Decomposition Analysis of Changes in Metropolitan Energy Consumption in China Based on the Logarithmic Mean Divisia Index

The energy consumption of Chinas metropolitan areas has expanded from production areas to other areas such as the service area, and characteristics differ across small and medium-sized cities. Using the Logarithmic Mean Divisia Index (LMDI) decomposition method, we conducted a factor decomposition analysis on changes in energy consumption for 32 Chinese metropolises from 1995–2012 that had a resident population of at least 1 million people in 1995. On the basis of a comprehensive consideration of urban economic growth, population expansion and spatial expansion, energy consumption factors can be decomposed into five indicators: economic scale, energy consumption per unit of output value, energy consumption per capita, population density and energy space support coefficient. We valued the contribution of each factor to metropolises' energy consumption to determine generality and regional differences.