🤖 AI Summary
This study addresses the limitations of conventional difference-in-differences methods in analyzing heterogeneous treatment effects across groups, which are often confounded by differences in covariate distributions, conservative inference procedures, and restrictive parametric interaction structures. To overcome these challenges, the paper proposes a novel estimator—the Balanced Group Average Treatment Effect on the Treated (BGATT)—which, under the parallel trends assumption, effectively disentangles covariate composition differences from genuine treatment effect heterogeneity. BGATT offers clear identifiability and interpretability while accommodating flexible, high-dimensional modeling. The authors construct an influence-function-based estimator that achieves √n-consistency and asymptotic normality, enabling efficient nuisance parameter estimation via machine learning. Both theoretical analysis and simulation studies demonstrate that the proposed method delivers superior finite-sample performance, substantially enhancing the accuracy and robustness of heterogeneity assessments.
📝 Abstract
Understanding how treatment effects vary across groups is central to policy evaluation. In Difference-in-Differences designs, heterogeneity is often studied using subgroup or triple-difference analyses, which can suffer from conservative inference, reliance on parametric interaction structures, and sensitivity to differences in covariate distributions across groups. We propose the Balanced Group Average Treatment Effect on the Treated (BGATT), a new estimand that isolates heterogeneity in treatment responses from differences in covariate composition and is identified under standard conditional parallel-trends assumptions. BGATT provides a transparent target for comparing group-specific treatment effects. We derive an influence-function representation and develop estimators that are $\sqrt{n}$-consistent and asymptotically normal under flexible machine-learning estimation of high-dimensional nuisance components, enabling valid inference on both group-specific effects and differences across groups. Simulation evidence shows favorable finite-sample performance.