Absence of conflict of interest.
Citation
Helfand, S.M., & Souza, A.P. (2010). The impact of conditional cash transfer program on human capital formation in Brazil. University of California, Riverside, CA: Sao Paulo School of Economics.
Highlights
- The study’s objective was to analyze the impact of Brazil’s Bolsa Escola program on school enrollment, school progression, and child labor for children ages 6 to 15 in rural areas.
- Using data from the 2001 and 2003 Brazilian Census, the authors used a structural model to examine the average program impact of the Bolsa Escola program.
- The study found that the Bolsa Escola program was significantly associated with an increase in school enrollment and school progression. However, the study did not find a statistically significant relationship between program participation and child labor.
- The quality of causal evidence presented in the study is low because the authors did not control for pre-intervention outcomes. This means we are not confident that the estimated effects are attributable to the Bolsa Escola program; other factors are likely to have contributed.
Intervention Examined
Bolsa Escola
Features of the Intervention
The Bolsa Escola Program was a conditional cash transfer program for poor families in urban areas of Brazil. The program gave cash grants to poor families with children between ages 7 and 14 conditional on school enrollment (15 Reais per month, equivalent to $6 USD). Households lose a month of benefits if their children miss too many days of school per month (the number of days is not specified in the article).
Features of the Study
The authors used a non-experimental analysis. The study measured the impact of Brazil's Bolsa Escola Program on school attendance, school progression, and child labor in rural Brazil. The study used the 2001 and 2003 Brazilian Census to collect data on outcomes for children ages 6 to 15 who received the program. The Census has information about families who participated in a social program. The authors used the information on program participation to form a treatment group. The sample consisted of all families with at least two children, aged 6 to 15, who received the program. The authors used ages 6 to 15 to include baseline data for children who started the program at age 7 and endline data for children who were age 14 when starting the program. There was no comparison group. The study used a structural model to estimate program impacts. The model controlled for characteristics of the child and the household.
Findings
Employment/Child labor
- The Bolsa Escola program was not significantly related to child labor.
Education (School participation/enrollment)
- The study found a significant positive relationship between the program and school attendance. The program raised the probability of school attendance by 10 percentage points among children in a two-child family in rural areas, and by 0.5 percentage points in urban areas.
- The Bolsa Escola program was significantly related to school progression. The program reduced the delay in school by 0.4 years among rural children in a two-child family and 0.115 years among urban children.
Considerations for Interpreting the Findings
The study used data from the Brazilian Census to identify children who received the program. The Census question does not explicitly ask about participation in a named program. Rather, it asks if the child participates in a social program conditional on school enrollment. The authors assumed that the social program was Bolsa Escola.
Although the structural model controlled for the child’s age, gender, and household socioeconomic status, the authors did not account for child outcomes at baseline. The children might be enrolled in school prior to the intervention. Also, the families chose to participate in the program rather than being randomly assigned to the program. It is possible that unobservable factors, such as individual motivation to attend school, might have contributed to the estimated effects.
Causal Evidence Rating
The quality of causal evidence presented in this study is low because the authors did not control for pre-intervention outcomes. This means we are not confident that the estimated effects are attributable to the Bolsa Escola program; other factors are likely to have contributed.