Adolescent mental health problems are associated with poor health and well-being in adulthood. We used data from a cohort of 2,264 children born in large US cities in 1998-2000 to examine whether neighborhood collective efficacy (a combination of social cohesion and control) is associated with improvements in adolescent mental health.We found that children who grew up in neighborhoods with high collective efficacy experienced fewer depressive and anxiety symptoms during adolescence than similar children from neighborhoods with low collective efficacy. The magnitude of this neighborhood effect is comparable to the protective effects of depression prevention programs aimed at general or at-risk adolescent populations. Our findings did not vary by family or neighborhood income, which indicates that neighborhood collective efficacy supports adolescent mental health across diverse populations and urban settings. We recommend a greater emphasis on neighborhood environments in individual mental health risk assessments and greater investment in community-based initiatives that strengthen neighborhood social cohesion and control.