Since the 1980s, the Chinese economy has developed rapidly, with an av- erage annual growth rate around 10%. Energy consumption in China has greatly in- creased as well. This paper investigates the relationship betwee...Since the 1980s, the Chinese economy has developed rapidly, with an av- erage annual growth rate around 10%. Energy consumption in China has greatly in- creased as well. This paper investigates the relationship between energy consumption, economic development and temperature in China by adopting provincial panel data from 1990 to 2011. Different from existing studies, in this paper, we use a panel smooth transition regression （PSTR） model to estimate the non-linearity relationship. Four different threshold variables including two lagged endogenous variables and two important exogenous variables have been considered. We find that energy intensity and the ratio of gross capital formation are suitable for the non-linearity model. The estimated elasticities of time dynamic indicate that energy consumption is income in- elastic and temperature inelastic. Elasticities of real income at first increase and then decrease, however, elasticities of temperature gradually increase after the year 1993. Last of all, we propose some policy implications.展开更多
基金The authors thank the reviewers for their careful reading and provid- ing some pertinent suggestions. The research is supported by the National Natural Science Foundation of China （No. 71073009 ）, Chinese science and technology supporting program （No. 2012BAC20B08） and Tianjin City High School Science ＆ Technology Fund Planning Project （No. 20130823）.
文摘Since the 1980s, the Chinese economy has developed rapidly, with an av- erage annual growth rate around 10%. Energy consumption in China has greatly in- creased as well. This paper investigates the relationship between energy consumption, economic development and temperature in China by adopting provincial panel data from 1990 to 2011. Different from existing studies, in this paper, we use a panel smooth transition regression （PSTR） model to estimate the non-linearity relationship. Four different threshold variables including two lagged endogenous variables and two important exogenous variables have been considered. We find that energy intensity and the ratio of gross capital formation are suitable for the non-linearity model. The estimated elasticities of time dynamic indicate that energy consumption is income in- elastic and temperature inelastic. Elasticities of real income at first increase and then decrease, however, elasticities of temperature gradually increase after the year 1993. Last of all, we propose some policy implications.
