Assessing renewable energy efficiency to identify improvement strategies: A network data envelopment analysis approach

China has been analyzing, assessing, and enhancing renewable energy efficiency as crucial steps in mitigating global energy depletion and climate change. Most studies evaluating renewable energy efficiency (REE) use a single-stage data envelopment analysis (DEA) model, and thereby, ignore the interactions among economic, social, and environmental subsystems. In contrast, this study develops a matrix-type structure network DEA model that considers the influence of undesirable outputs and mutual interactions among these three subsystems to measure REE in China's 30 provinces. Examining the 2005 to 2019 timeframe, the study shows that: 1) the changes in China's REE reflect a significant "V" shape during this timeframe, with 2013 marking the boundary, owing to an improvement in the efficiency of the environmental subsystem; 2) REE improved in 15 provinces, decreased in 13, and barely changed in the other two; 3) REE tended to converge between 2005 and 2011 and diverge between 2011 and 2019; and 4) for further REE improvement, total fixed asset investments, fossil energy imports, and household waste collected and transported need to decrease by 26 %, at least; while GDP per capita, excellent air quality, and years of education need to increase by >4 %. The expansion of the advanced provinces contributes to REE improvement for international economies and industrial communities. It will accelerate the free movement of production inputs, facilitating international energy saving and sustainable development.