Building on a growing literature showing that early college high schools substantially improve educational outcomes, we investigate possible spillover impacts of this intervention on civic outcomes in North Carolina, which houses an early college in most of its counties. We present both lottery and observational impacts on voting and criminal convictions. Our results suggest a modest increase in voting during early adulthood of about 4 to 5 percent, though lottery estimates do not rule out a null effect. For criminal convictions, lottery estimates are imprecise due to very low conviction rates, but observational evidence suggests a moderate decrease in convictions. We additionally identify stronger impacts on voting and conviction outcomes for key student subgroups, particularly black males and economically-disadvantaged white students. These results suggest that scaling up the early college program can improve youth civic outcomes and help to close key civic and political participation gaps. [This paper was published in "Social Sciences Research" v99 Article 10258 2021.]