I work with violent young men, from Africa to the Americas — guys so far into lives of crime that a natural reaction is hopelessness. If, 10 years ago, you told me that eight weeks of therapy plus a little cash could turn a significant proportion of them away from that life, I’d have scoffed. But it’s true — as three colleagues and I demonstrated in a new working paper. What we learned in Monrovia, Liberia, holds the potential to change the way America handles its own epidemics of crime and murder.
Cognitive behavioral therapy, or CBT, is a set of simple techniques for recognizing your problematic, “automatic” behaviors and training yourself to act differently. For instance, when an emotion like anger swells, CBT helps you recognize how it can distort your thinking. You can practice habits that put your rational brain back in charge, such as slow breathing, counting to 10 or walking away.
Research has shown that CBT can be effective in treating a range of disorders, including depression and anxiety, and CBT techniques have long been used to combat aggression and criminal behavior, too. Juvenile prisons use it routinely, for instance, although its long-term effectiveness remains a subject of study. In 2009, my colleagues and I, working with a program called Sustainable Transformation of Youth in Liberia, decided to explore, using a randomized trial, how much difference CBT could make for men deeply immersed in crime and daily violence. STYL recruited 1,000 men who fit that description. Half were offered therapy via lottery: For two months they were counseled in groups three times a week, four hours at a time, supplemented by individual sessions. Then, after the therapy ended, we held a second random draw to determine who got $200 cash. (We hypothesized that cash alone might make a difference in men’s lives.) This made four groups — therapy only, cash only, both therapy and cash, and a control group that received neither.
We looked at effects one month after the program, then one year later and, just recently, after 10 years. In an earlier paper, we showed that therapy plus cash had promising results after a year. Other research on CBT and violence prevention suggests that therapy’s effects fall off after that, and Liberia was no different — the effects of therapy alone began to fade with the first year.
We expected therapy plus cash to suffer the same fate over time (as did most of the experts we surveyed). But a decade later, after we tracked down and interviewed as many of men from the original study as we could find (some 830, 93 percent of those still alive), we learned that the large effects in the therapy-plus-cash group were sustained.
After 10 years, for example, 5 percent of the men who got both treatments (cash and therapy) reported selling drugs, compared with 10 percent for the control group. Also, each person who got therapy and money committed half as many thefts, on average, than control group men — results similar to those in our one-year study. We don’t have data for every intervening year, but the surveys we have suggest roughly 30 fewer robberies and muggings per person, per year, for a decade. Since together the cash and therapy cost $530 per person, the expense was less than $2 for every theft avoided (ignoring the declines in drug selling and interpersonal violence).
When it came to therapy alone, after a decade there were detectable signs of improvement, but they were substantially smaller than therapy plus cash, and not nearly as statistically precise. We think the difference is explained by the fact that even modest financial support gave the men enough stability in their lives to put into practice the behaviors they learned during therapy. The men who got money generally used the cash prudently, we found, starting small businesses like shoeshine stands. But these enterprises were fleeting, and within months, their goods were stolen or the businesses simply collapsed. So therapy plus cash didn’t mean they earned more money one year or 10 years later.
As with any study, there are weaknesses. Impoverished Liberia does not have crime statistics or arrest records, so our data on the men’s behavior was self-reported. (We asked the men about dozens of antisocial and illegal behaviors, including drug-dealing, theft, fighting, arrests, carrying weapons and domestic abuse.) We did what we could to validate this information, sending qualitative researchers to observe 100 of the men for four days; their impressions matched the survey data. We also measured outcomes such as patience with games played with real money, and still saw impacts there.
This 10-year research odyssey grew out of my relationship with Johnson Borh. He still runs STYL, as he did when we first met in 2008, when he was a smiling, meaty man in his 30s who moved easily among members of Monrovia’s underclass. He introduced me to men clustered in the thatch huts that served as drug dens, or standing by with empty wheelbarrows, ready to haul goods for hire (but making most of their money from robbery).
Borh had been offering a version of STYL for a decade. I started sitting in on the meetings that he and his self-taught social workers led. We rested on plastic chairs in an abandoned building. The men rehearsed handling raw emotions and peacefully negotiating their way through threatening situations. Counselors also urged them to try on, for size, a new social identity — trading shorts and sandals for pants and dress shoes, for instance, and practicing going to banks and supermarkets.
Borh had patched together the program from online materials, but it was clear that this was a cognitive behavioral program in all but name (indeed, many of the handbooks he downloaded were based on CBT). To see if it worked, Borh and I teamed up with a behavioral economist, Julian Jamison; a clinical psychologist, Margaret Sheridan; and a master of randomized trials, Sebastian Chaskel.
Beyond the large and persistent impacts of therapy and economic assistance, our results offer other good news. Even among the control group, half returned to more mainstream, nonviolent lives within a year. This suggests that even the young Liberian men who seem most lost revert to the mean over time. We also found that STYL’s impacts on crime and violence were concentrated among the men who reported, at the outset, the most anti-social behavior. This suggests that cities can make meaningful progress on violence with targeted interventions for the small number of the most at-risk people.
This is just one of several studies that show CBT’s promise in combating violence. At the same time we were studying STYL, a group of economists and psychologists in Chicago evaluated two versions of Becoming a Man — a CBT-based program for troubled adolescents in the city’s most violence-plagued high schools. Their research found that CBT alone — no economic help, in this instance — led to 45 to 50 percent reductions in violent crime arrests over the next one to two years. And even though the effects of therapy alone faded somewhat over a longer time frame, the authors demonstrated that BAM’s benefits were large enough to more than justify the program’s cost. The researchers also studied a BAM-like program in juvenile detention, in the same paper, and drew similar conclusions.
American cities need solutions to the problems of gun violence, especially ones that don’t involve hassling and imprisoning members of minority groups. CBT, in concert with economic assistance, offers a promising alternative, one that averts community violence rather than punishing and jailing offenders after it happens. President Biden has advocated putting more money into “community violence interruption” programs like Becoming a Man.
Controlling anger is a habit we all need to practice, along with handling contentious relationships, recognizing biases and seeing arguments from another point of view. Decades of social science show us that peaceful societies saturate citizens’ lives with these lessons, fortifying them with laws and norms. They socialize their young to nonviolence, in what sociologist Norbert Elias called “the civilizing process.”
In America we start early, with “Sesame Street,” “Mr. Rogers” and other television shows deliberately designed to impart these behaviors. And most schools teach the same socio-emotional skills that programs like STYL impart. But some young people don’t get these lessons in class, or at home — or they just need a little extra help to learn them. This includes disaffected young men in Chicago and street youth in Monrovia — but in principle the approach could work with anyone engaged in sustained and regular violence. We don’t yet know the full range of CBT’s benefits and limitations. But like Johnson Borh, we ought to try things, tinker and find out what works. He showed us something deeply hopeful: that we can make societies less violent not by using force, but by laying down a path to a different life.