Working Papers

The Heterogeneous Price of a Vote: Evidence from France, 1993-2014 (2019) With Julia CagéCEPR Discussion Paper, INET Working Paper. 

Work in Progress

Primary School Quality in Sub-Saharan Africa: A comparative study of primary
school systems in Ghana and Côte d’Ivoire

Why do primary school systems often fail to provide students with the most basic skills in developing countries, despite consequent educational investments? To answer this question, this paper exploits the colonial legacy of Sub-Saharan Africa. Relying on a comparative study of Ghana and Côte d’Ivoire, we are the first to explore geographical and temporal variations to estimate the returns to education in terms of literacy and numeracy.  We suggest a new methodology to account for differences in selection into and within primary school, and to quantify educational quality. Surprisingly, we show that the more inclusive educational policies implemented in Ghana did not lead, on average, to higher literacy rates than the more quality-oriented policies adopted in Côte d’Ivoire.  However, the Ivorian system does much better at increasing literacy in the very first years of the primary cycle. Hence, our study challenges the prominent focus on school enrollment for early education.

Colonial Origins of Education Systems and Student Performance in Primary School. Evidence from CameroonWith Yannick Dupraz.

How much do teaching practices impact learning achievements? To answer this question, we exploit the natural experiment provided by the partition of Cameroon between British and French empires after WWI. Since reunification at independence, two educational systems, Francophone and Anglophone, coexist. To our knowledge, we are the first to investigate the impact of teaching practices on learning achievements where the existence of two different subsystems can be thought of, at least locally near the border, as random. Our setting and data allow dealing with multiple treatment issues and identifying the determinants of the education production function responsible for differences in student achievement. Our border discontinuity analysis of 2004-2005 PASEC primary school survey data finds that Anglophone students might perform better in math at the end of the second year, but that Francophone students clearly perform better in the fifth year by half a standard deviation. We provide evidence that system-specific teaching practices play an important role in the accumulation of human capital.

The impact of rain shocks on students’ performance in Benin. With Kenneth Houngbedji and Oswald Koussihouede.

In this paper, we study the relationship between heavy rain during school years and students achievements. Our dataset allows us to measure progression between the beginning and the end of the school year. We argue that rainfall disrupts learning experience in two ways. First, rainfall during school days depresses school attendance for teacher and students, especially when the roads leading to school are degraded and become unusable. Second, depending on the type roof and ceilings of classroom, rain noise accompanying heavy rainfall disturbs the acoustic conditions of the classroom and the learning process. In the context of Benin, where weather varies widely across geographic areas, educational policies that accommodate school calendar to the region would be both cheap and highly efficient to increase student performances.

Ongoing Projects & Field Work

Qualitative Field for Survey Preparation: Benin, Côte d’Ivoire, Senegal.

Survey design, fund raising, organization and supervision of the data collection
through video recording, interviews and class observation.

Coordination of a classroom observation survey, World Bank: Cambodia.

Survey design, organization and supervision of the data collection through video
recording, interviews, teacher testing and class observation

Create a website or blog at

Up ↑