The Effect of Conditional Transfers on School Performance and Child Labor : Evidence from an Ex-Post Impact Evaluation in Costa Rica

Conditional transfer programs are becoming a common approach to influence household decisions. The evidence to date is that these programs are good at promoting certain outcomes such as school attendance, but that other outcomes such as reducing child labor are more difficult to achieve. This study examines the impact of Superémonos, a conditional transfer program in Costa Rica, which provides poor families with a subsidy for the purchase of food conditional upon children regularly attending school. Using three different empirical techniques—simple comparison of mean outcomes, regression analysis and propensity score matching—we examine the program's impact on school attendance, performance in school and child labor. We find strong evidence that the program achieves its goal of improving school attendance and much weaker evidence regarding school performance. The program does not reduce the likelihood that youth will work. These findings are discussed in the context of the results from impact evaluations of other conditional transfer programs