Despite recent figures indicating that 26% of public high schools in the United States require student involvement in community service or service learning, there is little empirical evidence to support such policies. In the present study, successive cohorts of high school students, one without (n = 174) and two with a community service requirement (n = 312), were compared longitudinally on measures of civic attitudes and behaviors. Each cohort was divided according to individual students’ inclinations to serve voluntarily. Students already inclined to serve scored high on all measures throughout and showed no advantage after meeting the requirement. However, students who were less inclined to serve showed marked gains on three of four civic measures after completing their requirement. Whereas they scored as low as their less-inclined counterparts who had no requirement during grade 11, they differed from their counterparts after serving the mandated 40 hours during grade 12. This quasi-natural experiment provides support for the argument that a well-designed service program can have a clear benefit in civic development for nonself-selected youth.