Revisiting the relationship between age, employment, and recidivism

Employment theoretically serves as a source of informal social control that can promote desistance from crime (Sampson & Laub, 1993). Findings from studies assessing the effects of employment, however, have been mixed. In a seminal study, Uggen (2000) reanalyzed data from the National Supported Work (NSW) Demonstration Project and found that employment significantly reduced the rate of recidivism among individuals aged 27 and older but had no impact on younger individuals. We reproduce and replicate Uggen's (2000) findings with data from four distinct employment programs: The National Supported Work Program (1975–1979), the Transitional Aid Research Project (1976–1977), the Employment Services for Ex-Offenders (1981–1984), and the Enhanced Services for the Hard-to-Employ Center for Employment Opportunities (2004–2008). We closely reproduced Uggen's original findings in the NSW but found evidence that the statistically significant interaction between age and employment in the NSW was only present at the year 3 follow-up and the observed effect is highly sensitive to minor threats to internal validity. Furthermore, a significant age–employment interaction was not observed in the three other data sources. These findings should encourage scholars to continue to investigate the age-graded nature of employment and crime, especially through a sociohistorical lens.