Absence of conflict of interest.
Citation
Benedetti, F., Ibarrarán, P., & McEwan, P. J. (2016). Do education and health conditions matter in a large cash transfer? Evidence from a Honduran experiment. Economic Development and Cultural Change, 64(4), 759-793. doi:10.1086/686583
Highlights
- The study’s objective was to examine the impact of a conditional cash transfer on school enrollment and child labor.
- The study was a randomized controlled trial conducted in in low-income villages in Honduras. Villages were randomly assigned to the treatment and control groups. Baseline and follow-up survey data were collected from randomly selected households and analyzed using regression.
- The Bono 10,000 program significantly increased school enrollment but did not significantly reduce child labor.
- The quality of causal evidence presented in this report is high because it was based on a well-implemented randomized controlled trial. This means we are confident that the estimated effects are attributable to the Bono 10,000 program and not to other factors.
Intervention Examined
Bono 10,000
Features of the Intervention
The Bono 10,000 program is a conditional cash transfer program operating in Honduras. It began in 2010 when a newly elected administration renamed and modified the former Programa de Asignacion Familiar (PRAF-II) conditional cash transfer programs. The Bono 10,000 program provided cash transfers to households in the 2,911 poorest of Honduras’ villages located primarily in the Western part of the country.
After signing a letter of commitment, households received Honduran Lempira (L) 10,000 (about $500 USD) if at least one child aged 6-18 enrolled in school. This was true regardless of how many children in the age range lived in the household. Households received L 5,000 ($250 USD) if children under age 6, and pregnant or nursing mothers, registered in a health center. The money was paid in three installments. The first was an unconditional payment of 1/12 the total amount after eligibility was determined. The second and third installments were contingent on school enrollment for at least one child 6-18, or registration in a health center for children under age 6 or pregnant or nursing mothers.
Features of the Study
The study used a randomized controlled trial to examine the impact of Bono 10,000 on school enrollment and child labor. The sample was drawn from 816 untreated, and slightly less poor, villages. Villages were assigned to treatment and control groups by selecting numbered balls randomly, alternating assignment to treatment and control groups, until 300 villages were selected (150 treatment and 150 control). To select households within the villages, the Honduran government conducted a household census, then applied a proxy means test to obtain a list of "nominally poor" households. Researchers at NORC at the University of Chicago randomly sampled poor households from each village's list.
Baseline household surveys were conducted between January and June of 2012 and follow-up surveys were conducted between March and June of 2013. In the baseline survey, there were 2,137 households in the treatment group and 2,108 in the control group. In the follow-up survey, there were 1,973 households in the treatment group and 1,866 in the control group Child labor was measured by assessing if the child participated in paid or unpaid work in the week before the follow-up survey. School enrollment was measured by assessing if the child attended school in the week before the follow-up survey. Impacts were assessed using regression, with multiple controls (household size, age, child birth order) and interaction effects (enrollment in different grade levels and poverty). Data were analyzed using regression. While the program included 18-year-old individuals, analyses were restricted to children aged 6-17 for school enrollment and child labor participation.
Findings
Employment/Child labor
- The study found that the Bono 10,000 program did not significantly impact child labor.
Education (School participation/enrollment)
- The study found that the Bono 10,000 program significantly increased school enrollment by 4.4 percentage points.
Considerations for Interpreting the Findings
The authors note that the government’s proxy means test used to determine eligibility based on poverty allowed "substantial leakage of nonpoor households into the sample." Additionally, 49 villages included in the study were located in municipalities served by the earlier PRAF-II program, which began in the late 1990s. The authors do not indicate the distribution of the 49 villages between the treatment and control groups. Preexisting differences between the groups—and not the Bono 10,000 program— could explain the observed differences in outcomes.
Causal Evidence Rating
The quality of causal evidence presented in this report is high because it was based on a well-implemented randomized controlled trial. This means we are confident that the estimated effects are attributable to the Bono 10,000 program and not to other factors.