Does Growth Clean Up After Itself? A Multi-Gas Test of the Environmental Kuznets Curve

Abstract

This study reexamines the Environmental Kuznets Curve (EKC) hypothesis in developing economies, identifying which greenhouse gas CO?, CH?, or N?O most influences environmental degradation, and combining all three into a composite index via Principal Component Analysis. Using a cleaned panel of 51 developing economies, 2000–2024. The study estimates two-way fixed effects models, an Anderson–Hsiao dynamic-panel check, and a cross-sectional dependence robust CCEMG estimator. The inverted-U EKC holds only for N?O under fixed effects (turning point ? US$6,000 GDP per capita, already crossed by 23% of the sample), though CCEMG motivated by a Pesaran CD test confirming strong cross-sectional dependence tempers this to a directional rather than significant result. Evidence for CO?, CH?, and the composite index remains weak throughout. Renewable energy consistently lowers CO? emissions; nonrenewable energy is the strongest, most robust driver of degradation across specifications; urbanization raises emissions significantly. These results show that EKC conclusions in developing economies depend heavily on which pollutant is used and which econometric technique is applied, with implications for energy and industrial policy.