Estimating Dynamic Treatment Effects in Event Studies With Heterogeneous Treatment Effects
Open accessThe DOI matches its Crossref record; 12 of the 12 references it cites were found in a registry or an index.
Abstract
Event studies are frequently used to estimate average treatment effects on the treated (ATT). In estimating the ATT, researchers commonly use fixed effects models that implicitly assume constant treatment effects across cohorts. We show that this is not an innocuous assumption. In fixed effect models where the sole regressor is treatment status, the OLS coefficient is a non-convex average of the heterogeneous cohort-specific ATTs. When regressors containing lags and leads of treatment are added, the OLS coefficient corresponding to a given lead or lag picks up spurious terms consisting of treatment effects from other periods. Therefore, estimates from these commonly used models are not causally interpretable. We propose alternative estimators that identify certain convex averages of the cohort-specific ATTs, hence allowing for causal interpretation even under heterogeneous treatment effects. To illustrate the empirical content of our results, we show that the fixed effects estimators and our proposed estimators differ substantially in an application to the economic consequences of hospitalization.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
Difference-in-Differences with multiple time periods
Callaway, B., & Sant’Anna, P. H. C. (2021) · Journal of Econometrics · doi:10.1016/j.jeconom.2020.12.001 · Cited by 9,126
Semiparametric Difference-in-Differences Estimators
Abadie, A. (2005) · The Review of Economic Studies · doi:10.1111/0034-6527.00321 · Cited by 2,286 · cited 5×
“We state the conditional parallel trends assumption in Online Appendix F, and present the doubly robust scores for estimating CATTe,l consistently and efficiently as an extension to the inverse propensity score reweighted estimator proposed by Abadie (2005) and Callaway and Sant’Anna (2018).”
BackgroundMethodInfluentialRevisiting Event-Study Designs: Robust and Efficient Estimation
Borusyak, K., Jaravel, X., & Spiess, J. (2024) · Review of Economic Studies · doi:10.1093/restud/rdae007 · Cited by 2,280
Estimating the Effect of Training Programs on Earnings
Ashenfelter, O. (1978) · The Review of Economics and Statistics · doi:10.2307/1924332 · Cited by 1,515 · cited 1×
“e should exclude them from the estimation to avoid violation of this assumption. While common in the applied literature, the parallel trends assumption is strong and oftentimes violated. For example, Ashenfelter (1978) documented that participants in job training programs experience a decline in earnings prior to the…”
Design-based analysis in Difference-In-Differences settings with staggered adoption
Athey, S., & Imbens, G. W. (2022) · Journal of Econometrics · doi:10.1016/j.jeconom.2020.10.012 · Cited by 917
Evaluating the Employment Impact of a Mandatory Job Search Program
Blundell, R., Dias, M. C., Meghir, C., & van Reenen, J. (2004) · Journal of the European Economic Association · doi:10.1162/1542476041423368 · Cited by 500 · cited 4×
“(1998, 1997); Blundell et al. (2004); Abadie (2005); Chernozhukov et al. (2013)), but more closely related to the recent literature on the interpretation of two-way fixed-effects regressions when the timing of treatment varies.”
BackgroundResultInfluentialRevisiting Event Study Designs
Borusyak, K., & Jaravel, X. (2016) · SSRN Electronic Journal · doi:10.2139/ssrn.2826228 · Cited by 463
Average and Quantile Effects in Nonseparable Panel ModelsPreprint
Fernández‐Val, I., Hahn, J., Chernozhukov, V., & Newey, W. K. (2013) · DSpace@MIT (Massachusetts Institute of Technology) · Cited by 236
Bailey, M. J., & Goodman-Bacon, A. (2015) · American Economic Review · doi:10.1257/aer.20120070 · Cited by 227
Average and quantile effects in nonseparable panel models
(2013) · Econometrica · doi:10.3982/ecta8405 · Cited by 170
Research works printed but not linked: 1
- H C Pedro, Sant'Anna, did: Treatment Effects with Multiple Periods and Groups 2020
Cited by
Items 101–123 of 123
Economic Impact of the Western Africa Ebola Outbreak - A Holistic ApproachPreprint
Liu, X. (2022) · SSRN Electronic Journal · doi:10.2139/ssrn.4145938 · Cited by 3
Pavlik, J. B., & Desierto, D. A. (2022) · Public Finance Review · doi:10.1177/10911421241248719 · Cited by 2
Pre-event Trends in the Panel Event-Study Design
Freyaldenhoven, S., Hansen, C., & Shapiro, J. M. (2019) · Working paper · doi:10.21799/frbp.wp.2019.27 · Cited by 3
The effect of exit exams on college enrollment : a state level analysis
Kyoore, J. E. (2019) · MOspace Institutional Repository (University of Missouri) · doi:10.32469/10355/72209 · Cited by 2
Policy Evaluation of Waste Pricing Programs Using Heterogeneous Causal Effect EstimationPreprint
Valente, M. (2021) · SSRN Electronic Journal · doi:10.2139/ssrn.3975304 · Cited by 2
Government Accountability and Development: Evidence from Young DemocraciesPreprint
Petkovic, A., & Rahman, T. (2021) · SSRN Electronic Journal · doi:10.2139/ssrn.3971900 · Cited by 2
Intergenerational Effects of Welfare Reform: Adolescent Delinquent and Risky Behaviors
Dave, D., Corman, H., Kalil, A., Schwartz‐Soicher, O., & Reichman, N. E. (2019) · National Bureau of Economic Research · doi:10.3386/w25527 · Cited by 2
“…standard in the applied DD literature, these should be viewed as suggestive given recent evidence indicating that lags and leads in event study-type models with heterogeneous treatment effects may spuriously reflect treatment effects from other periods (Goodman-Bacon, 2018; Abraham & Sun, 2018).”
