Date of Award




Document Type


Degree Name

Doctor of Philosophy (PhD)


Department of Sociology

Content Description

1 online resource (v, 137 pages) : illustrations.

Dissertation/Thesis Chair

Steven F Messner

Committee Members

Peter Brandon, Glenn Deane


Crime, TANF, Welfare, Public welfare, Welfare recipients

Subject Categories



The criminological literature has found that welfare is negatively associated with crime, but few studies have tested this association with data from after the Personal Responsibility and Work Opportunity Act (PRWORA) passed in 1996. PRWORA fundamentally altered the structure of the U.S. welfare state. The goals of this dissertation were to identify the research and theories on welfare and crime and test whether the relationship has changed since PRWORA. A review of the literature showed that there were several lines of research on welfare and crime but no studies drawing on the full extent of the welfare-crime literature. After combining the different lines of research, the theories of welfare and crime were expanded to incorporate the time dimension, which heretofore had receive little attention in empirical studies. In addition, methods of caseload reduction (application denials and expulsions) were proposed as new measures of welfare generosity more appropriate for the current welfare system. Using data on state TANF programs from 2000-2010, cross-sectional and longitudinal regression analyses were conducted to test the welfare-crime relationship in manners consistent with the time dimension of the theories of welfare and crime. The results of the cross-sectional analyses did not support the conservative argument that welfare leads to crime and instead found that states with higher per-recipient payments experienced lower property crime rates. When placed in a longitudinal framework, however, traditional measures of welfare such as expenditures and the percent of the poor on welfare did not predict changes in the crime rate. Caseload reduction strategies affected some crimes, but the relationship was weak and inconsistent across different specifications. The dissertation concluded that while TANF programs may be correlated with crime, changes in the programs are unlikely to affect crime rates.

Included in

Criminology Commons