← Search

American Economic Review Vol. 111 No. 12 2021

Synthetic Difference-in-Differences

Dmitry Arkhangelsky1; Susan Athey2; David A. Hirshberg3; Guido W. Imbens4; Stefan Wager5

1 CEMFI, Madrid (email: ) · 2 Graduate School of Business, Stanford University, SIEPR, and NBER (email: ) · 3 Department of Quantitative Theory and Methods, Emory University (email: ) · 4 Graduate School of Business and Department of Economics, Stanford University, SIEPR, and NBER (email: ) · 5 Graduate School of Business, and of Statistics (by courtesy), Stanford University (email: )

open access

Abstract

We present a new estimator for causal effects with panel data that builds on insights behind the widely used difference-in-differences and synthetic control methods. Relative to these methods we find, both theoretically and empirically, that this “synthetic difference-in-differences” estimator has desirable robustness properties, and that it performs well in settings where the conventional estimators are commonly used in practice. We study the asymptotic behavior of the estimator when the systematic part of the outcome model includes latent unit factors interacted with latent time factors, and we present conditions for consistency and asymptotic normality.

DOI
10.1257/aer.20190159
Volume
111
Issue
12
Pages
4088-4118
Language
en
Sources
bibtex:phds-export.bib openalex crossref

Cite