Fighting for not-So-Religious souls: The role of religious competition in secular conflicts.
Galindo‐Silva, H., & Tchuente, G. (2021) · Journal of Economic Behavior & Organization · doi:10.1016/j.jebo.2021.08.027 · Cited by 1
Xia, X., Wang, F., Xie, S., Xian, J., & Chen, X. (2026) · Frontiers in Public Health · doi:10.3389/fpubh.2026.1764174 · Cited by 1
“To address the potential bias arising from heterogeneous policy effects, this study employed the interactive weight estimation method (IW) as proposed by Sun et al. (13) to conduct a parallel trend test, thereby mitigating concerns associated with heterogeneous treatment effects.”
Local labor market effects of nuclear power plants
Altindag, D. T., Taha, R. E. C., Jones, J. U., & Seals, A. (2025) · Journal of Policy Analysis and Management · doi:10.1002/pam.70029 · Cited by 1
Information, Financing, and InvestmentPreprint
Bird, A., Ertan, A., Karolyi, S. A., & RUCHTI, T. G. (2018) · SSRN Electronic Journal · doi:10.2139/ssrn.3098916 · Cited by 1
Incentivizing Demand for Supply-Constrained Care: Institutional Birth in India
Andrew, A., & Vera‐Hernández, M. (2020) · doi:10.1920/wp.ifs.2020.520 · Cited by 1
#StayAtHome: Social Distancing Policies and Mobility in Latin America and the Caribbean
Aromí, J. D., Bonel, M. P., Cristia, J., Llada, M., Pereira, J. I., Pulido, X., & Santamaría, J. (2020) · doi:10.18235/0002685 · Cited by 1
Imperfect Competition and Rents in Labor and Product Markets: The Case of the Construction IndustryPreprint
Kroft, K., Luo, Y., Mogstad, M., & Setzler, B. (2020) · SSRN Electronic Journal · doi:10.2139/ssrn.3625752 · Cited by 5
MiFID II: The Impact of Double Volume Cap Mechanism on Market QualityPreprint
Ran, Z. (2021) · SSRN Electronic Journal · doi:10.2139/ssrn.3891160 · Cited by 1
Recreational Marijuana Laws and the Use of Opioids: Evidence from Nsduh MicrodataPreprint
Ali, M. M., McClellan, C. B., Mutter, R. L., & Rees, D. I. (2021) · SSRN Electronic Journal · doi:10.2139/ssrn.3897522 · Cited by 1
Buying voter support for unpopular policies: Evidence from German nuclear power plantsPreprint
Valentim, A., Klüver, H., & Erfort, C. (2023) · OSF Preprints (OSF Preprints) · doi:10.31219/osf.io/jnarh · Cited by 1
Bribe-SwitchingPreprint
Desierto, D. A., & Pavlik, J. B. (2021) · SSRN Electronic Journal · doi:10.2139/ssrn.3843425
MiFID II: the Impact of Double Volume Cap Mechanism on Market QualityPreprint
Ran, Z. (2021) · SSRN Electronic Journal · doi:10.2139/ssrn.3879142
Fighting against hunger: A country-wide intervention and its impact on birth outcomes
Beltran-Silva, F. (2023) · World Development · doi:10.1016/j.worlddev.2023.106202
Marijuana legalization and drug abuse as a cause for entry into foster care
Wallace, C. T., Clark, C., & Evans, B. A. (2024) · Review of Economics of the Household · doi:10.1007/s11150-024-09704-x
“We employ the DID M estimator from de Chaisemartin and D ’ Haultfoeuille (2022) as well as the interaction-weighted (IW) estimator from Sun and Abraham (2021) to estimate the change in entries related to drug abuse for states in which marijuana use is legal relative to states in which use is…”
Samper, J. A. (2022) · SSRN Electronic Journal · doi:10.2139/ssrn.4157881
Zhang, B., Dai, M., Guo, X., & Yuan, C. (2026) · Land · doi:10.3390/land15101932
Metadata
- Authors
- Sun, L., & Abraham, S.
- Affiliations
- Massachusetts Institute of Technology (US)
- Corresponding author
- Sun, Liyang
- Venue
- SSRN Electronic Journal
- Publisher
- Elsevier BV
- Published
- 2018
- ISSN
- 1556-5068 (ISSN-L)
- OpenAlex
- W3122004222
- Cited by
- 118
- References
- 10
- Citations by year
- 2026: 4 · 2025: 3 · 2024: 3 · 2023: 8 · 2022: 12 · 2021: 35 · 2020: 28 · 2019: 18 · 2018: 7
- Type
- posted content
- Language
- en
- Open access
- green
- doi.org
- Locations
- SSRN Electronic Journal (acceptedVersion)
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Social Sciences › Economics, Econometrics and Finance › Economics and Econometrics › Spatial and Panel Data Analysis
- Social Sciences › Public Administration › Evaluation and Performance Assessment
- Keywords
- dynamic treatment effects, treatment effect heterogeneity, event study, two-way fixed effects
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.