The Sustainable Transformation of Youth in Liberia (STYL) Program LOGiCA Policy Note - March 2015 Guéckédou 10° Koindu Macenta Mendekoma Voinjama Foya Kolahun Kailahun Buedu GUINEA 8° Pendembu Vahun LOFA Bo Zigida Kenema Daru Lola SIERRA LEONE Zorzor Nzérékoré Zuie Biankouma Gorahun Fassama Salayea Yekepa GBARPOLU Belle Yella Sanniquellie Man Zimmi Diéké Zorgowee Bombor Junction Belefanai Danané Gumgbeta GRAND Ganta Tahn Foequelleh Bopolu Flumpa CAPE Gbarma Gbarnga Bangolo Bo Water Side Suacoco Saclepea Sulima TieneMOUNT Tubmanburg Bong Town BONG Kpaai Gbatala NIMBA Butlo Binhouyé Mecca Salala Duékoué Robertsport Sinje BOMI Zuolay Zekepa Klay MARGIBI Botota Toulepleu Guiglo Kakata Tapita Bensonville Careysburg Gboyi Toe Town CÔTE Monrovia Harbel Kola GRAND BASSA Kangbo Towai Town D'IVOIRE Marshall Zleh Town Hartford Gboleken Zwedru (Tchien) Bolowhea Chayee Edina Garpu GRAND GEDEH Taï Buchanan Drubo (Dubwe) Trade Town RIVER CESS Duabo Galio Yepa Town Pennoken Timbo Gbokohn Shabli Yibuke Cestos City Jakakehn (Kaobli) Juazohn RIVER GEE SINOE Tiehnpo Fish Town Kopo Bame Tawake Tutuke Kodeke Greenville MA Nyaake Nana Kru GRAND KRU RY LA ND Sasstown Barclayville Nemeke Grand Cess Plibo Tabou Harper Behavioral therapy with cash grants leads to significant fall in crime, drug use, and violence among street youth C ities are home to more than half the population of Researchers from Columbia University, Harvard Medical developing countries. Many cities struggle to deal School, and the U.S. government partnered with the with large-scale urban violence, crime, and drugs, Peace & Recovery program at IPA to test and evaluate two especially among poor young men. In post-conflict and innovative approaches for reducing crime and violence other fragile states, poor young men are also targets among street youth in Liberia’s capital, Monrovia. The for mobilization into election intimidation, rioting, and Sustainable Transformation of Youth in Liberia (STYL), a rebellion. In Liberia, which saw two civil wars between pilot program, addressed both the economic and socio- 1989 and 2003, development experts and policymakers behavioral risks that these young men face. It tested two are seeking evidence on the most effective ways to programs, both together and individually: a cognitive reduce crime and violence among high-risk young men. behavioral therapy program of rehabilitation, and unconditional cash transfers. Two of the most common policy prescriptions---policing and job creation---try to reduce crime and violence by The evaluation found that the young men largely invested either changing the economic incentives facing young and saved a substantial fraction of the unconditional men or incarcerating them. Another approach is to cash transfer, yet the money only produced short-run improvements in investment and income. rehabilitate high-risk men, using therapy and counseling to foster “character skills” such as self control, as well The therapy program, while not affecting grant use or as shift preferences and values away from violence, income, led to large and persistent falls in crime, drug crime, and other anti-social behaviors. It’s unclear, use, and violence—especially in the group receiving however, whether adults (especially high-risk men) can cash in addition to therapy. Self-control seems to be a be rehabilitated in this way. Are character and values “character” skill that can be fostered among high-risk malleable in adulthood? adults Researchers Policy Goals Christopher Blattman (Columbia University), Julian C. Jamison (U.S. Cash Transfers, Post-Conflict Recovery, Savings, Vocational Training, Consumer Financial Protection Bureau), Margaret Sheridan (Harvard Youth Employment, Crime and violence reduction Medical School) Country Partner Liberia Innovations for Poverty Action (Peace & Recovery program), Global Sample Communities, Network for Empowerment and Progressive Initiatives 999 men aged 18 to 35 Context areas to provide individual advising and encouragement. Many of the facilitators were themselves reformed ex- combatants and street youth, and included graduates of Liberia underwent two civil wars between 1989 and past programs. 