from EVIDENCE to POLICY Learning what works, from the Human Development Network May, 2013 77086 Can Disadvantaged Kids Ever Catch up with Better-Off Peers? Early Childhood Development What’s the best way to help disadvantaged children The World Bank is focused on developing and reach their potential? What do they need to succeed in supporting programs that help children reach their school, in work and in a family? How can they receive potential and live lives free of poverty. To help build the necessary build- a body of evidence of what works, the World Bank ing blocks for a happy financed an evaluation of a program in Jamaica that and productive life, targeted mothers of babies stunted due to malnu- one free of poverty? trition. The mothers received either support and For policymakers and guidance on how to encourage their babies’ devel- development experts, opment through play and language, or nutritional the answers lie in early supplements, or a combination of the two. Twenty childhood develop- years later, the evaluation found that children who ment, when children’s received the extra stimulation, whether with nutri- brains and bodies tional supplements or not, were earning more money are still developing. than similarly stunted babies whose mothers received Proper healthcare, nu- just nutritional supplements or no intervention. The trition, psychosocial children whose mothers had received the extra guid- stimulation, and emotional support all play a role in giv- ance also were doing as well financially as the less dis- ing children the foundation they need to do well later on. advantaged (and non-stunted) children. This study, But what happens after? Do programs designed to bolster a rare look at the effects of early childhood interven- disadvantaged children’s cognitive, emotional and physi- tion over the decades, gives policymakers and devel- cal development really help over the long term? Or are opment experts tangible proof of the potential effects gains seen in the early years lost by adulthood? of early childhood development programs. Context The importance of early childhood development is well- through the SIEF trust fund are helping build evidence known. Children who don’t receive the proper nutri- of what works to give poor children the same advantages tion, stimulation, and emotional support in early years as their better-off peers: For example, in Mozambique, are more likely to lag behind more advantaged peers pre-schools were shown to be effective at boosting chil- even before they start school. Once in school, they are dren’s emotional, cognitive, and physical readiness for less likely to do as well and more likely to drop out be- school and getting them to start school at the right age fore graduating, hurting their chances to succeed later (see “Is Preschool Good for Kids,� Evidence to Policy in life. Impact evaluations financed by the World Bank note, March 2012). Evaluation In 1986-1987, a Jamaican study enrolled 129 children would be shared with other family members, families also aged 9-24 months and living in poor neighborhoods of were given supplementary milk powder and cornmeal. the capital, Kingston. The children, who were all stunted Two participants dropped out, leaving 127 to be based on height-for-age measurements, were stratified by surveyed at baseline, after two years, and again at aged age and sex, and then split into four groups. One group 7, 11, and 18. In 2007-2008, when the original partici- Early Childhood Development received psychosocial stimulation, one group received pants were around 22 years old, they were interviewed nutritional supplements, one received both, and the again. In the last survey, researchers were able to find fourth group, the control group, received nothing. The and interview 83 percent of the stunted children in the treatment continued for two years. Children in all groups program. received free access to health care. During the same intervals, researchers also surveyed Households in the psychosocial stimulation treatment a group of non-stunted and less disadvantaged children group received weekly, hour-long visits from trained com- from the same neighborhood. Non-stunted children munity healthworkers, who taught mothers how to play were deemed less disadvantaged based on a variety of educational games with their children and encouraged them measurements: they had taller mothers with higher vo- to converse with their children. They also were encouraged cabulary scores, higher birth weights, larger head cir- to praise their children and improve the self-esteem of the cumferences, and higher developmental scores on initial child. Homemade toys were brought to each visit and ex- tests. At the 20-year mark, 77 percent of the original changed the next week for other toys. Families that quali- non-stunted children were located and surveyed. fied for the nutritional supplement received one kilogram For more details on methods used, please see the of fortified formula. To reduce the chances that the formula original paper.* Findings Nearly 20 years after poor mothers were children whose families received nutritional supplements shown how to play and interact with their alone. In fact, they were doing so well that their earnings had children in ways that promote cognitive, caught up with earnings of their less-disadvantaged peers. physical, and emotional development, the gains were apparent. While giving mothers parenting support to aid early childhood development had a big impact, Children in this treatment group—stimulation or stimula- the nutritional supplements alone didn’t have tion plus nutritional supplements—had significantly higher any long-term effect. earnings as young adults than the control group, regardless of whether they held a part-time, full-time or permanent The supplements were designed to help offset nutritional de- job. They were doing equally better when compared with ficiencies that may