Policymakers have tried for decades to curb the amount of teen violence that plagues many U.S. cities. Now, a new study finds lasting effects from a surprisingly simple and relatively low-cost intervention: a summer job.
“It’s a great study,” says Steven Raphael, a professor of public policy at the University of California, Berkeley, who studies urban and labor economics. “It shows that a short, concentrated positive experience can have a long-lasting effect on this population.”
The research, by Sara Heller of the University of Pennsylvania (Penn), found that a summer jobs program significantly reduced violent crime by African-American teenagers for as long as a year after they stopped working and returned to school. However, it’s not clear why jobs can make such a difference for these high-risk students. “Something is going on,” agrees Dan Bloom, a researcher at the nonprofit MDRC in New York City who studies barriers to employment for marginalized groups. “But what exactly is the mechanism? You have a lot of things happening, so it’s hard to tease out what led to this effect.”
Heller was a public policy graduate student at the University of Chicago when she became interested in the city’s long-running summer jobs program for disadvantaged students. Researchers typically look at the economic impacts of these programs, notably their ability to raise income levels and prepare students for the adult workforce. But Heller wondered if giving the teens jobs could also lower sky-high crime rates among a group in which one in five has an arrest record and a similar proportion has been the victim of a crime.
The regular summer jobs program consists of working 25 hours a week for 8 weeks at a public sector job, such as being a camp counselor or tending a community garden. For Heller’s study, Chicago city officials agreed to modify that schedule so that half of participants would work only 15 hours and spend 10 hours being taught an approach to better manage their emotions and behavior, called cognitive behavioral therapy. Youths in both experimental groups were also given an adult mentor to discuss any work-related issues.
Using administrative records to track their activities over the 13 months after the program ended, Heller found a 43% reduction in arrests for violent crimes—murder, rape, robbery, and aggravated assault—among the 730 youths in the two treatment groups compared with the 904 youths in the control group. (The summer jobs program receives many more eligible applicants each year than it has slots, providing Heller with an ideal control group whose demographics matched the treatment groups.) There was a negligible difference in violent-crime arrest rates between the two treatment groups. Moreover, the number of arrests for property crimes, drugs, and other nonviolent offenses was roughly the same for both the treatment and control groups, Heller reports online today in Science.
“People have long had the sense that it takes a lot to move the needle,” Heller says, citing as an example the 8-month, residential, federally funded Job Corps program that she characterizes as a “lengthy, intensive, and costly intervention.” Heller was also pleasantly surprised to find that the summer jobs program had a lasting impact: The number of violent-crime arrests declined steadily over the entire 16 months that the youths were tracked. Such persistence, she says, suggests that the teenagers retained whatever it was they had learned.
Heller admits she doesn’t know why the summer jobs program suppressed violent crimes so effectively. But she suspects that both the job and the social-emotional learning somehow gave the teenagers the ability to deal with interpersonal conflicts without resorting to violence. And that skill can make all the difference, she says.
“Kids get shot every day over nothing—a dirty look, a curse, or a gesture of disrespect,” she notes. “Whatever the mechanism is, it’s something that both treatment elements share.” At the same time, the study found no meaningful improvement in school outcomes, including better attendance or academic performance.
The study leaves plenty for other researchers to chew on: How important is the mentor? Do the jobs also provide a network of positive role models and useful contacts? Are there certain subgroups of youths who are especially receptive to such an intervention? “Now that there’s something positive we can measure, people may try to study different permutations of summer jobs,” predicts MDRC’s Bloom. An attitudinal survey of the teenagers could also shed light on the reasons for their altered behavior, he says. “And, of course, you’d like to see it replicated elsewhere.”
Heller completed her Ph.D. last year and is now an assistant professor of criminology at Penn. She hopes to continue tracking these youths through and beyond high school to see how long the anticrime effects last. But having a version of her doctoral thesis published in Science, she admits, is not a bad start.
*Update, 4 December, 3:24 p.m.: This article has been updated to make clear that the study included both males and females.