94711 from EVIDENCE to POLICY Learning what works for better programs and policies UPDATED: March, 2015 LIBERIA: Does Giving People the Chance to Work Reduce Conflict and Boost Peace? Nearly one billion people will enter the labor force be- and the United Nations developed an innovative pro- tween now and 2030—the deadline that World Bank gram to help ex-combatants move into full-time farm President Jim Kim set for ending extreme poverty. Help- work by giving them training, counseling, and start-up ing young adults find capital. The World Bank, which is committed to help- productive work and ing countries improve work opportunities as part of giving them the tools the United Nations Millennium Development Goals, they need to succeed is helped support a randomized impact evaluation of the SKILLS AND JOBS crucial to meeting this program. The evaluation was carried out by researchers goal. But what hap- with Innovations for Poverty Action. pens when the young The evaluation found that the program successfully adults are mainly shifted high-risk men from criminal activities into farm- skilled in armed con- ing. Graduates earned more money than their counter- flict and illegal money-making activities and the countries parts who weren’t enrolled in the program, spent less where they live have devastated economies and few work time in illegal work and were less likely to consider opportunities? Development experts and policymakers fighting as mercenaries in neighboring conflicts. The are exploring options for improving employment oppor- evaluation also showed that skills training isn’t always tunities through programs that can boost job earnings enough—men who received training but didn’t get their while also reducing the likelihood of social instability. start-up capital didn’t do as well as those who did. As In Liberia, where civil wars were fought on and off policy makers in the region look to strengthen their between 1989 and 2003, the non-profit organization, economies and boost stability, the results of this evalu- Action on Armed Violence, the Government of Liberia, ation offer guidelines for crafting successful programs. Context By the end of the civil war in 2003, the majority of Libe- many years of war, often had little, if any, schooling. ria’s three million citizens were displaced and the coun- Although the security situation slowly improved and try had been devastated. People were poor and, after so the economy expanded, by 2008 there were still close to 10,000 ex-fighters living in remote, ungoverned parts of Participant Profile: the country. They earned money through illegal activi- • On average, 30 years old • $46 in savings ties such as unlicensed mining and rubber tapping. • 27 percent were literate • $7 in debts Action on Armed Violence, a non-profit based in • 71 percent lived with their • 47 percent spent some time spouse in potentially illicit work Britain, wanted to see whether it was possible to shift • Monthly cash earnings • 33 percent said this was former combatants from illegal work to steady agricul- (before program) of $47 their main source of income tural labor, the main source of income for most rural Liberians. There was also the hope that the program, implemented in cooperation with the government’s by the World Bank’s Strategic Impact Evaluation Fund, Ministry of Agriculture, would reduce the likelihood the Italian Children and Youth Trust Fund, and the of ex-fighters returning to combat by giving them legal United Nations Peacebuilding Fund, was built into the and sustainable work. An impact evaluation, supported program to measure the effect. Evaluation Between May and October 2009, Action on Armed Vio- gram, and the other half after they had started farming. lence recruited applicants from 138 communities with An unexpected supply issue related to the start-up high concentrations of ex-combatants. The program capital provided to participants created an important had a limited number of slots, making it possible for variation between the groups. Those who focused on researchers to create a control group among those who animal husbandry received training, but didn’t receive couldn’t be accommodated despite qualifying. In total, the promised pigs and chickens by the endline survey. SKILLS AND JOBS 1,123 men were enrolled in the study, with 640 men Instead, they were promised an equivalent cash trans- randomly assigned to be in the treatment group. A small fer (about $100), which was supposed to be transferred number of high-ranking officers were enrolled —it was within a few months after the survey. Participants who considered risky to leave them out of the program—but specialized in crop farming received the training and got they were not counted for the study. their promised start-up seeds and other materials in time. Men in the treatment group were placed in one of As there were almost no observable differences between two residential programs: Bong County in western Li- the two groups at baseline, this provided an opportunity beria or Sinoe County in eastern Liberia. They received to evaluate the effect of providing start-up capital and four months of agricultural training centered on grow- training versus providing training alone. In other words, ing rice and vegetables and animal husbandry, in addi- researchers were able to also conduct an additional “as- tion to basic literacy classes. A built-in “life skills” class if ” random evaluation. sought to teach participants non-violent approaches to A baseline survey was conducted between May and conflict. Former combatants who had trained as mentors October 2009. The endline survey was conducted from provided conflict management sessions, as well as more February to June 2011, 18 months after baseline and 14 informal counseling. When the program finished, par- months after training was finished. The qualitative evalu- ticipants received about $125 worth of start-up agricul- ation included extensive conversations with participants tural materials—seeds and tools to those men wanting to and all the information was self-reported. More than 90 specialize in crops, and animals and building materials percent of participants in both control and treatment for those wanting to specialize in pig or poultry raising. groups were available for the endline survey. Some of the materials were given at the end of the pro- The Findings Men who attended the program spent more control group. They were 15.5 percentage points more time farming and… likely to be currently farming, up from 61 percent for the More than a year after completing the program, partici- control