We provide a common framework that relates traditional event study estimation methods in finance to a modern approach for causal event studies. The framework provides a model for abnormal returns that nests the fitted market model (the traditional approach) and more recent approaches based on difference-in-differences and synthetic control methods. We show that a synthetic control method in this context can be understood as a synthetic portfolio. We provide a simulation exercise and an empirical application, using mergers and acquisitions as the event of interest, to evaluate the performance of the different models within the framework. Our results indicate that causal inference methods such as synthetic matching or difference-in-differences do not provide an improvement over the traditional approach based on the fitted market model. Although the fitted market model may not always abide by the conditions under which it is considered a proper counterfactual, its performance indicates that it is still a good potential outcome.
