Recent empirical research has found that children's noncognitive skills play a critical role in their own success, young children's behavioral and psychological disorders can severely harm their future outcomes, and disruptive students harm the behavior and learning of their classmates. Yet relatively little is known about wide-scale interventions designed to improve children's behavior and mental health. This is the first nationally representative study of the provision, financing, and impact of school-site mental health services for young children. Elementary school counselors are school employees who provide mental health services to all types of students, typically meeting with students one-on-one or in small groups. Given counselors' nonrandom assignment to schools, it is particularly challenging to estimate the impact of these counselors on student outcomes. First, cross-state differences in policies provide descriptive evidence that students in states with more aggressive elementary counseling policies make greater test score gains and are less likely to report internalizing or externalizing problem behaviors compared to students with similar observed characteristics in similar schools in other states. Next, difference-in-differences estimates exploiting both the timing and the targeted grade levels of states' counseling policy changes provide evidence that elementary counselors substantially influence teachers' perceptions of school climate. The adoption of state-funded counselor subsidies or minimum counselor-student ratios reduces the fraction of teachers reporting that their instruction suffers due to student misbehavior and reduces the fractions reporting problems with students physically fighting each other, cutting class, stealing, or using drugs. These findings imply that there may be substantial public and private benefits derived from providing additional elementary school counselors. (Contains 9 tables, 1 figure and 28 footnotes.)