"What's your shape?": A data-driven approach to estimating the environmental Kuznets Curve

The substantial literature on the existence of an inverted U-shaped relationship between environmental degradation and economic growth-known as the Environmental Kuznets Curve (EKC)-has produced very mixed evidence. This largely depends on model and variable selection uncertainty. We address these issues relying on Bayesian Model Averaging techniques. Our results indicate that the EKC has an inverted-N shape, with almost all emerging economies analyzed on the upward segment of the curve displaying a positive association between per capita gross domestic product and carbon dioxide emissions, and most advanced economies analyzed on the second downward segment of the curve. These findings are robust to the use of different measures of environmental pollution and (non-Bayesian) Least Absolute Shrinkage and Selection Operator (LASSO) regression techniques.