Sanctions are one of the most important policy tools by which work requirements are enforced among welfare recipients. Empirical evidence suggests that some welfare recipients are being sanctioned not because they lack the will to find and retain a job but because they encounter various ‘barriers' which impede on their ability to work. State level studies show that welfare recipients who encounter barriers such as low educational attainment, lack of work experience, or mental health problems are more likely to get sanctioned than welfare recipients who do not encounter these barriers. Moreover, and especially troubling, these studies consistently report that race is a strong predictor of sanction probability. Sanctioned welfare recipients also tend to be worse off in terms of their earnings and are more likely to encounter economic hardships after leaving welfare compared to recipients who were never sanctioned. Empirical evidence has been unavailable so far regarding the effects of sanction polices across states. This study is the first to examine the effect of state sanction policies (stringent vs. lenient) on the probability of being sanctioned and the likelihood of suffering adverse outcomes using a nationally representative sample of recipients. In this study I test the effects of state sanction policies on sanction probability and client well-being to clarify whether more stringent sanction policies result in more negative effects for welfare recipients. As some studies have shown that more stringent sanction policies reduce caseloads more efficiently, it is imperative to evaluate the consequences of sanction policies on welfare recipients.

I use data from the National Survey of America's Families (NSAF), a nationally representative survey conducted by the Urban Institute. NSAF is a multi-wave cross-sectional survey and provides information on more than 40,000 households for each wave. NSAF includes questions on individual and family well-being, employment, earnings and income, economic hardships and program participation. In order to study the effects of state level TANF policies, I merge the NSAF with data from the Welfare Rules Database (WRD), another dataset provided by the Urban Institute. WRD compiles detailed information about welfare rules across states, time, and geographic areas within states. It contains information on all relevant TANF components such as sanctions (including information on good cause exemption rules), work requirements, time limits, or family caps. In this study I use ordinary least square and logistic regression techniques as well as simultaneous equation models to account for existing selection bias.

Sanction polices are an integral part of the central goal of TANF which is to move welfare recipients into the workforce. The results of this study highlight the fact that it is necessary to develop TANF rules that are better able to account for barriers among certain welfare recipients. Stringent sanction policies that are unable to account for ‘barriers' potentially hurt rather than help welfare recipients who struggle to find and retain a job. I therefore recommend that good cause exemptions rules and other relevant TANF rules should be amended to better account for those welfare recipients who encounter barriers.