During the middle 1970s, a number of prestigious commissions convened to study the problems of adolescents (e.g. President's Science Advisory Committee, 1974; National Commission on the Reform of Secondary Education, 1973; National Panel on High Schools and Adolescent Education, 1975) reached the common conclusion that additional early work experience would foster the development of personal responsibility, smooth the transition from the youth to adulthood, and improve educational performance and occupational attainment. Shortly thereafter, a number of federal initiatives (e.g. the Career Education Incentive Act of 1977) were passed with the goal of increasing the employment experience of youths.
These recommendations were made in the absence of any hard empirical evidence that increased job-holding causes or even is correlated with favorable outcomes. Economic theory also fails to provide unambiguous predictions concerning the efficacy of youth employment. For example, the human capital model identifies both potential benefits and costs of working. On the one hand, time devoted to jobs could detract from potentially more productive educational investments. On the other, the employment might provide skills and knowledge which increase future productivity and complement in-class learning. Early work experience could also speed the process by which youths obtain positions where there is a good match between job requirements and work qualifications.
Previous research suffers from two fundamental shortcomings which make it difficult to determine the net benefits or costs of job-holding by students. First, most studies treat youth employment as an exogenous variable, ignoring the selection process determining which student work and, conditional upon doing so, how many hours they are employed. Indeed, much of the prior investigation has used unrepresentative samples and held constant few, if any, individual characteristics. Second, analysts have focused upon educational achievement and employment outcomes shortly after the completion of high school but have obtained little information on long-run labor market success.
Using data from the National Longitudinal Survey of Youth (NLSY), the paper improves in both areas. Several strategies are used to account for difficult-to-observe differences between workers and nonworkers. These entail controlling for an unusually comprehensive set of background characteristics, examining whether reduced estimates are biased by the potential endogeneity of high school employment, and testing the robustness of key results to changes in samples and specifications. The dependent variables are employment consequences 6 to 9 years after the scheduled date of high school graduation, thus providing the best available information on long-term effects of the student job-holding. In addition, this study examines a wider variety of economic outcomes and utilizes better information on high school employment status than previously been available. The investigation focuses upon the number of hours worked by high school students. Examining the role of job characteristics or of employment by college students is beyond the scope of this analysis and is reserved for future study.