Criminologists from diverse theoretical perspectives have long asserted tha
t the quality of employment is more strongly associated with criminal behav
ior than its presence or absence. By this argument, "good jobs" or "meaning
ful work" are necessary to induce offenders to desist from crime. This pape
r constructs a satisfaction based measure of job quality using data from th
e National Supported Work Demonstration and the 1977 Quality of Employment
Survey and tests whether employment in high quality jobs reduces the likeli
hood of criminal behavior among offenders. After statistical corrections fo
r selection into employment, job quality is found to reduce the likelihood
of economic and non-economic criminal behavior among a sample of released h
igh-risk offenders. None of the most salient alternative explanations-sampl
e selection, human capital accumulation, personal expectations, external la
bor market effects, or prior criminality-appear to diminish the job quality
effect.