was told to finish the task by 1993, but it has put off the deadline until 1998. Some reasons for the delay were political. Development of the measures was disrupted by Bush's defeat in 1992. The priority of the new Clinton administration was to draft its own welfare reform plan, not to craft indicators for JOBS. Health and Human Services was held up, as well, by disputes among federal and local officials about what purposes the new measures should serve (U.S. Department of Health and Human Services, 1994). Most states criticize federal planners for failure to define indicators, but most also want flexibility in setting norms for themselves (U. …