Context

Ethiopia’s PSNP is one of the largest social protection programmes in Sub-Saharan Africa, providing participating communities with a mix of public works employment and unconditional transfers. The programme’s objective is social protection for food-insecure communities, so as to prevent asset depletion at the household level, to create community assets and to stimulate markets. The household-level impacts of public works have been widely studied, but much less is known about their effects on intra-household resource allocation and the consequences of these decisions for children’s human capital formation. 

Since 2006, IFPRI has been working with the Ethiopian government and its PNSP development partners on the design and implementation of a large-scale, longitudinal household survey of beneficiaries and non-beneficiaries. Development partners include a consortium of donors, such as the World Bank, UK aid, USAID, the Canadian International Development Agency and Irish Aid, which are represented on the Food Security Coordination Directorate that implements the PSNP. 

Although the objectives of the PSNP explicitly include contributions to reduced malnutrition of children under 5 years, nutrition outcomes did not feature in the regular evaluations. There was also concern that public works could contribute to increased child labour. The 3ie-supported evaluation – led by researchers from IFPRI, Ethiopian Development Research Institute and Cornell University – responded to these issues and came at a time when the donors were looking to work on nutrition and the government was reviewing its national nutrition strategy. 

The PSNP was set to undergo a major government and donor review in 2013 and 2014 to assess whether it should change, continue or end. The 3ie-supported work that began in 2012 was expected to feed into this process, showing whether the programme had had an impact on human capital formation. 

The evaluation examined the PSNP’s impacts on the health status of preschool children, children's schooling, and the incidence and level of child labour. The evaluation also looked into whether these impacts varied by age and sex of the child and whether they varied by sex of the household head.

Evidence

In 2008, when PSNP payments were low, participation in the programme lowered the completion of grades for both girls and boys. Child labour on the family farm also increased, although for boys this was offset by reductions in domestic labour. However, as PSNP payments increased, these adverse outcomes reversed. In 2012, the programme increased girls’ grade attainment between 6 and 14 per cent (depending on the age of the child), improved schooling efficiency by 10 to 20 per cent and reduced labour for boys. 

On nutrition, the evaluation found the PSNP had not reduced stunting or wasting amongst children from participating communities. The evaluation highlighted issues of dietary quality and the fact that the programme had not encouraged consumption of pulses, oils, fruits, vegetables, dairy products or animal-sourced proteins. Most mothers had not had contact with health extension workers, nor had they received information on good feeding practices. They also did not regularly boil drinking water.

Evidence impacts

Suggested citation

International Initiative for Impact Evaluation (3ie), 2019. Using evidence to link nutrition to social protection in Ethiopia (online summary), Evidence Impact Summaries. New Delhi:3ie.