You are here

Evidence Doesn’t Support Claims of Success of TANF Work Requirements

April 3, 2018

Those who argue that work requirements in the Temporary Assistance for Needy Families (TANF) program have been a major success often cite rigorous random-assignment studies — the gold standard for assessing a program’s effectiveness — of mandatory work programs conducted before TANF’s creation in 1996 or during its early years.  But these claims usually don’t account for these studies’ full findings, or for changes that have occurred since TANF’s early years.  In addition, a recent descriptive study of the employment and earnings of Kansas parents who left TANF due to work sanctions shows how poorly they fare over time.  While this descriptive study cannot tell us how recipients would have fared without work sanctions, it provides detailed information on how much they worked, and how little they earned, even four years later.

  • Employment increases among recipients subject to work requirements were modest at best and faded over time.  Rigorous studies of work requirements implemented under TANF’s predecessor (Aid to Families with Dependent Children) and during TANF’s early years found that employment for those facing work requirements rose modestly in the first two years after leaving the program, but by the fifth year, at least three-quarters of former recipients worked regardless of whether they faced work requirements.

    Work effort among Kansas parents who left TANF due to work sanctions was just as high as in those earlier studies: 84 percent worked in the quarter in which they exited the program or in the year before or after their exit quarter.  Moreover, the share with any employment was virtually the same in the year before exit (67 percent) as the year after exit (68 percent).

  • Stable employment among recipients subject to work requirements proved the exception, not the norm.  Studies consistently show that although work is common among TANF recipients, they often work in low-wage jobs where employment is unstable.  In the early rigorous studies, the share of recipients subject to work requirements who worked stably (defined as working in 75 percent of the calendar quarters in years three through five after the study began) ranged from just 22 percent to 41 percent.

    In the recent Kansas study, the share of work-sanctioned TANF leavers working in any single quarter was much lower than the share that worked at any point in the year before or after exit.  In any given quarter before or after exiting TANF, more than half of work-sanctioned TANF recipients had no earnings.

  • The large majority of individuals subject to work requirements remained poor, and some became poorer.  In the early rigorous studies, recipients’ work rates rose in the first two years after becoming subject to work requirements, but their earnings were not sufficient to lift them out of poverty.  In some programs, the share of families living in deep poverty (i.e., at or below 50 percent of the federal poverty level) increased.  (These studies based poverty and deep-poverty status on an income measure that includes earnings, cash assistance, and SNAP benefits.)

    The findings from the recent Kansas study are especially sobering.  In the fourth year after exiting TANF due to work sanctions, nearly 40 percent of parents had no earnings and another 30 percent had earnings below the deep-poverty level.  Their median annual earnings were just $2,175, or about 11 percent of poverty.

  • Most recipients with significant barriers to employment never found work even after participating in work programs that were otherwise deemed effective.  Many people turn to public assistance due to significant personal or family challenges that limit their ability to work or to compete for the limited supply of jobs.  A rigorous two-year study of New York City’s PRIDE (Personal Roads to Individual Development and Employment) program — a comprehensive mandatory assessment, work experience, and job search program for TANF recipients with significant employment barriers — found that most never found jobs.  Just 34 percent of recipients required to participate in PRIDE found jobs over a two-year period.  Many program participants actually ended up worse off, as sanctions for not meeting the work requirements took away their only source of cash income.
  • Over the long term, the most successful programs supported efforts to boost education and skills, rather than simply requiring people to search for work or find a job.  In welfare reform’s early years, proponents of a “work first” approach, which focuses on quick entry into the job market, declared victory over proponents of a “human capital development” approach, which focuses on building education and skills.  They did so partly by characterizing two important programs — in Portland, Oregon and Riverside, California — as “work first” programs, even though both supported participants’ efforts to improve their education or skills.

    More recent studies demonstrate that post-secondary education and training programs can significantly improve disadvantaged people’s long-term employment trajectories.  For example, within two years, those randomly chosen to participate in three sectoral employment programs that prepare people for in-demand jobs and were evaluated as a part of the Sectoral Employment Impact Study earned 29 percent more than those not selected to participate.  Participants also were more likely to be employed, be employed consistently, work in jobs with higher wages, and work in jobs that offered benefits. 

For more information, see LaDonna Pavetti, “Work Requirements Don’t Cut Poverty, Evidence Shows,” CBPP, updated June 7, 2016, https://www.cbpp.org/research/poverty-and-inequality/work-requirements-dont-cut-poverty-evidence-shows and Tazra Mitchell and LaDonna Pavetti, “Life After TANF in Kansas: For Most, Unsteady Work and Earnings Below Half the Poverty Line,” CBPP, updated February 20, 2018, https://www.cbpp.org/research/family-income-support/life-after-tanf-in-kansas-for-most-unsteady-work-and-earnings-below.