According to data from the National Longitudinal Survey of Youth 1997, employment during high school is ubiquitous. For example, fully 87 percent of students have worked during at least one month of high school by May of their senior year, and many employed students work at high intensity, generally considered over 20 hours per week. In addition, almost one in four employed seniors (23%) works full time at over 35 hours per week. But empirical work stretching back to the early 1980s establishes that long work hours in combination with school enrollment are correlated with elevated involvement in a variety of problem behaviors, not the least of which are crime and substance use. Partly on the basis of the research showing adverse effects of extensive work involvement on problem behavior, the National Research Council (1998:227) proposed extending federal limits on youth work involvement during the school year-which currently apply to 14 and 15 year olds only-to 16 and 17 year olds. This recommendation was adopted in the Youth Worker Protection Act, a bill introduced in the 108th Congress.
This recommendation represents an example of how sociological research can lead directly to public policy recommendation, and at the very least to a change in attitudes on the part of policymakers. Indeed, in the 1970s, three federal government panels recommended an increase in employment for adolescents. Yet the research in question does not directly address the proposal made by the National Research Council. In their recommendation, the NRC infers that the movement into the labor market that accompanies the relaxation of child labor laws at age 16 leads to increased problem behavior. But the research that led to this recommendation is based on the cross-sectional correlation between work intensity and problem behavior with, at best, a control for previous offending. Knowing that youths who work longer hours have more problem behavior than youth who work fewer, controlling for observables, does not in itself prove that youths will increase their problem behavior as a result of an increase in work involvement. Indeed, recent econometric work cautions that the adverse work intensity effect observed in prior studies may be driven by inattention to unobserved heterogeneity, and that longer work hours may have no causal impact or may under some circumstances reduce delinquent involvement.
Broader Impacts
The PI maps state-to-state variation in child labor laws at ages 16 and 17 onto changes in self-reported work and problem behavior for youths in the National Longitudinal Survey of Youth 1997. The study will examine what happens to problem behavior, job quality, and other dimensions of youth work as a result of the change in policy regime from federal to state laws at age 16. The study pays special attention to the interaction between labor market conditions and these laws, in part because it is unlikely that the laws will be as relevant in loose labor markets as they are in tight labor markets where there is a demand for youth labor. The general procedure gives the PI the ability to shed light on the potential consequences of expanding federal restrictions to cover 16 and 17 year olds. This the first study that so directly explores the policy recommendation from the NRC. This approach will also provide new quasi-experimental evidence for the sociological question regarding the causal relationship between work and youth behavior. Study results will be unique in that they derive their power from legislative variation that is of direct interest to policymakers, and as such, will be of interest to both social scientists and policymakers.
.
