Sources of Size Distortion in Statistical Inference for Average Treatment Effect by Synthetic Controls : A Cautionary Note

The recent literature has underscored the importance of properly accounting for nonstationarity in the application of the synthetic control methods and proposed inference methods to do so. This note studies the size properties of the method proposed by Masini and Medeiros (2022) through simulations. Our results reveal two sources of size distortions that applied researchers should be aware of. While one source may be specific to the inference method under study, the other likely applies more broadly