Answers are grounded in this paper and cite their sources.
Carbon intensity is a key indicator of progress towards decarbonisation in fast-growing economies. This study examines the long-run and short-run determinants of carbon intensity in ten large emerging economies and Japan over 1990-2023 (374 annual observations). The variables show strong cross-sectional dependence and heterogeneous slopes, so we estimate a cross-sectionally augmented autoregressive distributed lag (CS-ARDL) model with the dynamic common correlated effects mean group (DCCE-MG) estimator. We then test how sensitive the results are to the specification, the sample and the estimator. A Kaya decomposition separates the behavioural effects of income from its mechanical link with carbon intensity. The results show that per capita energy use and the renewable share of final energy consumption are the most robust determinants of carbon intensity across all specifications. The near-unit negative income coefficient largely reflects the Kaya identity: per capita emissions still rise with income, so decoupling is relative rather than absolute, and there is no evidence of an inverted-U relationship. Trade openness has a small positive effect that is not robust to excluding Japan, and FDI, urbanisation and population growth are not significant. Carbon intensity adjusts quickly to its long-run equilibrium, although adjustment speeds and elasticities differ across countries. The findings suggest that decarbonising the energy system, rather than relying on income growth, is the main policy lever for reducing carbon intensity in these economies.
No takes yet. Share an insight, caveat, or question.
Gouraram et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: