『Abstract
We employ decomposition analysis and econometric analysis to
investigate the driving forces behind China's changing energy
intensity using a provincial-level panel data set for the period
from 1995 to 2009. The decomposition analysis indicates that:
(a) all of the provinces except for a few experienced efficiency
improvement, while around three-fourths of the provinces' economics
became more energy intensive or remained unchanged; (b) consequently
the efficiency improvement accounts for more than 90% of China's
energy intensity change as opposed to the economic structural
change.
The econometric analysis shows that the rising income plays a
significant role in the reduction of energy intensity while the
effect of energy price is relatively limited. The result may reflect
the urgency of deregulating the price and establishing a market-oriented
pricing system in China's energy sector. The implementation of
the energy intensity reduction policies in the Eleventh Five-Year
Plan (FYP) has helped reverse the increasing trend of energy intensity
since 2002. Although the Chinese Government intended to change
the industry-led economic growth pattern, it seems that most of
the policy effects flow through the efficiency improvement as
opposed to the economic structure adjustment. More fundamental
changes to the economic structure are needed to achieve more sustainable
progress in energy intensity reduction.
Keywords: Energy intensity; Energy price; Policy evaluation』
1. Introduction
2. Decomposition analysis of China's changing energy intensity
at the provincial level
2.1. The decomposition method
2.2. The decomposition results
3. Econometric analysis of China's changing energy intensity at
the provincial level
3.1. Model specification and explanatory variables
3.1.1. Price
3.1.2. Income
3.1.3. Capital-labor ratio
3.1.4. Annual investment
3.1.5. Urbanization
3.1.6. Energy resource endowment
3.1.7. Policy
3.2. Estimation results and discussion
4. Policy implications and conclusion
Acknowledgments
Appendix A. instrumented variable tests and estimation results
References