POLICY EVALUATION

This course provides an introduction to modern methods for policy evaluation and causal inference in applied microeconomics and the social sciences. The course focuses on empirical strategies used to identify the causal effects of public policies, programs, and institutional interventions using both experimental and observational data.

Students attending this course are expected to be proficient with the fundamentals of probability and mathematical statistics, as well as with the tools of causal reasoning and identification.

Building on the contemporary toolbox of applied econometrics, students will learn how researchers move from correlation to credible causal interpretation in real-world settings.

The course combines theoretical foundations with an applied and design-oriented perspective. Particular attention is devoted to understanding the assumptions underlying each empirical strategy, assessing the credibility of identification designs, and interpreting empirical results critically. The course also emphasizes replication, reproducibility, and the practical implementation of methods using statistical software. Topics include randomized experiments, matching methods, difference-in-differences designs, instrumental variables, synthetic control methods, and regression discontinuity designs. Throughout the course, students will engage with empirical applications in labor economics, public policy, political economy, development economics, education, and related fields.

Angrist, Joshua and Jorn-Steffen Pischke (2008). Mostly Harmless Econometrics: An Empiricists’s Companion, Princeton University Press. (MHE) Cunningham, Scott (2020). Causal Inference: The Mixtape, Yale University Press. Cameron, Colin and Pravin Trivedi (2009). Microeconometrics Using Stata, Stata Press. (MUS)

Lecture slides and additional readings will be made available through the course platform. Since no single textbook fully covers the material presented in the course, the readings will combine textbook chapters, methodological papers, and empirical applications discussed during lectures.

Compliant students: Written Exam (2/3 or about 66,7%) In-class written exam covering conceptual questions, interpretation of empirical results, and applied exercises on causal inference methods. Group Presentation (1/3 or about 33.3%) Students present and critically discuss an empirical paper using one of the methodologies covered in class. Students are required to replicate and extend part of the empirical paper using real data and statistical software. The project may be completed individually or in small groups. Presentations should focus on: – research question; – identification strategy; – validity of assumptions; – interpretation of results; – potential limitations and extensions. Presentations will take place throughout the course and will cover papers related to the topics discussed each week. The presentation schedule and paper assignments will be communicated by the instructor at the beginning of the semester. Presentation materials (slides, code, and an approximately 8-page report) must be submitted via the course platform at least 48 hours before the presentation. Each presentation will last 15–20 minutes and will be followed by a discussion. Non Compliant students: Written Exam (100%) In-class written exam covering conceptual questions, interpretation of empirical results, and applied exercises on causal inference methods. The exam will be longer and will consist of two additional questions designed to assess a more advanced level of understanding. Retake exams follow the same rules as those for non-compliant students.

Students who pass the exam with a grade of at least 25/30 may request a thesis under this professor's supervision.