The heterogeneous impact of conditional cash transfers

Central America

Honduras

Social Protection

Conditional Cash Transfers (CCT)

S Galiani; PJ McEwan

2013

Galiani, Sebastian, and Patrick J McEwan. “The Heterogeneous Impact of Conditional Cash Transfers.” Journal of public economics 103 (2013): 85–96.
View publication

Secondary Data: 2001 Honduran Census data. Available at: https://geodata.mit.edu/catalog/nyu-2451-34924.

2013

Programa de Asignación Familiar (PRAF), or Family Allowances Program

2000-2002 (beginning and end of intervention).

Improve health and education outcomes of poor families in Honduras.

Children between 6 and 12 who were enrolled and regularly attending grades 1 to 4 between the school year of February and November, up to the completion of the fourth grade. A health transfer of 644 Lempiras per year (about US$40) was available to children under 3 and pregnant mothers who regularly attended health centers. Households were eligible to receive up to 3 education transfers and up to 2 health transfers.

During the two years of implementation, transfers were distributed in November 2000, May-June 2001, October 2001, and late 2002. The transfers were locally distributed as cash by PRAF personnel. In addition to CCTs, two kinds of direct investments in schools and health centers were done. The education interventions consisted of payments of approximately US$4,000 per year, depending on school size, to parent associations in primary schools. The payments were conditioned on obtaining legal status and preparing a quality-improvement plan. The health interventions consisted of payments of approximately $6,000 per year to local health centers, depending on the client base. The health payments were conditioned on the formation of a health team (with members of the community and health personnel) and the preparation of a budget and proposal.

• Enrollment in school
• Child labour (household chores and outside the house)

RCT

Study design: to identify the experimental sample, 298 municipalities were ordered from lowest to highest values of the mean height-for-age z-score of first graders, obtained from the 1997 Height Census of First Graders. Seventy-three eligible municipalities had z-scores below a cut-off of −2.304. Of these, 3 were excluded because of distance and cost considerations, yielding a final sample of 70 municipalities.
Sample: The geographic concentration of child stunting produced a sample dominated by western Honduras. Three treatment groups and one control group were identified. G1 was to receive CCTs in education and health. G2 was to receive CCTs in addition to direct investments in education and health centers, while G3bwould receive only direct investments. G4 would receive no interventions. The 70 municipalities were divided into five quintiles of 14 municipalities, based on mean height-for-age. The random assignment occurred on October 13, 1999, during a public event. Within each quintile, 4 municipalities were randomly assigned to G1, 4 to G2, 2 to G3, and 4 to G4. The final sample consisted of 20 municipalities in G1, 20 in G2, 10 in G3, and 20 in G4.
Survey Instrument: The paper uses the 2001 Honduran Census, rather than the official evaluation data.

Eligible children were 8 percentage points more likely to enroll in school and 3 percentage points less likely to work with the implementation of the Honduran PRAF experiment. The effects were much larger in the two poorest strata, and statistically insignificant in the other three (the latter finding is robust to the use of a separate regression-discontinuity design). Heterogeneity confirms the importance of judicious targeting to maximize the impact and cost-effectiveness of CCTs.

Child Labour

Households

National

Eligible children were 3 percentage points less likely to work with the implementation of the Honduran PRAF experiment.

Eligible children were 8 percentage points more likely to enroll in school with the implementation of the Honduran PRAF experiment.

There is no consistent evidence of effects on ineligible children or adult labor supply