Uncovering the non-linear effects of renewable energy on energy intensity: Evidence from a dynamic panel threshold model

This study examines whether the energy-efficiency payoff from renewable energy deployment depends on a country's level of development by modeling a nonlinear relationship in a balanced panel of 104 countries over 2010–2020. Energy intensity serves as a widely used proxy for economy-wide energy efficiency and is closely linked to progress toward the United Nations Sustainable Development Goals (SDGs). Methodologically, we employ a dynamic panel threshold GMM framework that captures persistence in energy intensity and addresses the dynamic endogeneity induced by the lagged dependent variable, using GDP per capita as the threshold variable to delineate income regimes. The estimates reveal a statistically significant income threshold at USD 6112 per capita (2017 PPP). Above this threshold, an increase in the renewable energy share is associated with a reduction in energy intensity that is significant at the 10% level; below the threshold, the estimated effect is smaller and statistically insignificant. The results imply that lower-income countries may need to prioritize system-readiness investments and capacity building to convert renewable energy scale-up into economy-wide efficiency gains, whereas higher-income economies can amplify these gains by focusing on innovation, smart-grid modernization, and demand-side efficiency measures.