Disentangling Selection From Causation in the Empirical Association Between Crime and Adolescent Work

Robert Apel, University of Maryland at College Park

An extensive literature finds that youths who work intensively (particularly those that work intensively while in high school) are more likely to engage in a variety of problem behaviors such as minor and serious delinquency, substance use, and school misconduct. Although there is widespread agreement about the presence of this adverse "work effect" on problem behavior, there remains considerable ambiguity about its causal significance. What complicates research on the effect of adolescent work on problem behavior is the fact that youth employment is driven, in part, by individual choices. For this readon, analysts must confront the problem that adolescents are not randomly allocated into the youth labor market. Consequently, there exists the possibility that the estimate of the "work effect" is biased in the absence of rigorous controls for enduring differences between youths in the likelihood of being employed. In this paper, we explore in greater detail the nature of the association between youth employment and problem behavior. We rely on methods that directly confront the endogeneity of youth employment by estimateing two-stage selectivity and instrumental variable models.

(Return to Program Resources)

Updated 05/20/2006