Publications

We study the effect of opening a new high school on individual schooling decisions at the end of lower secondary education. The working sample covers all ninth graders between 2007–2008 and 2012–2013 in France. The two-way fixed-effect estimation strategy uses variation in time and space to estimate the causal effect of an increase in school supply. Opening a new high school significantly increases the probability of pupils from neighboring middle schools continuing in higher secondary education. The effect is exclusively due to new high schools proposing a vocational track. Furthermore, the effect is mainly driven by low-achieving students.

This paper analyzes the relationship between students grades and their evaluations of teaching. We exploit an original data set from almost 100 courses during 7 academic years in a French higher education institution. We use teacher fixed effects to rule out any simultaneity or endogeneity bias. We find that students take their exam grade into account when they evaluate teaching. A better grade is associated with a better evaluation of a teacher's pedagogy, although the size of the effect is relatively small. A one-point increase in by-course mean grade corresponds to a less than one percentage point decrease in the proportion of students giving bad evaluations. These results suggest that it is possible to manipulate evaluations through grade or exam leniency.

By analyzing a French program that targeted low-achieving and socially disadvantaged junior high schools we provide evidence that school-based compensatory education policies create sorting effects. We use geocoded original data and a regression discontinuity framework to show that the program decreases the individual probability of attending a treated school and symmetrically increases the probability of attending a private school. The effects are driven by pupils from high socioeconomic backgrounds, resulting in an increase in social segregation across schools.

This study analyzes the effect of opening a new high school on pupils’ allocation to different tracks at the end of lower secondary education. We use high school openings to highlight the constraint local school supply exerts on individual schooling decisions. The working sample covers all pupils enrolled in 9th grade between the school year 2007-2008 and the school year 2010-2011 in France. The estimation strategy (a generalized difference in differences) takes advantage of the variation in time and space of the openings of high schools to estimate the causal effect of an increase in school supply on the allocation of pupils to tracks at the end of 9th grade. We show that opening a new high school increases the probability to continue in high school for pupils from the nearest middle school. This increase is driven by the vocational track, and by low achieving students.

Apprenticeship is unequally distributed over the French territory, with strong regional disparities. The objective of this paper is to study the relationship between regional expenditures on professional training, and access to apprenticeship in each region. To do that, we constructed an original panel data set on regional expenditures in metropolitan France. We also use individual-level data (Céreq Génération 2010) to study the individual decision to complete an apprenticeship, once regional characteristics are taken into account.

“Réseaux ambition réussite” (RAR), or “ambition success networks,” are a compensatory education policy that took place in France between 2006 and 2011. To be eligible, a collège (equivalent to junior high school) had to enroll at least 67% of pupils from a disadvantaged background, and at least 10% of pupils having repeated twice or more when entering junior high school. About 3% of junior high school pupils were enrolled in a rar. Evaluating the impact of such a policy is complicated by two types of bias. First, the selection of schools into the program makes that the pupils enrolled in a rar school would have lower academic achievement than the other pupils on average, even in the absence of the program. Second, the school is not the good level of analysis, since pupils may select themselves into the program. To analyze the causal effect of the program on individual academic achievement at the end of junior high school, we use a regression discontinuity method and an original exhaustive data set. Overall, we do not find any significant effect of enrollment in a RAR on individual educational outcomes at the end of junior high school. The absence of average effect may be due to differentiated effects according to pupils’ characteristics.