The popular saying that “nothing stops a bullet like a job” reflects a common intuition that employment programs should help reduce the enormous youth violence problem in the United States. But most existing evidence suggests that youth job training and placement programs are actually a bad investment, requiring so much intensity and expense to have any effect that the benefits do not outweigh the costs. However, the research focuses almost exclusively on out-of-school or out-of-work populations. This may be the wrong approach; considerable evidence in domains like education suggests that intervening before youth leave school can generate larger improvements at smaller costs than trying to reverse the negative trajectories of dropouts after the fact. Summer jobs are a widespread version of this kind of primary-prevention approach to employment, but there is almost no evidence beyond observational studies on their effects.

This paper presents the first experimental evidence on whether summer jobs can reduce youth violence and delinquency. I randomly assign 1,634 8^th – 12^th grade applicants from 13 low-income, high-crime schools in Chicago to be offered a part-time subsidized summer job, a job plus a social-emotional learning curriculum, or the status quo (no additional services). I find that the program generates an enormous decrease in violence 16 months after random assignment: 8.7 fewer incidents per 100 participants, a 32 percent decline. This drop is driven by a significant decrease in offending – violent-crime arrests decline by 44 percent (5.2 per 100 participants) – and a substantively large (23 percent) though not statistically significant decrease in violent victimization (3.5 per 100 participants, p = 0.18). There is no difference in other types of crime, nor across treatment arms. Although a preliminary benefit-cost analysis is very sensitive to the social cost assigned to a lost life, the results show that an 8-week employment intervention can have a considerable and lasting impact on youth violence. Counter to prevailing pessimism about youth employment programs, it appears that intervening before youth spend time out of school can generate large improvements at a relatively low cost.