Cognitive Behavior Therapy Reduces Crime and Violence over 10 Years: Experimental Evidence
In most societies, a small number of people commit the most serious violence. Short-term studies have shown that cognitive behavioral therapy (CBT) can reduce such antisocial behaviors. These behavior changes may be temporary, however, especially from therapy on its own. This is unsettled, however, for there has been little randomized, long-term research. We follow 999 high-risk men in Liberia 10 years after randomization into either: 8 weeks of a therapy; a $200 grant; both; or a control group. A decade later, both therapy alone and therapy with economic assistance produce dramatic reductions in antisocial behaviors. Drug-selling and participation in thefts and robberies, for example, fall by about half. These impacts are greatest among the highest-risk men. The effects of therapy alone, however, are smaller and more fragile. The effects of therapy plus economic assistance are more sustained and precise. Since the cash did not increase earnings for more than a few months, we hypothesize that the grant, and the brief legitimate business activity, reinforced the habit formation embodied in CBT. Overall, results suggest that targeted CBT plus economic assistance is an inexpensive and effective way to prevent violence, especially when policymakers are searching for alternatives to aggressive policing and incarceration.
The Network for Empowerment and Progressive Initiatives (NEPI) and Global Communities implemented the program, and NEPI provided ongoing support. Innovations for Poverty Action coordinated all research activities. For research assistance in this latest round, we thank G. Dackermue Dolo, Victor Gamarra, Sebastián Hernández, Walker Higgins, Andreas Holzinger, Bruno Aravena Maguida, Laura McCargo, Anna Mysliwiec, Albert Nyuangar, and Newton Toe. We thank Sara Heller for comments. This round of the study was funded by the National Science Foundation (SES 1919399) and the Wellspring Philanthropic Fund. The contents of this study are the responsibility of the authors and do not necessarily reflect the views of their employers, funding agencies, or governments. The authors have no relevant or material financial interests that relate to the research described in this paper. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.