Asymmetric roles of innovation and renewable energy in shaping carbon emissions in China

Achieving sustained economic growth while reducing carbon emissions remains a central challenge for China's sustainability transition. Despite the rapid expansion of renewable energy capacity and innovation activity, carbon emissions continue to rise, suggesting that the environmental effects of technological progress and energy restructuring may be nonlinear and asymmetric. Understanding whether positive and negative changes in innovation and renewable energy exert different impacts on emissions is therefore fundamental for effective climate and development policies. This study examines the asymmetric relationships between innovation, renewable energy consumption, economic growth, trade openness, and CO 2 emissions in China from 1990 to 2023. The analysis employs a Nonlinear Autoregressive Distributed Lag (NARDL) model as the base framework, allowing positive and negative shocks in innovation and renewable energy to affect emissions differently in the short and long run. Long-run cointegration is examined using the NARDL bounds testing approach, while dynamic multiplier functions trace adjustment paths following asymmetric shocks. Robustness is assessed through alternative lag specifications, a linear ARDL benchmark model, and extensive diagnostic and stability tests, including CUSUM and CUSUMSQ. The results reveal asymmetries. Positive shocks to innovation and renewable energy reduce CO 2 emissions, while negative shocks increase emissions. Economic growth continues to raise emissions. Trade openness raises emissions in the short run. Policies should stabilize R&D, prioritize grid integration and storage, and strengthen links with the emissions trading system.