Beyond the Efficiency Dividend: The Non-Linear Nexus of Renewables and Energy Intensity in Emerging Economies

Energy intensity is a core indicator of economy-wide energy efficiency and a central concern for energy-transition policy. This study examines whether renewable energy penetration is monotonically associated with lower energy intensity or whether this association changes across penetration levels, using a balanced panel of 27 emerging economies from 2002 to 2021. The empirical strategy estimates a two-way fixed-effects quadratic renewable-share specification, controlling for GDP per capita, manufacturing value added, urbanization, and natural resource rents, and validates the non-linear pattern using robust inference. We further validate our findings via the Lind-Mehlum U-shape test, dynamic GMM (Generalized Method of Moments) estimators, and spline-based functional-form checks. The results indicate a U-shaped association: renewable expansion is associated with lower energy intensity at low-to-moderate shares, but the marginal association weakens and becomes positive at higher shares. Static and spline specifications place the main turning region roughly between 44% and 54%. Quantile Regression estimates show that the turning point rises across the conditional energy-intensity distribution. Higher GDP per capita is associated with lower energy intensity, while manufacturing share and natural resource rents are associated with higher energy intensity; urbanization shows weaker and distribution-dependent effects. The findings imply that renewable deployment can support energy-intensity reductions in emerging economies, but sustaining these gains at higher renewable shares requires complementary investments in grid flexibility, integration capacity, and structural upgrading.