Human Development, Industrialization, Renewable Energy, and Environmental Taxation as Determinants of CO2 Emissions: Evidence from Heterogeneous Panel and Quantile Analysis

Abstract

Carbon dioxide emissions remain the principal driver of anthropogenic climate change, yet the socioeconomic forces that shape per capita emissions differ markedly across the distribution of emitting economies. This study examines the determinants of carbon dioxide emissions per capita using a balanced structure of 1189 country–year observations drawn from a panel of 41 countries over the period 1995–2023. Four explanatory variables are considered: the Human Development Index, the share of manufacturing value added in gross domestic product, the share of primary energy sourced from renewables, and environmentally related tax revenue expressed as a percentage of gross domestic product. Cross-sectional dependence and slope heterogeneity are formally tested and confirmed, motivating the use of second-generation panel techniques throughout. The Pesaran cross-sectionally augmented Dickey–Fuller test reveals a mixed order of integration, and the Westerlund error-correction-based cointegration test confirms a long-run equilibrium relationship among the series. Long-run marginal effects estimated through the Common Correlated Effects Mean Group and Augmented Mean Group estimators show that human development, manufacturing intensity, and environmentally related tax revenue raise per capita emissions on average, while renewable energy penetration lowers them. Method of Moments Quantile Regression uncovers substantial distributional heterogeneity: the manufacturing effect changes across the conditional distribution, shifting from emission-enhancing at lower quantiles to emission-reducing above the median. Human development and environmentally related tax revenue remain positively associated with emissions, although their effects weaken at higher emission levels. By contrast, the mitigating influence of renewable energy is stable across all quantiles. Dumitrescu–Hurlin causality results indicate bidirectional relationships between emissions and all covariates, with the feedback effect from emissions to human development representing the weakest causal linkage. The findings qualify simple average-based policy prescriptions and support differentiated, distribution-sensitive climate and fiscal policy design.