The unequal impact of matching students with schools – MATCHINEQ

In all projects, we use the state-of-the-art in structural modeling, combining theory with rich datasets on choices, outcomes, and characteristics.
Administrative data from Chile allow us to see which schools parents choose and how they rank them against each other. We also see characteristics of such schools, including the average teacher experience and class size. We combine this data with new estimators we develop for education production functions and choice models to be able to assess how effective school characteristics are and how matching systems can be improved.
In France, we focus on the dynamic decision process students have. Similarly, we observe how students rank college programs against each other, but we also model what makes them decide to wait for a better offer in a multi-round matching mechanism. This then allows us to assess the costs and benefits of such a system.
Finally, the Canadian data allow us to see the college programs students applied to, but also the high school history of classes and grades, which allow us to construct peer groups. Such data allow us to investigate how peers impact application decisions.

Project 1 (Chile) focuses on school choice and its implications for educational quality and inequality.
In a first paper, the authors develop a new education production function and use it to show that class size significantly reduces test scores on average, but effect sizes differ a lot by schools. They also find important effects of financial resources but not so much for teacher experience.
In a second paper, the authors show that a reform to improve access to school for disadvantaged students had limited impact on segregation and access to quality for students of low socio-economics status—mainly because families submitted too few school choices. Simulations show the reform’s potential for improvement if students had ranked more schools and quotas for disadvantaged students were expanded.
Project 2 (France) explores centralized college admissions.
College admissions in France take place in different rounds. While students dislike waiting, it also helps to match them to a program they like better. We find that the net gains are substantial. While they are lower for socio-economically disadvantaged students, we do find that they also obtain large gains in welfare and educational attainment.
Project 3 (Canada) examines peer effects in college applications and how they relate to differences in application behavior between students of high and low socio-economic status.
This research shows that students’ application choices are strongly influenced by peers, in their school or neighborhood. Socio-economically advantaged students tend to have peers that apply to more selective programs, explaining partially why disadvantaged students do not apply to programs that could be beneficial for them.

This project focused on how to better match (disadvantaged) students to the educational programs that are available. More work remains to be done on the optimal provision of programs and choices.
Personal choices are not necessarily in line with society’s objectives because of externalities. Apart from financial externalities that are inevitable in subsidized educational systems, peer effects are crucial in this context as the student composition of an educational program affects its quality. Furthermore, from an individual perspective, the right choice is difficult to make, and behavioral biases (or “internalities”) are common.
As a result, such choices are often restricted. Targeted vouchers open spots for some groups of children at the cost of others, college admissions prevent completely free access. However, there is little guidance for policymakers in knowing and realizing the optimal size and composition of different options. What is the right balance between choice, ability and diversity requirements in school choice? Are seats in college programs optimally provided according to society’s needs? How to incentivize students to align their choices with a societal optimum?
These questions gave rise to the PIs ERC starting grant that now replaces this ANR grant. In this project, he and his team aim to develop a methodology to empirically analyze the optimal provision of choice using novel dynamic discrete choice models that take into account externalities and internalities. The objective is to apply this methodology to similar contexts and data to be able to bring this type of research to a higher level with a large impact on science and policy.

3 TSE working papers:
De Groote, O. and Gazmuri A., “School choice and class-size effects: Unintended consequences of a targeted voucher program”, TSE working paper, n. 1526, April 2024.
De Groote, O. and Rho, M., “Matching disadvantaged children to daycare: Evidence from a centralized platform”, TSE working paper, n. 1600, December 2024.
Fabre, A., “Robustness of Two-Way Fixed Effects Estimators to Heterogeneous Treatment Effects”, TSE Working Paper, June 2023.
4 unpublished manuscripts:
Fabre, A., “The Geography of Higher Education and Spatial Inequalities”, December 2023.
De Groote, O., Fabre, A., Luflade, M., Maurel, A., “Sequential College Admission Mechanisms and Off-Platform Options”, June 2025.
Gazmuri, A., Prager, E., “Peer Learning in College Applications”, December 2024.
Botbol, L., Gazmuri, A., “Effects of centralizing school admissions on education equality”.

2 works in progress:
Pal, J. “Optimal Policy Design for Teacher Recruitment”
Foucault, G. and Pal, J. “Effort Choices and Funding Instruments in Higher Education”

The participants in these projects also presented this work at dozens of seminars and conferences, detailed in the acknowledgements of the papers. These include department seminars at top economics departments such as Duke University and the University of Pennsylvania, as well as selective conferences and the RESTud tour for the best job market candidates from Europe.
While some papers were directly targeted by the ANR project, the experience with the data, context and topic also further inspired the researchers involved in these projects leading to a large output of scientific research.

School choice programs allow students (and their parents) to select a school of their preference in an attempt to improve access to high-quality education. As school capacities are limited, it is not possible to assign every student to their most preferred choice. Therefore, it is important to define a mechanism for student assignments. There is little evidence on the impact that these mechanisms have on equal access to quality education. The way in which they are implemented could hurt socially disadvantaged groups for several reasons. For example, it could reduce the options available for disadvantaged students if schools are able to select students. Moreover, misinformation about the rules, or limited resources to apply and hold out for the best options, could disproportionally harm students from low socioeconomic backgrounds. These are featured frequently across many different school choice settings without a clear understanding of the consequences in terms of educational outcomes and inequality. Our team proposes a project in which we will analyze education programs that aim to improve outcomes among the socially disadvantaged. We will study how the characteristics of the allocation system impact educational outcomes for different socioeconomic groups both in compulsory and higher education. We will use and extend structural econometric models to study student behavior, which will allow us to study some of the determinants of social inequality and to simulate outcomes under alternative matching policies.