EURO 2025 Leeds
Abstract Submission

1930. CAUSAL EFFECTS OF POLICY INTERVENTIONS ON PRODUCTIVE PERFORMANCE: INTEGRATING SYNTHETIC CONTROL METHODS INTO THE NONPARAMETRIC APPROACH

Invited abstract in session WA-60: DEA applications in Policy Making and Planning I, stream Data Envelopment Analysis and its applications.

Wednesday, 8:30-10:00
Room: Western LT

Authors (first author is the speaker)

1. Xun Zhou
Surrey Business School, University of Surrey

Abstract

Examining the causal effects of policy interventions on productive performance is critical for policy analysis and managerial decision-making. A typical method in the nonparametric approach is to regress efficiency estimates on policy interventions and contextual variables. However, this method has been criticized for ignoring the correlations between inputs and policy interventions and/or contextual variables and hence subject to endogeneity in the efficiency estimation. To address this issue, this study proposes to integrate synthetic control methods (SCM), a data-driven causal inference technique, into the nonparametric approach. SCM estimates the causal effect by constructing a counterfactual of the treated unit (i.e., synthetic control) using a convex combination of control units, analogous to the rationale in DEA. In the integrated framework, 1) SCM ensures the treated unit and the synthetic control closely match before the treatment, with respect to the input, output, and contextual variables; 2) the nonparametric approach estimates the post-treatment efficiency both for the original sample of units and for the synthetic sample where the treated unit is replaced by the synthetic control while retaining all the control units. The causal impact of the policy intervention is then estimated by comparing the efficiency of the treated unit with that of the synthetic control, that is, the counterfactual efficiency of the treated unit in the absence of the intervention.

Keywords

Status: accepted

Back to the list of papers