Using difference-in-differences to identify causal effects of COVID-19 policies

Andrew Goodman-Bacon and Jan Marcus

Policymakers have implemented a wide range of non-pharmaceutical interventions to ﬁght the spread of COVID-19. Variation in policies across jurisdictions and over time strongly suggests a difference-in-differences (DD) research design to estimate causal effects of counter-COVID measures. We discuss threats to the validity of these DD designs and make recommendations about how researchers can avoid bias, interpret results accurately, and provide sound guidance to policymakers seeking to protect public health and facilitate an eventual economic recovery.

MoreLess

Year of publication:	2020
Authors:	Goodman-Bacon, Andrew ; Marcus, Jan
Publisher:	Berlin : DIW Berlin, German Institute for Economic Research
Subject:	difference-in-differences \| non-pharmaceutical interventions \| COVID-19 \| causal inference \| Coronavirus \| Wirkungsanalyse \| Impact assessment \| Kausalanalyse \| Causality analysis \| Induktive Statistik \| Statistical inference \| Gesundheitspolitik \| Health policy \| Lockdown \| Lock-down