Natural disasters pose major challenges to economic development, yet empirical evidence on their growth effects remains mixed. This study examines whether the macroeconomic impact of disasters depends on infrastructure development, using electricity access as a conditioning variable. Employing a balanced panel of 108 countries from 1996 to 2022, we estimate a panel threshold regression to capture nonlinear disaster–growth relationships. The results indicate that disaster impacts vary systematically across infrastructure regimes. In very low electricity-access contexts, disaster intensity is associated with short-run growth responses consistent with reconstruction-driven dynamics in a very limited subset of observations. Once basic electricity access is achieved, this association weakens and becomes statistically indistinguishable from zero. Beyond this lower regime, the analysis also reveals a broader structural transition in which disaster impacts gradually decouple from economic growth as infrastructure coverage expands. These findings suggest that disaster effects are nonlinear and contingent on development stage. By distinguishing between statistical thresholds and economically meaningful transition zones, the study reconciles mixed evidence in the disaster–growth literature and highlights electricity access as a key determinant of economic resilience.
Keisuke Kokubun (Fri,) studied this question.