This paper examines the impacts of work experience acquired while youth were in high school (and college) on young men's wage rates during the 1980s and 1990s. Previous studies have found evidence of sizeable and persistent rates of return to working while enrolled in school, especially high school, on subsequent wage growth. Such findings may represent causal effects of having acquired work experience while still enrolled in school, but they may also be the result of failure to fully account for individual differences in young adults' capacities to acquire such skills and be productive in the work force later in life. We re-examine the robustness of previous attempts to control for unobserved heterogeneity and selectivity. We explore more general methods for dealing with dynamic forms of selection by explicitly modeling the educational and work choices of young men from age 13 through their late twenties. Using data on young men from the 1979 National Longitudinal Survey of Youth (NLSY79), we find that the estimated returns to working while in high school or college are dramatically diminished in magnitude and statistical significance when one uses these dynamic selection methods. As such, our results indicate a decided lack of robustness to the inference about the effects of working while in school that has been drawn from previous work.