TrainingsDE 2ème/3ème cycleDESU Magistère Ingénieur EconomisteCoursesPublic policies evaluation

DESU Magistère Ingénieur EconomisteUE Public policies evaluation


This lecture aims to provide the different tools for formal evaluations of « treatment effects », that is, the causal effect of a public policy or program on economic decisions (labor supply, consumption, savings, retirement, etc.) in developed or developing countries.

Course overview :

1. The evaluation problem and Rubin’s framework

2. Randomized Control Experiments

3. Selection bias

4. Instrumental Variables and LATE

5. Difference-in-difference and panel estimations

6. Regression discontinuity

7. CIA, Matching and PSM

8. The revolution of administrative data

Professional skills

To have a good command on evaluation tools and notably the conditions of success for experiments (randomization protocol and verification, sample size and power of a test, attrition and other issues) and natural experiments (condition of « common trend » for difference-in-difference, density tests for regression discontinuity, etc.) as well as for classic econometric methods and their re-interpretation (IV and LATE, panel FE, linear estimations versus matching and PSM).

Language used

Main language used by this course: Anglais.


  • Imbens, G., and Rubin, D., (2012) .Causal Inference in Statistics and Social Science

Recommended prerequisites

Basic Econometrics.

Structure and organisation

6 lectures of 4 hours ; final exam of 2 hours with questions and MCQ