The impact of Ethiopia’s productive safety net program on schooling and child labour (Chapter 5 of Social Protection for Africa’s Children)

Northeast Africa

Ethiopia

Labour Market

Public Works

J Hoddinott, DO Gilligan, AS Taffesse

2010

Hoddinot et al (2010). The impact of Ethiopia’s productive safety net program on schooling and child labor. In: Devereux, Stephen, et al. Social Protection for Africa’s Children (Chapter 5). London: Routledge, 2011.
View publication

Secondary Data: Food Security Program Survey. Available at: https://www.fao.org/food-agriculture-microdata/en/.

2010

Ethiopia’s Productive Safety Net Program (PSNP)

June to August 2006 (data collected by the government)

The objective of the program is to provide food or cash transfers to households through large PW projects or a smaller program of direct transfers to households with elderly or disabled members. The objective of the study is to examine the impact of the program on child outcomes.

The eligibility for the PSNP is restricted to poor households (no further details are provided).

The PSNP operates in 262 “chronically food-insecure” woredas (districts), defined by the government as those woredas that had been significant recipients of food aid between 2002 and 2004. The PW program under the PSNP pays individuals from selected beneficiary households 6 birrs ($US 0.61) per day or food of equivalent value (at 2005 prices) to work on labor-intensive projects designed to build community assets. The most common projects included soil conservation activities and road and school construction. Each beneficiary household was allocated a labor quota of up to 30 days of work for each household member per year, for a maximum transfer of 180 birrs per member per year. Because only working-age, able-bodied household members could perform the work, the number of days worked by PW participants often exceeded 30 days per year.

• School attendance
• Child labour (hours worked in household chores, agriculture, others).

Propensity Score Matching

Study design: to estimate the effect of the PSNP and OFSP on schooling and child labor, it is necessary to construct an empirical “identification strategy” – a method for isolating causal effects that are a direct result of participating in the programs. Developing such measures requires comparing outcomes for beneficiary households to the counterfactual; that is, what those outcomes would have been in the absence of the program. The fundamental challenge in this and all evaluation

studies is that the counterfactual cannot be directly observed. As a result, the authors constructed a proxy for these counterfactual outcomes from data on a selected comparison group of non-beneficiary households using propensity score matching (“nearest neighbors”).
Sample: The Food Security Program (FSP) survey sample was representative of woredas served by the PSNP in the four principal regions. Within each region, the number of food-insecure woredas selected for the sample was proportional to the number of chronically food-insecure woredas in that region relative to the number of chronically food-insecure woredas in all four regions. Within regional strata, woredas were selected with probability proportional to size (PPS) based on the estimated chronically food-insecure population (beneficiaries). The woredas were ordered geographically, to ensure geographic dispersion of the sample and to cover the range of agroecological conditions in each region. In total, 68 out of 190 woredas were selected across the four regions. Within each woreda, two to three kebeles were sampled proportional to size from a list of all kebeles implementing the PSNP, yielding 148 kebeles in the sample. Within each kebele, one to three villages or village groupings were selected as enumeration areas for household sampling. From separate lists of PSNP beneficiary and non-beneficiary households in each kebele, 15 PSNP beneficiary households and 10 non-beneficiary households were selected for the sample, resulting in a target sample of 3700 households.
Survey Instrument: The survey collected data on participation in the PSNP covered a wide range of outcomes, including detailed modules on child school participation and child time allocation.

Public works (PW) participation under the PSNP has no beneficial effects on school attendance. There were no improvements in average attendance rates for boys and attendance rates decreased for girls aged 6 to 10. However, there were significant reductions in child labor hours for PW participant households on average, leading to reduced child labor time in agriculture for boys and lower total labor hours for girls aged 11 to 16 years.

Child Labour

Households

Local

There were significant reductions in child labour hours for Public Works participant households on average, leading to reduced child labour time in agriculture for boys and lower total labour hours for girls aged 11 to 16 years.

Public works (PW) participation under the PSNP has no beneficial effects on school attendance. There were no improvements in average attendance rates for boys and attendance rates decreased for girls aged 6 to 10.

One shortcoming of the FSP survey for studying the impacts of the program is that this was a cross-sectional survey conducted 18 months after the PSNP began. No pre-program data were available, and the intervention period was relatively short given some of the delays in implementation that occurred during the process of scaling up the program. Also, the PSNP is complemented by a series of other food security activities, collectively referred to as the Other Food Security Program (OFSP). This includes access to credit, agricultural extension, technology transfer (such as advice on food crop production, cash cropping, livestock production, and soil and water conservation), and irrigation and water harvesting schemes.