What if someone told you that you could dramatically reduce the crime rate without resorting to coercive policing or incarceration? In fact, what if they said you could avert a serious crime — a robbery, say, or maybe even a murder — just by shelling out $1.50?
That’s such an incredibly good deal that it sounds too good to be true. But it’s been borne out by the research of Chris Blattman, Margaret Sheridan, Julian Jamison, and Sebastian Chaskel. Their new study provides experimental evidence that offering at-risk men a few weeks of behavioral therapy plus a bit of cash reduces the future risk of crime and violence, even 10 years after the intervention.
Blattman, an economist at the University of Chicago, never intended to conduct this study. But in 2009, he was hanging out with an acquaintance in Liberia named Johnson Borh, who showed him around the capital city of Monrovia. Since Blattman studies crime and violence, Borh took him to visit the pickpockets, drug sellers, and others living on the margins of society.
Along the way, they kept running into guys who were sitting on street corners, eking out a meager living by shining shoes or selling clothes. When these men spotted Borh, they’d run to give him a hug. Blattman recalls that when he asked the men how they knew Borh, they’d say something like, “I used to be like them,” and point to the nearby pickpockets or drug sellers. “But then I went through Borh’s program.”
That’s how Blattman learned about the program Borh had been running for 15 years: Sustainable Transformation of Youth in Liberia. It offered men who were at high risk for violent crime eight weeks of cognitive behavioral therapy. CBT, as it’s called, is a popular, evidence-based method of dealing with issues like anxiety, but Borh adapted the therapeutic strategy to deal with issues like violence and crime.
Meeting with a counselor in groups of around 20, the men would practice specific behavioral changes, like managing anger and exerting self-control. They’d also rehearse trying on a new identity unconnected to their past behavior, by changing their clothes and haircuts and working to reintegrate themselves into mainstream society through community sports, banks, and more.
Blattman wanted to formally study just how effective this kind of program could be. He decided to run a big randomized controlled trial with 999 of the most dangerous men in Monrovia, recruited on the street. The results were so promising that they’ve already inspired a sister program in a very different city: Chicago.
In Chicago, the murder rate is troublingly high, and the police fail to solve 95 percent of all shootings. Finding a way to prevent shootings and other violent crimes is an urgent priority — not only in that city, but across the US, as the recent mass shootings in Buffalo, New York, and Uvalde, Texas, remind us. Given that direct interventions like removing guns are largely blocked by political polarization, and trying to crack down on crime after the fact carries with it risks of policy brutality, we desperately need new solutions to the problem of violence.
The 999 Liberian men were split into four groups. Some received CBT, while others got $200 in cash. Another group got the CBT plus the cash, and finally, there was a control group that got neither.
A month after the intervention, both the therapy group and the therapy-plus-cash group were showing positive results. A year after the intervention, the positive effects on those who got therapy alone had faded a bit, but those who got therapy plus cash were still showing huge impacts: crime and violence were down about 50 percent.
But Blattman didn’t dare to hope that this impact would persist. Experts he surveyed predicted that the effects would steeply diminish over the years, as they do in many interventions.
So it was a great surprise when, 10 years later, he tracked down the original men from the study and reevaluated them. Amazingly, crime and violence were still down by about 50 percent in the therapy-plus-cash group.
Blattman estimates that there were 338 fewer crimes per participant over 10 years. Given that it had cost just $530 per participant to implement the program, that works out to $1.50 per crime avoided.
In short, it worked extremely well. But why did the combination of CBT and some cash work?
The most plausible hypothesis, according to Blattman, is that the $200 in cash enabled the men to pursue a few months of legitimate business activity — say, shoe shining — after the therapy ended. That meant a few extra months of getting to cement their new non-criminal identity and behavioral changes. “Basically, it gave them time to practice,” Blattman told me.
A couple of caveats: The study relied largely on self-reported data about what behaviors participants were and weren’t engaging in, which could raise concerns of experimenter demand (where participants tell experimenters what they want to hear). Also, of the 999 men initially recruited into the study, 103 had died by the time of the 10-year follow-up.
That might make you wonder whether the more violent men, who could have been more resistant to the effects of the program, were just missing from the reevaluation, artificially making it look as if violent crime had dropped more than it really had.
But there are caveats to the caveats. For one thing, the study authors didn’t rely only on self-reported data; they also observed how participants acted in incentivized games where, for example, they’re given a choice between getting $1 now or $5 next week (a good example of self-control and future-oriented thinking). “Our treatment effects are strong and persistent in these outcomes,” the study notes.
By interviewing friends and relatives of each participant who died, the authors also determined the cause of death. They identified only 26 violent deaths. And even when they modeled what would happen to their results if they plugged in “good” outcomes for missing control group members and “bad” outcomes for missing treatment group members, the positive treatment effect for therapy-plus-cash largely remained.
Inspired by the program in Liberia, Chicago has been implementing a similar but more intensive program called READI. Over the course of 18 months, men in the city’s most violent districts participate in therapy sessions in the morning, followed by job training in the afternoon. The rationale for the latter is that in a place with a well-developed labor market like Chicago, the best way to improve earnings is probably to get people into the market, whereas in Liberia, the labor market is much less efficient, so it made more sense to offer people cash.
“We’ll have more results this summer,” said Blattman of the READI program, which he is helping to advise. So far, “it doesn’t look like a slam dunk.”
Still, Chicago is eager to try these therapy-based approaches, having already had some success with them. The city is also home to a program called Becoming a Man (BAM), where high schoolers do CBT-inspired group sessions. A randomized controlled trial showed that criminal arrests fell by about half during the BAM program. Even though effects dissipated over time, the program looks to be very cost-effective.
But this isn’t just a story about the growing recognition that therapy can play a useful role in preventing crime. That trend is part of a broader movement to adopt an approach to crime that is more carrot, less stick.
“It’s all about a progressive, rational policy for social control. Social inclusion is the most productive means of social control,” David Brotherton, a sociologist at the City University of New York, explained to me in 2019.
Brotherton has long argued that mainstream US policy is counterproductively coercive and punitive. His research has shown that helping at-risk people reintegrate into mainstream society — including by offering them cash — is much more effective at reducing violence.
To give one striking example from Brotherton’s research: In 2007, the crime-riddled nation of Ecuador legalized the gangs that had been the source of much of the violence. The country allowed the gangs to remake themselves as cultural associations that could register with the government, which in turn allowed them to qualify for grants and benefit from social programming.
Can you guess what happened to the murder rate over the next few years?
That’s right. It plummeted.