Assessing political bias and value misalignment in generative artificial intelligence
Measures political preferences and value alignment in generative AI systems.
Victor Rangel
Economist working on causal inference, public policy, political economy, and large language models.
I have a Bachelor's degree in Economics and I'm pursuing a Master's in Public Policy at Insper. I have professional and academic experience in data analysis, focusing on applied econometrics and machine learning.
I spend most of my days running acceptably biased OLS models and, additionally, I work as a science communicator and teach econometrics on my Instagram profile @econometriafacil (200,000+ followers). I also write the Econometria Fácil column at Folha de São Paulo.
Research Interests
Applied Microeconometrics
Causal Inference
Large Language Models
Public Policy Evaluation
Political Science & Political Economy
Selected Research
Economic conditions and voting behavior
Studies how local economic conditions shape electoral behavior.
Natural disasters and financial technology adoption
Examines whether natural disasters accelerate adoption of digital financial services.
Do Health Policy Conclusions Survive Modern DiD? Re-estimating TWFE Studies
Revisits health-policy evidence using modern difference-in-differences estimators.
Teaching & Media
Econometria Fácil
Educational platform with 200,000+ followers, focused on econometrics, causal inference, and data analysis.
InstagramFolha de São Paulo
Columnist for Econometria Fácil, writing about evidence, policy, data, and public debate.
ColumnCourses & Workshops
Hands-on teaching in econometrics, R, Python, Stata, causal inference, and applied research workflows.
Recent Updates
- 2026 New column Econometria Fácil at Folha de São Paulo
- 2026 "Masking the identification problem" received R&R at Journal of Comments and Replications in Economics
- 2025 "From Association to Explanation" accepted at Advances in Scientific and Applied Accounting (ASAA)
- 2025 "Economic conditions and voting behavior" forthcoming at Public Choice
- 2025 "Assessing political bias and value misalignment in generative AI" published in Journal of Economic Behavior and Organization