group. Participants in the program were also 35 pants were 14.1 percentage points more likely to have percentage points more likely to grow their own seedlings farmed the previous season, up from 69 percent for the and 12 percentage points more likely to have sold crops. … fewer hours every month engaged in crop raising and therefore received their capital. Hours of potentially illegal activities. agricultural work did not increase among men who trained Participation in the program didn’t end illicit activities to become animal specialists. Start-up capital therefore ap- completely—around 37 percent of men still engaged in peared to be crucial to the men’s ability to successfully shift such activities, a small decline compared with the control into new occupations. group. Nor did it lead to any change in the overall num- ber of hours worked in the previous month. However, But the program didn’t appear to reduce men who had gone through the program were spending aggressive behavior or violence. 33 percent more of their time farming and 24 percent less The program included a life skills component and coun- in potentially illicit work, such as rubber tapping or unli- seling intended to help participants curb aggressive be- censed gold and diamond mining. In short, men shifted havior, such as physical fights and destroying property, their portfolio of activities from illegal to legal ones. They also showed less interest in being recruited as mercenaries. An impetus for the program was to reduce the like- lihood that ex-combatants would return to fighting, whether in Liberia or as mercenaries in a neighbor- ing country. When the civil war ended in 2003, the United Nations ran a traditional Disarmament, De- mobilization and Reintegration program, giving cash and training vouchers to former fighters. However the lack of legal opportunities limited the impact and when the program ended in July 2009, thousands of former fighters still lived in remote areas, involved in illegal or unlicensed work. and also taught methods of non-violent conflict reso- Action on Armed Violence hoped that a more di- lution. The study showed no effect of the program on rected program, and one that specifically trained men these behaviors, however, suggesting that the life skills to enter the country’s dominant agricultural sector, component was not sufficient to change behavior, and would reduce the chances of criminal or armed ac- that the economic incentives alone likely led to the shift tivities. In fact, participants reported less interest in away from illicit and mercenary work. One possibility is joining the war that broke out in neighboring Cote that behavior was changed in a temporary and artificial d’Ivorie in December 2010. They were 43 percent environment, and may not have survived the transition less likely to attend a secret recruitment meeting and to a new village life. 51 percent less likely to consider joining, even at a going recruitment rate of $1,000. Nor did the program affect existing chains of command and ex-combatant relationships. Ensuring that program participants received The researchers hoped that relocating ex-combat- the agricultural supplies they were promised ants to a new village for training would encourage turned out to be important to success. them to sever ties with their former commanders, The shift from illegal resource extraction to farming was but the program didn’t seem to break down existing concentrated among the young men who specialized in chains of military command. The men in the pro- This policy note is based on the report “Can Employment Reduce Lawlessness and Rebellion? A Field Experiment with High-Risk Youth in a Fragile State,” May 2014, Christopher Blattman (Columbia University) and Jeannie Annan (International Rescue Committee) in collaboration with Innovations for Poverty Action. gram continued to socialize combatants and fewer ranking officers could poten- with other ex-combatants, tially also break armed group linkages and further but that was also because deter mercenary activity. of the program’s design: The residential component The program seems to make economic sense of the program meant that when the higher incomes are factored in. participants spent all of Assuming a permanent income boost of almost $11 a their time with other high- month, the per person program cost of $1,275 would be risk men. This implies that paid back in nine years. This doesn’t take into account the reduction in mercenary national economic gains from a drop in illicit activity, in activity likely arose from the economic incentives addition to the potentially stabilizing effect in the coun- alone rather than by disconnecting men from risky try and in neighboring countries from a lack of steady social networks. A program that included more non- recruits for mercenary work. SKILLS AND JOBS Conclusion In post-conflict countries, work training programs have But the findings also shed light on some of the con- the potential to help rebuild a shattered economy, pro- straints that youth face when trying to pursue work vide job opportunities, and boost peace-building efforts. opportunities: Training alone may not be enough Indeed, as this impact evaluation shows, training and if it isn’t supplemented with start-up capital. These access to capital does help young people shift away from findings will likely help development experts as they illegal activities into legal, non-violent, and productive continue to create programs that help the millions of work. This should be welcome news for policy makers young people entering the labor force find meaningful in post-conflict regions looking to rebuild the econo- and productive work. mies—and social fabric—of their countries. The Strategic Impact Evaluation Fund, part of the World Bank Group, supports and disseminates research evaluating the impact of development projects to help alleviate poverty. The goal is to collect and build empirical evidence that can help governments and development organizations design and implement the most appropriate and effective policies for better educational, health and job opportunities for people in developing countries. For more information about who we are and what we do, go to: http://www.worldbank.org/sief. The Evidence to Policy note series is produced by SIEF with generous support from the British government’s Department for International Development. THE WORLD BANK, STRATEGIC IMPACT EVALUATION FUND 1818 H STREET, NW WASHINGTON, DC 20433 Produced by the Strategic Impact Evaluation Fund Series Editor: Aliza Marcus; Writer: Daphna Berman