Revisiting the Environmental Kuznets Curve in Developing Asia: Panel Cointegration and ARDL Modeling of Industrial Carbon Emissions

This study re-examines the validity of the Environmental Kuznets Curve (EKC) hypothesis across 15 developing Asian economies over the period 1995–2023. Utilizing Panel Autoregressive Distributed Lag (ARDL) cointegration techniques, Pooled Mean Group (PMG) estimators, and Cross-Sectionally Augmented ARDL (CS-ARDL) models, we evaluate the long-run and short-run dynamics between GDP per capita, industrialization intensity, renewable energy consumption, and carbon dioxide emissions. The empirical findings strongly support an inverted U-shaped EKC relationship, identifying a mean income turning point at $14,500 GDP per capita. Contour surface mapping and stacked sector decompositions demonstrate that power generation and heavy manufacturing drive peak emissions. Renewable energy integration exerts a statistically significant negative long-run impact on carbon emissions (β = -0.284, p < 0.001), whereas industrial structural expansion accelerates emissions growth in low-income sub-samples. Policy implications highlight the necessity of combining carbon pricing with green technology transfer.