Schools, Neighborhoods, and the Long-Run Effect of Crime-Prone Peers

Abstract

There is a large and growing literature on how childhood peers shape outcomes. However, within this literature little is known about the effects of the different neighborhood factors, such as the relative importance of school versus neighborhood peers. This paper, authored by Stephen B. Billings and Mark Hoekstra, examines the effect of elementary-aged peers on adult crime and other outcomes by exploiting cohort variation in the proportion of peers with an arrested parent. Importantly, this data enables the authors to distinguish between the effect of school and neighborhood peers. Results indicate that a five percentage point increase in school and neighborhood crime-prone peers increases adult arrest rates by 6.5 and 2.6 percent, respectively. Additional evidence indicates that adult crime is primarily driven by inter-actions in schools rather than in neighborhoods. This paper also documents how school and neighborhood peers affect cognitive and non-cognitive outcomes during adolescence.