Did Chrysler Benefit from Government Assistance? Making Causal Inferences in Small Samples Using Synthetic Control Methodology
Abstract: We assess the impact of U.S. government intervention on the performance of Chrysler which, following the financial crisis in 2008, accepted government support in return for on-going Treasury oversight. To conduct our analysis, we introduce a recently developed statistical methodology—synthetic control—to management research. The method overcomes challenges to causal inference in contexts like ours that are constrained by small samples or few occurrences of the phenomenon of interest. Our synthetic control analysis constructs a replica of the focal firm based on a weighted combination of other firms with similar attributes. We quantify the magnitude and direction of the treatment effect by comparing the actual performance of Chrysler to its counterfactual replica without treatment. We estimate that Chrysler, or its surviving components, would have sold approximately 20% more vehicles in the U.S. through summer 2011 had the company instead relied on private financing and market forces.
