Dynamic evolution of innovation efficiency of China’s new energy enterprises

Introduction Analyzing the dynamic evolution and convergence of innovation efficiency in China’s new energy enterprises is critical for optimizing energy structures and guiding high-quality development under dual-carbon goals. This study examines spatiotemporal patterns and drivers from 2015–2023. Methods Innovation efficiency was measured via SFA model, differentiating R&D and transformation phases. σ - and β -convergence tracked disparities and catch-up dynamics. A threshold regression model identified nonlinear macroeconomic impacts on convergence, using 2,182 firm-year observations across 30 provinces. Results (1) The innovation efficiency of China’s new energy enterprises is relatively low, with significant spatial and temporal differences but a consistent upward trend. Specifically, the innovation efficiency of China’s new energy enterprises ranges from 0.55 to 0.71 in the R&D phase and from 0.13 to 0.51 in the transformation phase. (2) 1/3 of the new energy enterprises show a high R&D-high transformation mode, while another 1/3 operate under a low R&D-low transformation mode. The σ -convergence of innovation efficiency across provinces is not evident, except for the R&D phase of enterprises in the eastern and western regions, where substantial β -convergence is observed. (3) The threshold model suggests that urbanization construction and economic development play a crucial role in influencing the convergence of innovation efficiency among China’s new energy enterprises. Discussion Persistent R&D-transformation gaps necessitate region-specific policies. Western China should enhance technology absorption, while central/eastern regions require optimized innovation ecosystems.