Bridging Disciplinary Boundaries (January 11 - 14, 2007)

Sanctions in the welfare system are financial penalties that affect the amount of cash aid provided to recipients when they fail to comply with requirements. As a behavior modification tool, the purpose of sanctions is to motivate/enforce participation with welfare-to-work requirements, the assumption being that participation will facilitate the transition of recipients from assistance into employment. There is some evidence to suggest, however, that sanctioned recipients may have worse outcomes in terms of employment and earnings, when compared to non-sanctioned recipients, but findings are mixed regarding welfare usage. This growing body of research generally measures sanction receipt as a dichotomous variable, which could potentially mask differences within the sanctioned population, thus leading to conflicting findings. For example, some recipients receive a sanction and then subsequently comply; it is currently not known whether these individuals are similar to non-sanctioned recipients or those who receive sanctions and never comply. The research question addressed in this study is whether compliance with welfare-to-work requirements, as measured by pattern of sanction receipt, affects future outcomes. Using survey data and matched administrative records from a sample of 701 welfare recipients in California, this study examines the impact of sanction status on subsequent welfare usage, employment, and earnings outcomes, for three groups of recipients: those who are never sanctioned, those who receive a sanction but then comply with requirements, and those who receive a sanction and never comply. The findings from a multinomial regression analysis of welfare usage indicate the following: 1) regardless of compliance after sanction receipt, sanctioned recipients are more likely to return to aid after leaving assistance (versus remaining off aid) compared to non-sanctioned recipients, 2) recipients who are sanctioned but who eventually comply with requirements are more likely to remain continuously on assistance (versus remaining off aid) compared to non-sanctioned recipients, and 3) there are no differences between sanctioned recipients who remain in sanction status and non-sanctioned recipients in remaining continuously on aid (versus remaining off aid). In terms of employment and earnings outcomes, results from logistic and Ordinary Least Squares (OLS) analyses, respectively, indicate that sanctioned recipients who eventually comply with requirements have worse outcomes when compared to non-sanctioned recipients, but there are no differences when comparing sanctioned recipients who remain in sanction status and those non-sanctioned. Perhaps because California is a partial sanction policy state, where only the adult portion of the total grant amount is removed for noncompliance, there are different dynamics underlying the decision/ability to comply between sanctioned recipients who eventually comply and those who do not. Overall, the findings from this study suggest that sanctioned recipients who eventually comply with requirements have poorer welfare, employment, and earnings outcomes compared to non-sanctioned recipients. Implications for sanction policy and future research are discussed.