Reducing child labour through conditional cash transfers: evidence from Nicaragua’s Red de Protección Social

Central America

Nicaragua

Social Protection

Conditional Cash Transfers (CCT)

KA Gee

2010

Gee, Kevin A. “Reducing Child Labour Through Conditional Cash Transfers: Evidence from Nicaragua’s Red de Protección Social.” Development policy review 28.6 (2010): 711–732. Web.
View publication

Secondary Data: Red de Protección Social (RPS) Evaluation Dataset, 2000-2002. Available at: http://www.ifpri.org/dataset/nicaragua.

2010

Red de Protección Social (RPS)

2000-2002 (range of the dataset)

The purpose of the paper is to present the results of an analysis of the Red de Protección Social and compare them with those of other CCT programs, discussing how the structure of each program’s incentives, including differences in targeting, subsidy amounts, and educational requirements, contributes to the variation in their effectiveness at reducing child labour.

Eligibility for the RPS program was determined by ranking rural districts by a ‘marginality index’, based on a weighted average of 4 poverty measures for these districts (average family size, percentage without piped water, percentage without a latrine, and percentage of individuals over 5 who are illiterate). Areas with higher index scores were deemed most impoverished. The 42 districts selected for inclusion in the evaluation study had the highest marginality index scores.

The RPS was implemented in 2000 as a cluster-randomized experiment that included 42 eligible districts in rural Nicaragua. Twenty-one districts were randomly assigned to each of the treatment and control settings. A baseline wave of data on individuals, households, and districts was collected in 2000, before the implementation of the CCT program, and two follow-up waves in 2001 and 2002. The baseline sample was obtained by stratified random sampling at the district level. Forty-two households were randomly selected from each of the 42 eligible districts, comprising a total of 1,764 households. Due to attrition and nonresponse, the final wave of data in 2002 included only 1,397 households, of which 722 were in intervention villages and 675 in control villages.

• Child work (whether CCT prevents child work or not)
• The number of hours worked by a child

RCT

Study design: Because of the joint dichotomous and continuous distributional properties of the outcome variable, the author uses Tobit regression to address the research questions.
Sample: There is a total of 2,028 children aged 9-15 (these are the children who were aged 7-13 at the baseline in 2000, and therefore comprise the applicable age group for RPS). Of these, 1,049 are from control villages and 979 from treatment villages. In residual, the author includes the random effects of
Survey Instrument: A baseline wave of data on individuals, households, and districts was collected in 2000, before the implementation of the CCT program, and two follow-up waves in 2001 and 2002. No further detail on the type of information collected is provided.

The RPS subsidy caused the estimated probability that a child will engage in work activities to be reduced by approximately 10.6%. The offer of an RPS subsidy reduces the hours that a child will engage in work activities, given that the child is currently working, by approximately 3.65 hours.

Child Labour

Households

Local

The RPS subsidy caused the estimated probability that a child will engage in work activities to be reduced by approximately 10.6%. The offer of an RPS subsidy reduces the hours that a child will engage in work activities, given that the child is currently working, by approximately 3.65 hours.

-

There are two primary reasons why Nicaragua’s program provides a good setting. First, as mentioned above, the randomized offer of a CCT subsidy provides a rigorous design that allows the estimation of the causal impact of such a subsidy offer on both the incidence and duration of child labour. Nicaragua’s RPS program is one of only three CCT schemes in Latin America that have utilized such random assignment, and it has also made the data collected under this assignment publicly available for further analysis.