have led to the stunting and contained 66 *This policy note is based on “Labor Market Returns to Early Childhood Stimulation: a 20-year Follow up to an Experimental Intervention in Jamaica,� by Paul Gertler, James Heckman, Rodrigo Pinto, Arianna Zanolini, Christel Vermeersch, Susan Walker, Susan Chang, Sally Grantham-McGregor. Research was supported, in part, by SIEF. percent of daily recommended calories and 100 percent of not stunted as babies. Based on an analysis of earlier daily recommended protein. Although families also received surveys at age 18, the now-young adults scored higher extra milk powder and cornmeal, in hopes of minimizing on tests measuring cognitive abilities, including math, sharing of the formula, the supplements were shared, diffus- reading, and verbal skills, and socio-emotional skills, ing their effect. But the follow-up surveys did not show that giving them the tools needed for doing well in school, children in families that received nutritional supplements jobs, and relationships. were doing any better than the control group. Because the These children—now young adults—had more nutritional supplement-only group didn’t show any schooling than the control group and did better on the long-term differences with the control group, the au- exams that students in Jamaica take in high school and thors group them together, and the phrase “control those taken to attend college. They were three times more group� now refers both to the control and the nutri- likely to have some college education and, at the time of tional supplement-only group. the survey, they were twice as likely to still be in school and three times more likely to be a full-time student. Children in the stimulation treatment groups earned more money than the control group, even after the effect of relatively higher wages for those who had migrated was removed. They also managed to catch up to their non-stunted peers on earnings. The differences in earnings of children in the stimulation treatment groups compared with the control group were marked. Average monthly lifetime earnings for young adults in permanent jobs (as opposed to temporary work- ers) were 60 percent higher than that of the control group. Overall, earnings were 49 percent higher for all workers in the stimulation treatment group, compared with the control group. When the effect of wages of those young adults who had migrated to the United States, Britain, or As was the case with earnings, the treatment group Canada was removed from the calculations, average life- caught up with their less disadvantaged peers in edu- time earnings were still higher: 38 percent higher across cation and socio-emotional skills, even though they all jobs and 45 percent higher for full-time non-tempo- didn’t completely catch up in cognitive skills. rary jobs. Interestingly, those in the stimulation treatment groups were not any more or less likely to be working Children in the study were also more likely to than those in the nutritional-only or control groups. have emigrated to the U.S. or to Britain, which may have affected earnings (as described earlier). Children in the treatment group also were more educated and did better on cognitive Compared with the control group, children in the tests, two areas that help with earnings. treatment group were 10 percentage points more likely to have emigrated, according to the baseline sample. Young adults whose mothers received advice on how When only the follow-up sample is considered, the to interact with their children did much better based rate of migration was the same as that for the non- on tests of cognitive function than those in the control stunted children in the comparison group. Migration group—in fact, they did as well as children who were clearly is a way to attain better earnings. The treatment was designed to improve mother- while the score of the stunted group was significantly child interaction during the critical early years. It lower than the non-stunted. At the end of the two- succeeded and that early success appears to be year treatment period, the HOME score of the treat- the reason behind the gains later in life. ment group was higher than the control group, and had caught up that of the non-stunted group. Using the infant toddler HOME measure, which re- By ages 7 and 11, the home environments of the lies on home observation of various activities related to treatment and control groups were the same when it stimulation, including verbal engagement of caregiver came to stimulation. Given the differences in earnings and availability of play materials, the treatment inter- and education that appeared later between the two vention did raise scores. At baseline, the HOME score groups, it appears that it’s the early childhood inter- of the treatment and control groups was the same, vention that made an impact. Conclusion A rare long-term study of the effects of an early child- For policymakers and development experts looking hood development program shows that children’s lives to improve people’s lives and reduce poverty, the mes- can be improved by ensuring that they have the right sage is clear: parental training groups that engage moth- stimulation and emotional support as babies and tod- ers in activities to promote development can succeed. dler. A simple, two year intervention gave children the What’s key is setting the framework for children to re- foundation to develop the skills that allowed them to ceive the support and skills they need to succeed as stu- earn more later on, catching up even to their less-disad- dents and young adults. As the Jamaica study showed, vantaged peers. the pay-off is forever. The Human Development Network, 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, HUMAN DEVELOPMENT NETWORK 1818 H STREET, NW WASHINGTON, DC 20433 Produced by Office of the Chief Economist, Human Development Network, Communications/Aliza Marcus amarcus@worldbank.org