2004. During the wars, nearly 10 percent of Liberia’s 3.5 million people died, a majority were displaced, and tens NEPI’s program encouraged men to practice new skills of thousands of young men were recruited into combat. of self-control and discipline. They practiced making Since 2003, Liberia has been at peace, but poor and and executing modest plans, improving their dress unemployed young men are among the greatest public and hygiene, and managing their anger. NEPI’s therapy safety concerns. program also presumed that the men knew what constituted acceptable behavior in mainstream Liberian This study targeted “hard-core street youth”—men society (e.g. that drug use or stealing are not acceptable), typically ages 18 to 35 who were commonly homeless or but because they did not consider themselves part of lived in extreme poverty. The study ultimately recruited mainstream society, they did not feel subject to its norms 999 young men to participate. Thirty-eight percent or values. Thus facilitators worked to persuade the men had been members of armed groups in the past. On that they could be welcomed back into mainstream average the men were age 25, had roughly eight years society, showed them it was possible, and walked of schooling, earned about 68$ in the previous month, them through the steps. For instance, “homework” and worked nearly 50 hours per week (mainly in low assignments included visiting banks, supermarkets, skill labor and illicit work). More than half reported mobile phone companies, and reintroducing oneself to committing theft in the previous two weeks, and %20 family and community leaders. sold drugs. Half used marijuana or hard drugs daily and many engaged in violent altercations with police other Practicing being a part of mainstream society and community members. adopting mainstream norms was part of helping the men develop impulse control, emotional regulation, and The therapy program long term planning—all crucial, facilitators believed, to changing behaviors such as drug and alcohol abuse, crime, and establishing stable social and economic For over ten years, a Liberian non-profit organization, relationships within a community. the Network for Empowerment and Progressive Initiatives (NEPI) has been conducting a behavioral change therapy program with small groups of high- Evaluation risk men. The therapy program is inherently grounded in Liberian culture, beliefs, and conceptions of society, The programs were assigned by public lottery among but also has a firm grounding in clinical psychology, the 999 men recruited. Half were offered the therapy. especially cognitive behavioral therapy. Following the therapy program, in another lottery, half received an unconditional cash grant of 200$. Thus, Cognitive behavioral therapy, or CBT, is a common, participants were randomly assigned to receive therapy, short-term psychological intervention that helps people cash grants, both or neither. recognize and reduce unhelpful behaviors and beliefs. This form of therapy actively teaches people new The grant was both a tool to see the effects of the ways of thinking about situations, and also has them therapy on how men spent cash and also an intervention practice new skills and behaviors. Its premise is that in itself. Participants could spend the money however the relationship between thought and behavior goes they wanted. Global Communities (formerly CHF two ways: changing thoughts can change behaviors, but International), an international non-profit organization, practicing new behaviors can also change how a person implemented the cash lottery and distribution. thinks about himself and reacts to events. To assess impacts, IPA surveyed participants at the time For eight weeks, men met with NEPI facilitators in groups of recruitment (before random assignment) and then two of 20, three times a week for about three to four hours weeks, five weeks, 12 months and 13 months after grant a day. On days when the group did not meet, facilitators disbursement. Although the men were extremely mobile occasionally visited the men at their homes or work and difficult to find, IPA successfully tracked 92 percent of them. IPA collected self-reported data on economic we randomly sampled roughly 300 endline surveys and activity, expenditures, investment, and criminal and performed as in-depth, qualitative, community-based anti-social behaviors. IPA staff also carried out games validation of survey responses. Liberian qualitative with real money to test changes in decision-making. researchers spent several hours a day for several days Furthermore, we validated self-reported behaviors talking and hanging out with the men, and validated six through in-depth interviews and direct observation of a types of behavior measured in the survey. We found that random sample of participants the men generally did not underreport stealing, theft, or other “bad” behaviors. If anything, these behaviors were Results underrerported by the control group. Men offered the program also reported being less One of the surprising initial successes was interest and impulsive and less oriented towards immediate attendance. About two thirds of high-risk men agreed rewards in general. In general the men reported to enter the study, and nearly all of those offered the greater self-control, perserverance, and