Harvesting differences-in-differences and event-study evidence

This paper by Blueprint Director Joshua Angrist, Alberto Abadie, Brigham Frandsen, and Jörn-Steffen Pischke surveys econometric innovations related to differences-in-differences estimators and to event-study models with time-varying treatment effects. The researchers discussion highlights tricky normalization issues, heterogeneous policy effects, the interpretation of exposure designs, pretrends pretesting, and the ever-bothersome question of logs versus levels. Key ideas are illustrated with applications.