[Workshop] Causal Inference Workshop
From Financial Event Studies to Causal Inference
Thursday, May 20th, 2027 on IÉSEG Lille Campus
—————————————–
Many research questions in the social sciences aim to establish a causal relationship between a particular element and its expected effects. In recent years, the quality of empirical studies designed to address these questions has significantly improved, particularly through more systematic use of identification strategies previously developed in economics. These strategies stem from economists’ growing awareness of the limitations of certain methods and were recognized with the Nobel Prize in Economics in 2021[1]. Consequently, this “credibility revolution” in causal inference has naturally led to the recent and rapid adoption of the most sophisticated identification techniques, particularly in the field of corporate finance[2].
workshop topic
Financial event studies are among the most widely used empirical methods in finance and accounting. They measure stock price reactions to corporate events, such as mergers and acquisitions, earnings announcements, and regulatory decisions. But do abnormal returns necessarily identify causal effects? What assumptions are needed to interpret them causally? And how can modern causal inference methods help researchers construct more credible counterfactuals? The AIC 2027 workshop will address these questions by connecting traditional financial event studies with recent developments in causal inference.
The workshop will be organized around two main references: Goldsmith-Pinkham and Lyu, Causal Inference in Financial Event Studies, which provides the central framework, and Abadie (2021), Using Synthetic Controls, which offers a comprehensive introduction to synthetic control methods, their assumptions, and practical implementation. The day will combine theoretical foundations, identification issues, methodological extensions, and a hands-on application to mergers and acquisitions (M&A) announcements.
Targeted audience
The workshop is intended for academic researchers interested in empirical work, primarily in causal inference. Although the empirical framework presented is mainly based on examples from economics, finance, and accounting, the fields of application are broad and encompass all areas of management and social sciences (accounting, marketing, strategy, human resources, organization, political science, etc.).
Practicalities & Registration
- The day will be moderated by faculty members from IAE Lille and IÉSEG (accounting and finance departments). Registration is free, but compulsory by April 20, 2027 (11:59 pm).
Organizing committee
Jean-Gabriel Cousin (Univ. Lille IAE France – LUMEN), Marion Dupire (IÉSEG – LEM), Jean-Yves Filbien (Univ. Lille, IAE France – LUMEN), Gael Imad’Eddine (Univ. Lille, IAE France – LUMEN), Mohamed Zakriya (IÉSEG – LEM), Romain Oberson (IÉSEG – LEM), Francesco Roccazzella (National Bank of Belgium).
program
- Session 1 — Financial Event Studies: A Causal Perspective (1 hour) Traditional event studies and abnormal returns; potential outcomes and causal effects; the missing counterfactual; expected-return models and identifying assumptions; CAR and BHAR. Main reference: Goldsmith-Pinkham & Lyu.
- Session 2 — Identification Assumptions and Sources of Bias (1 hour) Limited anticipation; random assignment versus random event timing; misspecified expected-return models and omitted factors; short-run versus long-run event studies; identification threats and empirical diagnostics. Main reference: Goldsmith-Pinkham & Lyu.
- Session 3 — Synthetic Controls and Modern Counterfactual Methods (1 hour) Synthetic control estimators; donor pool selection and pre-treatment fit; identification assumptions and diagnostics; generalized synthetic control and extensions; links with traditional financial factor models. Main references: Abadie (2021); Goldsmith-Pinkham & Lyu.
- Session 4 — Hands-on Application: M&A Announcement Returns (2 hours) Using M&A announcements, participants will compare traditional event study estimators (market-adjusted, market model, and multifactor models) with synthetic control approaches. The session will focus on counterfactual construction, estimation results, diagnostics, and causal interpretation, using R or Stata.
Main References
- Abadie, A. (2021). Using Synthetic Controls. Journal of Economic Literature, 59(2), 391–425.
- Goldsmith-Pinkham, P., & Lyu, T. Causal Inference in Financial Event Studies. Working paper (version used for the workshop).
Additional Reference
- Bowen, D. E., Frésard, L., & Taillard, J. P. (2017). What’s Your Identification Strategy? Innovation in Corporate Finance Research. Management Science, 63(8), 2529–2548.
[1] David Card, Joshua Angrist et Guido Imbens “for their methodological contributions to the analysis of causal relationships” The Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel 2021 – NobelPrize.org
[2] The proportion of academic articles using modern identification strategies was close to zero until the late 1980s, then steadily increased to over 50% by 2012 (Bowen, Frésard et Taillard, 2017)