Synthetic Controls with Multiple Outcomes

TL;DR

This paper extends the synthetic control method to handle multiple outcomes simultaneously, enhancing the accuracy of treatment effect estimates especially with limited pre-treatment data or multiple outcomes.

Contribution

The paper introduces a novel multi-outcome synthetic control framework that incorporates related outcomes to improve estimation reliability.

Findings

Enhanced treatment effect estimation with multiple outcomes

Improved robustness with limited pre-treatment periods

New perspective on German reunification analysis

Abstract

We generalize the synthetic control (SC) method to a multiple-outcome framework, where the conventional pre-treatment time dimension is supplemented with the extra dimension of related outcomes in computing the SC weights. This generalization improves the reliability of treatment effect estimation, and can be particularly useful for evaluating the effect of a treatment on multiple outcomes or when only a small number of pre-treatment periods are available. To illustrate our method, we provide a new perspective on the classic SC application to the 1990 German reunification.