Objective: To evaluate the predictive power of executive functions, in particular, response inhibition, in relation to alcohol-related problems and illicit drug use in adolescence. Method: A total of 498 children from 275 families from a longitudinal high-risk study completed executive function measures in early and late adolescence and lifetime drinking and drug-related ratings at multiple time points including late adolescence (ages 15-17). Multi-informant measures of attention-deficit/hyperactivity disorder and conduct disorder were obtained in early childhood (ages 3-5), middle childhood, and adolescence. Results: In multilevel models, poor response inhibition predicted aggregate alcohol-related problems, the number of illicit drugs used, and comorbid alcohol and drug use (but not the number of drug-related problems), independently of IQ, parental alcoholism and antisocial personality disorder, child attention-deficit/hyperactivity disorder and conduct symptoms, or age. Multivariate models explained 8% to 20% of residual variance in outcome scores. The incremental predictive power of response inhibition was modest, explaining about 1% of the variance in most outcomes, but more than 9% of the residual variance in problem outcomes within the highest risk families. Other measured executive functions did not independently predict substance use onset. Conclusion: Models of alcoholism and other drug risks that invoke executive functions may benefit from specifying response inhibition as an incremental component. (Contains 3 tables.)