Same Bureaucracy, Different Outcomes in Human Capital? How Indigenous and Rural Non-Indigenous Areas in Panama Responded to the CCT

This paper estimates the impact of the conditional cash transfer program, Red de Oportunidades, on human capital outcomes in areas with different incidences of poverty: indigenous and rural non-indigenous areas of Panama. The analysis relies on data from the Living Standards Measurement Survey of 2008. It uses a propensity score matching technique to identify the impact of the program by replicating the selection criteria followed by the government. Our results show that the program increased school enrollment and was able to reduce child labor in both areas. In rural areas, the impact of the program on education outcomes was restricted to enrollment in middle school where we estimated an increase of 10.2 percentage points (pp). The program also increased the proportion of children that completed elementary school by 13.8 pp, although it did not have an impact on enrollment in elementary or high school. In indigenous areas, the impact of the program on education outcomes was restricted to enrollment in elementary school where we estimate an increase of 7.9 pp. Additionally, the results show that the program reduced child labor in children ages 12 to 15 in both areas: by 10.1 pp in rural areas and 15.8 pp in indigenous areas. With regard to preventive health care services, we found no evidence of impact on the numbers of visits to health care providers or the number of vaccines that children received. However, we estimated that the proportion of women who had a Papanicolau test screening because of the program increased by 11.7 pp in rural areas and 14.7 pp in indigenous areas. Sadly, we found that the program might also have led to an increase in the number of pregnancies in rural areas: the proportion of pregnant women in 2008 was 3.2 pp higher in the beneficiary group than in the control group, and the number of pregnancies since the beginning of the program increased by 0.44 among beneficiaries---despite the fact that the number of children ages 3 to 6, who were born before the implementation of the program, was not statistically different between the groups.