Abstract: The authors use the structural decomposition method to develop the energy intensity decomposition model and to decompose energy intensity into five determinants: energy consumption coefficient, Leontief inverse coefficient, final demand structure, final demand level, and final energy consumption coefficient. Then the RAS method is used to decompose energy consumption coefficient and Leontief inverse coefficient into changes both in structure and efficiency. Empirical study is carried out on the constant price energy-input-output tables from 1987 to 2005. One main conclusion is that between 1987 and 2002, input structure accounts for most of the decline in energy intensity. However, the total demand of production per unit final demand and final demand structure explain the energy intensity increase between 2002 and 2005.

Key words: energy intensity, structural decomposition method, input-output technology, RAS method

摘要： 利用指数分解法, 将影响能源强度的因素分解成能源消耗系数、完全需要系数、最终需求结构系数、最终需求和最终能源消耗系数, 并建立能源强度分解模型。然后, 利用 RAS 方法将能源消耗系数和完全需要系数分别分解成结构和效率系数。通过 1987?2005 年的实物价值型能源投入产出可比价序列表, 进行实证分析, 得出结论: 1987?2002 年中国能源强度下降主要受到投入结构变动的影响, 2002?2005 年能源强度反向上升是受到完全生产完全需要结构和最终需求结构双重作用的结果, 真实的能源效率和真实技术进步对其影响很小。

关键词: 能源强度, 结构分解技术(SDA), 投入产出, RAS法