The influence of conditional cash transfers on eligible children and their siblings

Central America

Nicaragua

Social Protection

Conditional Cash Transfers (CCT)

JA Lincove; A Parker

2016

Lincove, Jane Arnold, and Adam Parker. “The Influence of Conditional Cash Transfers on Eligible Children and Their Siblings.” Education Economics 24.4 (2016): 352–373.
View publication

https://www.povertyactionlab.org/evaluation/long-term-effects-conditional-cash-transfer-program-nicaragua.

2016

Red de Proteccion Social (RPS)

2000-2002

The objectives of the RPS were to supplement income, reduce school dropouts, and improve the health and nutrition of children. The objective of the study was to evaluate whether a large cash transfer would induce measurable income effects that are not offset by displacement.

Three types of transfers: 1) food security transfer (the Bono alimentario): households with children between 7 and 13 years old - this amount was fixed per household regardless of the number of eligible children; 2) school attendance transfer (the Bono escolar), which provided a fixed amount per household per year contingent on enrollment and regular school attendance for each child aged 7–13 years who had not yet completed fourth grade; 3) school suppliers transfer, eligible children include those of ages 7–13 (fixed amount per child).

A series of three transfers were provided to eligible households. The first of these was a food security transfer (the Bono alimentario), which provided US$224 per household per year. This transfer was contingent on attendance at monthly health education workshops and bringing children to well-child checkups (monthly for children under 2 and every two months for children 2–5). The second transfer was a school attendance transfer (the Bono escolar), which provided US$112 per household per year contingent on enrollment and regular school attendance for each child aged 7–13 years who had not yet completed fourth grade. This amount was fixed per household regardless of the number of eligible children. Finally, the program provided a school supplies transfer of US$21 per eligible child per year, also contingent on school enrollment. In this case, eligible children include those ages 7–13, and ineligible children of age 14 and up.

• School attendance
• Hours of work (child labor)

RCT

Study design: The first empirical strategy was to estimate the direct effects of RPS on eligible and ineligible children, as well as the indirect effects of sibling eligibility. The authors observed the latter effects by comparing households with and without siblings of different ages. In addition to predicted treatment effects, they expected that school enrollment and child labor would change as children age regardless of the treatment. The authors estimate the effects of the program through difference-in-differences (DID) models that estimate the effects of the treatment on changes in outcome measures for the treated. The first difference is the change over time, and the second difference is the gap in outcomes between the treated and untreated. The second empirical strategy examines the effects of sibling eligibility. To do this, the authors divided the sample into children who were eligible for RPS by age and those who were not. They also identify the presence of eligible and ineligible siblings by age and include dummy variables to indicate the presence of each type of sibling. They added interactions between the post-treatment variable and the presence of an eligible sibling and the presence of an ineligible sibling. These interaction effects measure the effects of RPS through sibling eligibility. Children without siblings provide the appropriate comparison within treatment and control groups as any indirect effects of RPS should be absent in these families. While RPS may influence the family composition, it is reasonable to assume that the composition of school-aged siblings is exogenous, since these children were born at least six years before the program.
Sample: Forty-two eligible comarcas (communities) were selected based on a high poverty rate and sufficient administrative and school infrastructure to successfully implement the program. Comarcas were randomly assigned to treatment and control groups during the pilot period from 2000 to 2002. During this time, the International Food Policy Research Institute conducted an annual survey of households with children in all 42 comarcas to gather baseline and follow-up data. The data set included 2302 children ages 6–17 years old during the baseline year, who matured to ages 8 to 19 years during the two-year implementation. A small number of children were excluded if they were married, had children, were not residing in the household, or had missing data.
Survey Instrument: One baseline and two follow-up surveys. The surveys ask whether the child attended school, whether the child worked in the previous week, and weekly hours worked.

CCTs may have the largest effects on eligible children, but older brothers also benefited through increased schooling and fewer hours worked. The results of the study suggest that the income effects of CCTs apply to both eligible and ineligible children.

Child Labour

Households

National

CCTs may have the largest effects on eligible children, but older brothers also benefited through fewer hours worked and increased schooling in Nicaragua.

The results of the study suggest that the income effects of CCTs apply to both eligible and ineligible children.

The CCT of Nicaragua is the largest as a percent of income (approximately 20%). In addition, the rate of child labor is quite high and the rate of schooling for ineligible children is lower than the rate for eligible children. This provides the best opportunity to observe both large income effects and large labor displacement effects.