As an important single source to carbon emissions, China’s power industry should bear social responsibility for mitigating climate change. To explore what low-carbon development means for the industry, a novel approach that combines the extended multilevel LMDI model with Tapio algorithm was conducted to study the drivers of carbon emissions in the power industry and whether CO₂ emissions from power output is out of sync with economic development, covering the period from 1996 to 2016. Our results come to the following:
1. Carbon emissions from electricity output are characterized by increases and volatility, with an average annual growth rate of 7.05%. The carbon emission factor of electricity, facilitating to compute CO2 data, shows a decline.
2. The positive driving factors are economic activity effect (169.53%), population scale effect (9.29%), fuel mix structure effect (0.41%), and electricity trade effect (1.05%); the negative driving factors are electricity intensity effect (-46.38%), power generation efficiency effect (-24.93%), and power generation structure effect (-8.97%).
3. Weak decoupling and expansive decoupling are the main status during the research period. The electricity intensity effect is the main force to promote the decoupling process.
4. The market-oriented reform in the power industry in 2003 has a significant effect. The generationside competition mechanism successfully changes the historical developmental trend of the decoupling elastic index.