Currículo

Policy Evaluation PEV-LECO

Contextos

ECTS

6.0 (para cálculo da média)

Objectivos

Policy Evaluation introduces students to the foundations of public policy design. Students will learn methods that allow establishing causal inference between exposure to programs and public policies with their outcomes, as well as the importance of behavioural science in the promotion and design of public policies in different areas such as health, education, social security, taxation, and the environment. This discipline has two main objectives: 1) a broad understanding of various social and economic issues that policymakers face, and 2) obtaining essential tools for future public policy researchers, as well as for policymakers, program managers, consultants, and advisors so that public policies are designed, implemented, and evaluated efficiently and effectively. This course is designed to give students the opportunity to choose, according to their own academic and professional interests, the topics they want to delve into. For each topic/lecture, a more specialized complementary bibliography will be indicated. This bibliography can be read individually or in groups through the creation of "reading and discussion groups" that enhance understanding and discussion of the topics. This discipline uses recent empirical examples from studies published in the most prestigious international scientific journals to motivate the use of (quasi-) experimental frontier scientific methods in the evaluation of public policies. These real-world policy applications will be presented in a non-technical and analytical manner. In this context, the course will provide an introduction to basic methods in data science, including regression, causal inference, and machine learning. Without neglecting economic theory and in-depth knowledge of institutional details, the goal is to estimate the causal impact of events and choices on a specific outcome of interest. The increase in computing power, the advent of big data, and new statistical and econometric techniques have given a new impetus to our better understanding of the effects of public policies. This discipline will take into account the importance of behavioural science in the design, implementation, and evaluation of public policies. We will also familiarize ourselves with concepts such as nudges. Finally, we will address recent methods for calculating the cost-benefit of different policies. The course will use Stata software to showcase the main methods, but R and Phyton are also accepted in the evaluation components.

Programa

- Topic 1 (Classes 1 and 2): Part I: Why are Big Data and administrative micro-data transforming the evaluation of public policies? The importance of doing evaluation. The differences between monitoring and conducting proper impact evaluation. Introduction to the Experimental Methodology and to Randomized Controlled Trials (RCTs). Part II: Basic concepts in experimental design and procedures to consider in the implementation. Using RCTs to study perceptions, knowledge and beliefs, attitudes, and reasoning. Problems and limitations of this method. - Topic 2 (Classes 3 and 4): What to do when randomization is not possible or ethically acceptable? Difference-in-differences (DiD): Introduction, Main assumptions, Staggered interventions. Running examples: The importance of geographical context (Place-based policies); the child penalty. - Topic 3 (Class 5): Introduction to Regression Discontinuity Designs. Main assumptions. Running examples: Impact of attending elite universities; public policies and political support. - Topic 4 (Class 6): Saliency, enforcement, incidence, and asymmetry issues when designing public policy. - Topic 5 (Class 7): Introduction to Synthetic Control Methods and Synthetic DiD. Running examples: How can taxes and other public policies be designed and implemented to discourage socially harmful behaviours and negative externalities such as pollution? Pigouvian taxes and sin taxes. - Topic 6 (Class 8): Part I: Bunching. Running example: Taxes on labor. Part II: Behavioral responses to taxation: evasion and migration. Taxes on capital. - Topic 7 (Classes 9 and 10): Introduction to machine learning. Running example: Can computer assistance improve judicial decisions? Will these decisions be fairer, or do they raise technical and ethical issues? - Topic 8 (Classes 11 and 12): Essay presentations (research design)

Método de Avaliação

The teaching approach will strive to integrate, whenever possible, the exposition of concepts and theories with methodological analysis based on real cases of implementation and evaluation of public policies. Lecture slides will be made available on the course page on FÉNIX. Attendance and active participation in the majority of classes are expected from students. To encourage participation in the assessment, the evaluation during the regular period will consist of: • In-class presentation of a recently published research paper from a scientific journal, followed by discussion, worth 20% of the final grade. • In-class presentation of a research project, followed by discussion, in the last week of the term: 30% of the final grade. This should include: (i) a relevant research question related to the evaluation of a real public policy, explaining why this policy is important and why it should be studied (including a brief review of the literature); (ii) an evaluation design, with an explanation of the data and method(s) (iii) (preliminary) results (iv) potential limitations of the proposed approach. • Written final exam: 50% of the grade.

Carga Horária

Bibliografia

Não foi definida bibliografia principal

Disciplinas de Execução

2025/2026 - 2 Semestre