Working Paper

Estimation and Inference for Synthetic Control Methods with Spillover Effects (with Connor Dowd) Feb 2019. [matlab code]

Abstract: The synthetic control method is often used in treatment effect estimation with panel data where only a few units are treated and a small number of post-treatment periods are available. Current estimation and inference procedures for synthetic control methods do not allow for the existence of spillover effects, which are plausible in many applications. In this paper, we consider estimation and inference for synthetic control methods, allowing for spillover effects. We propose estimators for both direct treatment effects and spillover effects and show they are asymptotically unbiased. In addition, we propose an inferential procedure and show it is asymptotically unbiased. Our estimation and inference procedure applies to cases with multiple treated units or periods, and where the underlying factor model is either stationary or cointegrated. In simulations, we confirm that the presence of spillovers renders current methods biased and have distorted sizes, whereas our methods yield properly sized tests and retain reasonable power. We apply our method to a classic empirical example that investigates the effect of California’s tobacco control program as in Abadie et al. (2010) and find evidence of spillovers.


On the Empirical Content of the Beckerian Marriage Model (with Xiaoxia Shi and Matthew Shum) Economic Theory, 2019, 67(2), 349–362.

Abstract: This note studies the empirical content of a simple marriage matching model with transferable utility, based on Becker (J Polit Econ 81:813–846, 1973). Under Becker’s conditions, the equilibrium matching is unique and assortative. However, this note shows that when the researcher only observes a subset of relevant characteristics, the unique assortative matching does not uniquely determine a distribution of observed characteristics. This precludes standard approaches to point estimation of the underlying model parameters. We propose a solution to this problem, based on the idea of “random matching.”