Publication

This publication explains the risks of using the generalized method of moments (GMM) estimators for dynamic panel models and illustrates them with reference to two early applications of the estimators to economic growth. The author argues that endogenous causation proves hard to rule out in both papers. The publication further suggests that for the GMM estimators to be credible, researchers must report the instrument count and aggressively test estimates and specification test results for robustness to reductions in that count.