We examine the work experiences during middle school and high school of U.S. females and males and find that most of the child-oriented work such as babysitting and camp counseling is done by females. If the type of work undertaken while young affects either development of specific human capital or preferences, then these early work experiences may have measurable effects on later life outcomes. This paper examines whether or not having a job as a teenager, and whether or not it is a child-oriented job, causes differences in labor market behavior among young adults. In addition to a set of standard controls, in order to account for the endogeneity of students’ work decisions, we utilize a set of state-level instruments, including state-level child-labor laws and indicators of relative demand for, and supply of, child-oriented workers. While the effects we find are complex and sometimes hard to interpret, they suggest that work in 10th grade has a positive causal effect on later labor market outcomes and delays family formation, but to a lesser extent when jobs were child-oriented.