goal orientation therapy attended at least six days of it. Most finished the after the therapy. However, the researchers did not see program. any sustained effect on future-orientation (patience) in economic decision-making. Men offered the therapy reported large and sustained falls in criminal, violent, and other anti- Changes were much stronger when therapy was social behaviors both in the short run (a few weeks) coupled with the cash. The reductions in anti-social and in the long run (a year later). For instance, after one behavior and improvements in self control were larger, year, incidents of theft were one quarter lower in the more sustained, and more statistically significant therapy group than the control group. Drug selling was among those who received cash as well as therapy. The one third lower. Interpersonal disputes and carrying a qualitative data and CBT theory suggest that the cash weapon (usually a knife) fell by nearly half compared to bought men time to practice being productive members the control group. All of these effects were largest when of society, and thus served to reinforce the change in therapy was followed by cash. character and values promoted by the therapy. Men did not underreport “bad” behaviors in the The cash grant led to an initial rise in petty business surveys. To investigate whether men who received and earnings. Initially, earnings rose as the men therapy were falsely underreporting negative behaviors, engaged in petty buinsess. In the weeks and months after the grant was administered, the men lived a little and cash had independent effects, it appears the self- better, with better housing and clothing. Even this high control and values fostered by the therapy are specific risk group tended to use cash productively and wisely. to anti-social behaviors and social life, not economic decision-making. Men offered and not offered the therapy saved, invested and spent cash similarly. Whether they The effect of cash alone on poverty didn’t last. A year received the therapy or not, most of the men saved later the men who received cash had the same assets and invested the cash grant (which was equal to about and earnings as men who did not. Qualitative evidence three months earnings). Little of the grant was spent on suggests that many businesses failed after a major theft, alcohol or drugs or otherwise misused. Because therapy confiscation, loss, or bad deal. Conclusions In Liberia, behavioral therapy programs have the cash transfers may have potential and should be potential to reduce crime and anti-social behavior, explored further. The qualitative stories of business especially in combination with cash transfers. The failure suggest that, in addition to an absence of capital, results suggests that character skills, self-image, and these men’s ability to increase their income is held back values are malleable into adulthood, and that even short, by the poor protection of property rights and high levels non-expert programs of cognitive behavioral therapy of risk without insurance. can be an effective tool for instilling such changes. Although the cash did not have a long-term impact, More research is needed to test and optimize these supposedly irresponsible men demonstrated an ability programs in other contexts, but a body of evidence now to save and invest the cash. They spent little on drugs suggests that psychosocial therapy, when done right, is and alcohol, and lived better for a short time, meaning a useful tool for reducing crime in various settings. This project was led by Christopher Blattman (Columbia University SIPA & Political Science) Julian C. Jamison (U.S. Consumer Financial Protection Bureau), and Margaret Sheridan (Harvard Medical School & Boston Children’s Hospital). The program and study were funded by the National Science Foundation (SES1317506-), the World Bank's Learning on Gender and Conflict in Africa (LOGiCA) trust fund, the World Bank's Italian Children and Youth (CHYAO) trust fund, the UK Department for International Development (DFID) via the Institute for the Study of Labor (IZA), a Vanguard Charitable Trust, and the American People through the United States Agency for International Development’s (USAID) DCHA/CMM office. The Robert Wood Johnson Health and Society Scholars Program at Harvard University (Cohort 5) also provided financial support for Sheridan. The contents of this study are the sole responsibility of authors and do not necessarily reflect the views of their employers or any of these funding agencies or governments. Innovations for Poverty Action oversaw the fieldwork in Liberia, and Global Communities and a Liberian NGO, the Network for Empowerment and Progressive Initiatives (NEPI), implemented the interventions.