Using data from two long-term longitudinal projects, we investigated reciprocal relations between maternal reports of physical discipline and teacher and self-ratings of child externalizing behavior, accounting for continuity in both discipline and externalizing over time. In Study 1, which followed a community sample of 562 boys and girls from age 6 to 9, high levels of physical discipline in a given year predicted high levels of externalizing behavior in the next year, and externalizing behavior in a given year predicted high levels of physical discipline in the next year. In Study 2, which followed an independent sample of 290 lower income, higher risk boys from age 10 to 15, mother-reported physical discipline in a given year predicted child ratings of antisocial behavior in the next year, but child antisocial behavior in a given year did not predict parents' use of physical discipline in the next year. In neither sample was there evidence that associations between physical discipline and child externalizing changed as the child aged, and findings were not moderated by gender, race, socioeconomic status, or the severity of the physical discipline. Implications for the reciprocal nature of the socialization process and the risks associated with physical discipline are discussed.