Citation

Jacob BA, Lefgren L. NBER Work. Pap. Ser. 2003; 2003(online): w9653.

Copyright

(Copyright © 2003, National Bureau of Economic Research)

DOI

unavailable

PMID

unavailable

Abstract

This paper examines the short-term effect of school on juvenile crime. To do so, we bring together daily measures of criminal activity and detailed school calendar information from 29 jurisdictions across the country, and use the plausibly exogenous variation generated by teacher in-service days to estimate the school-crime relationship. We find that the level of property crime committed by juveniles decreases by 14 percent on days when school is in session, but that the level of violent crime increases by 28 percent on such days. These results do not appear to be driven by inflated reporting of crime on school days or substitution of crime across days. Our findings suggest that incapacitation and concentration influence juvenile crime - when juveniles are not engaged in supervised activities, they are more likely to engage in certain anti-social behaviors; at the same time, the increase in interactions associated with school attendance leads to more interpersonal conflict and violence. These results underscore the social nature of violent crime and suggest that youth programs - particularly those with no educational component such as midnight basketball or summer concerts - may entail important tradeoffs in terms of their effects on juvenile crime.