Association for Public Policy Analysis & Management

Youth violence, especially among disadvantaged minority youth, is a pressing social problem: Homicide kills more young black men than the 9 other leading causes of death combined. The first randomized controlled trial to study the effects of summer jobs on violence found surprising results: high-school youth offered an 8-week summer jobs program, who were living in some of Chicago’s most dangerous neighborhoods, had 43 percent fewer violent-crime arrests than their control counterparts over the following 16 months. One hypothesis for the program’s success is that – unlike many well-studied youth employment programs – it served youth prior to school exit, acting as unemployment prevention rather than remediation. Yet with no out-of-school youth in the sample, the study could not test that hypothesis directly. This paper is a follow-up to that study, testing the prevention question directly by randomly assigning a new cohort of 5,228 in- and out-of-school disadvantaged male youth in Chicago to the same summer jobs program. The results will report program effects on violence, measured by administrative arrest records, as well as other types of crime. Policy implications about which populations to target – and which might require more intensive intervention than a summer program can provide – will be discussed. One key implication of this set of studies is that it may be time to rethink the way in which employment programs work.