You get what you pay for: Schooling incentives and child labour

South Asia

Nepal

Education

Vouchers / Scholarships / School Subsidies

EV Edmonds, M Shrestha

2014

Edmonds, Eric V. and Maheshwor Shrestha. 2014. “You get what you pay for: Schooling Incentives and Child Labor”. Journal of Development Economics.
View publication

-

2014

Nepal Goodweave Foundation (NGF)

April 2010– March 2011

“encouraging school enrollment and attendance and reducing child labor in the carpet sector”

“children whose guardians worked in establishments that were NGF licensees or sub-contractors of licensees […] that (a) had attended school within the last 18 months, (b) had not received education support from NGF or other sponsors and (c) were in families who faced an elevated risk of transitioning to child labor”.

The effects of two interventions were tested. The first intervention compensated households for education related expenses. The second intervention also compensated households for education related expenses and in addition provided a stipend for regular school attendance. The value of the stipend was equal to about one third of the income earned by youths active in weaving.

• School participation: various measures of attendance and participation in exams
• Performance in school: failed grade and exam scores
• Child labor: carpet weaving

Randomized controlled trial

Study design: identified eligible children were allocated to a control group and the two intervention groups by means of a lottery
Estimation strategy: the effect of the interventions is identified by means of regressions that control for age, gender, age and gender interaction, outcome variables observed at baseline, and fixed effects for strata used in the randomization
Data: two sources of administrative data were collected: data collected to identify eligible children and classroom attendance records. In addition, four waves of household surveys were collected: baseline (before the 2010/2011 survey), midterm (during school year), yearend (end of school year), follow-up (16 months after end of school year)

The effects of the first intervention, which compensated households for education related expenses, were generally not statistically significant. The second intervention, which also provided a stipend for regular school attendance, had beneficial effects on school participation and performance in school during the period of the intervention: school attendance increased by about 5 percentage points (92% in the control group), participation in exams increased by about 9 percentage points (87% in the control group), and reduced grade failure conditional on taking the final exams by 6 percentage points (14% in control group). The second intervention also reduced children’s participation in weaving during the period of the intervention by about 4 percentage points (7% in the control group). The beneficial effects of the second intervention are not visible in the follow-up survey, 16 months after the second intervention had ended.

Child Labour

Households

Local

Results are reported for three different points in time: during the program, shortly after the program ended, and 17 months after the program ended. The stipend leads to decline in child involvement in weaving a few weeks past the end of the period of support.

The impact of schooling support is limited to the period of support. Scholarship funds were largely exhausted at the start of the school year, and the scholarship treatment seems to promote school attendance only at the first part of the school year. The stipend treatment has a large impact on attendance that seems to last even into the weeks past the end of the program. There is no evidence of effects of either intervention on schooling or work 17 months after the end of the program.

The authors also examine the effects of the intervention separately for boys and girls. Generally, they find that the effects are stronger for girls, although differences between boys and girls are not necessarily statistically significant. The authors also examine intra-household spillover effects, i.e. indirect effects on fellow household members of children participating in the intervention, we refer the readers to the paper for further discussion.