Absence of Conflict of Interest.
Citation
Lopez-Calva, L. F., & Patrinos, H. A. (2015). Exploring the differential impact of public interventions on indigenous people: Lessons from Mexico's conditional cash transfer program. Journal of Human Development and Capabilities, 16(3), 452-467. doi:10.1080/19452829.2015.1072378
Highlights
- The objective of the study was to examine the impact of Mexico’s conditional cash transfer program (PROGRESA/Oportunidades) on child labor and school attendance for children ages 8 to 17 across 3 indigenous, Spanish-speaking, and bilingual households.
- Using survey data from a cluster randomized controlled trial, the authors analyzed the average program impact of the PROGRESA/Oportunidades program on indigenous households versus individuals who were bilingual or Spanish-speaking.
- The study found that the PROGRESA/Oportunidades program was significantly related to a decrease in child labor and an increase in school attendance for indigenous speaking children.
- The quality of causal evidence presented in this study is low because it was based on a randomized controlled trial with unknown attrition and the authors did not control for the required variables. As a result, we are not confident that the estimated effects are attributable to PROGRESA/Oportunidades; other factors are likely to have contributed.
Intervention Examined
PROGRESA/Oportunidades
Features of the Intervention
Since 1997, the PROGRESA/Oportunidades program has provided monthly conditional cash transfers, which is approximately 20 percent of the average total household income. By 2012, PROGRESA/Oportunidades had served 5.8 million households in Mexico. The cash transfer is conditional upon child school attendance (85% of days) for children aged 8-17, and health clinic visits for household members, as well as the requirement for pregnant women and lactating women to take nutritional supplements and go to five pre-natal visits. The amount of the cash transfer varies by gender as girls are more likely to drop out of school at younger ages and increased with the grade of the child. However, if a child repeats a grade twice, he or she will permanently lose the cash transfer.
Features of the Study
The study used survey data from a randomized controlled trial across rural communities in seven states in Mexico (320 in the treatment; 185 in the control). Poor households, based on a marginality index, were selected to participate in the program. In this study, authors used the 1997 baseline survey of Household Socioeconomic Characteristics (ENCASEH) as well as the Mexican national household survey, Encuesta de Evaluacion de los Hogares, in 2000, 1999, and 1997. The authors use multinomial logit, sequential probit, and difference- in-difference regressions, looking at differences between individuals who speak the indigenous language, those were bilingual, and those who were Spanish-speaking on the outcomes of child labor and school attendance. The difference-in-differences (DID) model explored the effect of PROGRESA/Oportunidades on the three language groups (indigenous, Spanish, and bilingual).
Findings
Employment/Child labor
- The study showed that participation in PROGRESA/Oportunidades was significantly related to a decrease in the probability of working among children aged 8-17 in all language groups, with the largest reduction among indigenous speaking children.
Education (School participation/enrollment)
- The study showed that participation in PROGRESA/Oportunidades was significantly related to an increase in the probability of school attendance among children aged 8-17 in all language groups, with the largest increase among indigenous speaking children.
Considerations for Interpreting the Findings
The authors do not include attrition data for the randomized controlled trial. In cases of high or unknown attrition, a study can receive a moderate causal evidence rating if the analysis controls for possible differences in background characteristics of the analytic treatment and control groups. For the difference-in-differences model, the authors include a vector of household characteristics but not child gender.
Causal Evidence Rating
The quality of causal evidence presented in this study is low because it was based on a randomized controlled trial with unknown attrition and the authors did not control for the required variables. This means we are not confident that the estimated effects are attributable to PROGRESA/Oportunidades; other factors are likely to have contributed.