The Effects of Cognitive Behavioral Therapy and Financial Assistance on Crime, Violence, and Employment: Experimental Evidence from Allegheny County

We are collaborating with Allegheny County to develop and evaluate a series of social programs to assist individuals who were previously incarcerated. We request funding to support the randomized evaluation of the first social program developed through this collaboration, which tests two conceptually different ways to support individuals exiting jail and transitioning back into the community. The first intervention is a cognitive behavioral therapy program designed to help individuals avoid high-risk situations and triggers. The second intervention is a series of unconditional financial grants totaling $2,400 to help individuals solve immediate financial needs. We hypothesize that the first intervention will improve outcomes if poor decision-making drives adverse outcomes and high-risk individuals attend the classes. Meanwhile, the second intervention will improve outcomes if short-term financial constraints drive adverse outcomes and high-risk individuals are willing to accept financial assistance. With an anticipated sample size of 3,100 individuals, we can detect intent-to-treat effects of 7.4-11.0 percentage points for recidivism, 5.8-8.6 percentage points for violence involvement, and 0.38-0.57 quarters of non-employment. We will also develop new methodological approaches to identify individuals who will benefit most from our interventions using a combination of predictive algorithms and human insights, which will support our on-going work with Allegheny County.