School grants and student performance in Senegal: an experimental analysis

Written on 16 Oct 20 by Pedro Carneiro, Oswald Koussihouèdé, Nathalie Lahire, Costas Meghir , Corina Mommaerts
Microplanning of education
Educational quality


In the last 50 years, primary school enrolment has increased dramatically in developing countries.  Even in the poorest areas of Sub‐Saharan Africa, gross enrolment rates in primary school are close to 80% (Glewwe and Kremer, 2006). However, the quality of education in these countries remains low. Access to education does not translate into corresponding increases in productivity and well-being. This is consistent with recent evidence suggesting that education quality, not quantity, matters most for growth (Hanushek and Woessmann 2011; Glewwe et al. 2013).

Is it possible to improve the quality of learning in a disadvantaged context through school grants?

The assumption is that local decision‐makers, such as principals and community leaders, are likely to have a deeper understanding of the needs of their schools than central education authorities, and are thus best placed to put these resources to their most efficient use.

Mechanism for allocating school grants

For many years, Senegal allocated grants (for school projects) in the belief that schools were best placed to identify their own needs and come up with practical solutions.

In 2009, the Senegalese Ministry of Education launched a call for proposals to apply for school grants funded by the World Bank. To rigorously evaluate the programme the government also sought technical and financial assistance from the World Bank. As school grant schemes had existed previously, many schools were undoubtedly aware of their existence.

However, to ensure that all schools benefited from the same level of information about the scheme, field agents were sent to each regional authority to ensure they had indeed received the call for proposals and understood their role. During the call for projects, the emphasis was placed on pedagogical concerns, contrary to previous calls that focused on improving the physical school environment. Each school was allocated a grant of 1,500,000 FCFA.  

Following the call for proposals, 712 schools submitted a grant application; during a workshop, the winning proposals were selected on the basis of their objectives: (a) to improve French, mathematics and science, (b) to reduce grade repetition and drop-out rates, and (c) to increase effective learning hours.
Acceptable proposals were placed into two categories. The first category comprised proposals that were immediately eligible for funding. The second comprised proposals with potential but which required revision. These were returned to the schools with comments from the evaluation committee, prior to being re-submitted. In the end, 633 projects were chosen to be funded in three randomly selected cohorts of 211 schools.

Data and identification strategy

To evaluate the effect of the school grant scheme, three surveys were conducted among grade 2 and 4 students, and their families, teachers, and principals over two academic years in 2009. The research team first conducted a baseline survey at the start (November) of the 2009 academic year, i.e. when the first round of grant payments was sent to the schools. Later surveys (follow-up surveys) were carried out among the same students at the beginning and end of the 2010-2011 school year.
The collected data focused on learning and the learning context (school and family). The schools were randomly placed in three cohorts; balance tests revealed strong levels of similarity between all the cohorts. Due to this similarity, the calculation of the impact was carried out using a simple linear model.


We found impacts of school grants on student learning, especially on girls and students with high ability levels at baseline. These effects last for at least two years. These effects are seen in grade 3 but not in grade 5, and are more pronounced in the south than the north of the country. These findings suggest that resources distributed in a decentralized manner can have positive effects on student learning.

It is not easy to explain the diversity of the results, in particular the fact that they are only visible in grade 3. One supposition is that headteachers focus on the early years as they see this as a foundation for future learning, while higher grades may already have sufficient resources compared to lower grades so that the marginal improvement in resources is less effective at higher grades. Further research would shed light on this question.

However, we have learned a little more about the north-south divide in terms of the effects of the grant scheme. While schools in the north focus on information technologies and other educational materials, schools in the south tend to focus on human resources, in particular the training of teachers and school administrators. Our results suggest that the latter type of investment, although perhaps less visible to the local community (and therefore less preferred for example by local decision-makers or even local school authorities), may be more effective than the former types of investment.


Carneiro, P., Koussihouede, O., Lahire, N., Mommaerts, C., and Meghir, C. 2020. School Grants and Education Quality: Experimental Evidence from Senegal. 2020. Economica. Volume 87, No 345, pp. 28-51.
Glewwe et Kremer, M. (2006). “Schools, teachers, and education outcomes in developing countries”. In: E. Hanushek et F. Welch (éd.), Handbook of the Economics of Education, vol. 2. Amsterdam: The Netherlands, pp. 945–1017.
Glewwe, P., Hanushek, E. A., Humpage, S. et Ravina, R. (2013). School resources and educational outcomes in developing countries: a review of the literature from 1990 to 2010”. In: P. Glewwe (éd.), Education Policy in Developing Countries. Chicago, USA: University of Chicago Press.
Hanushek, E. et Woessmann, L. (2011). “The economics of international differences in educational achievement”. In: E. A. Hanushek, S. Machin and L. Woessmann (eds), Handbook of the Economics of Education, vol. 3. Amsterdam: The Netherlands, pp. 89–200.


Bookmark this