2 Research on Urbanization and Low-Carbon Development in China
47
regional disparities for this impact? What should be the proper size and distribution
of future Chinese cities? Answers to these questions will be of great help in driving
scientific urbanization, energy conservation and emission reduction in the country.
2.3 Impact Mechanism of China’s Urbanization on Carbon
Emissions and Empirical Analysis
2.3.1 Analysis Framework Based on STIRPAT Model
The factors driving the growth of carbon dioxide emissions include economic development, industrial structure, technological progress, energy consumption, total population, people’s life, cultural mindset, etc, which are constantly shaped and changed in
the process of urbanization. Therefore, it is a challenging systematic project to study
the mechanism of urbanization on carbon emission. Ehrlich & Holdren (1971) first
put forward the famous IPAT model, which explored the effect of human activities
on the natural environment from the three main dimensions of population, affluence
and technology. Subsequently, Dietz & Rosa (1997) modified the IPAT model, and
took into account the Stochastic Impact of population growth, wealth change and
technological progress on the environment, and proposed the STIRPAT model:
I it = aP
b
it A
c
it T
d
it ε it
The logarithmic form is:
ln I it = a + b(ln P it ) + c(ln A it ) + d(ln T it ) + ε it
Among them, variables P, A and T have the same meaning as before, and I represents the impact of the environment, usually expressed by carbon dioxide or other
environmental indicators; a represents the fixed constant term, while b, c and d
represent the elastic coefficients of the three variables, which are all undetermined
parameters. In addition, i stands for the observed individual, t is the observed year,
and ε is the error term.
The study of York et al. (2003) found that socio-economic factors or other control
factors can join into the STIRPAT model, assuming these are conceptually consistent with the multiple settings of the model. As a result, STIRPAT is widely used in
environmental and economic research due to its adaptability and inclusiveness. For
instance, Wang and Liancheng (2015) used the fixed assets stock per capita and the
added value per capita in manufacturing as proxy variables of population and wealth
in STIRPAT, and analyzed the correlation between the growth of China’s manufacturing output and the level of energy consumption and carbon dioxide emissions.
Taking OECD countries as an example, Emrah & Ulucak (2019) incorporated energy
R&D expenditure into the technical dimension of STIRPAT model, and explored the
47
regional disparities for this impact? What should be the proper size and distribution
of future Chinese cities? Answers to these questions will be of great help in driving
scientific urbanization, energy conservation and emission reduction in the country.
2.3 Impact Mechanism of China’s Urbanization on Carbon
Emissions and Empirical Analysis
2.3.1 Analysis Framework Based on STIRPAT Model
The factors driving the growth of carbon dioxide emissions include economic development, industrial structure, technological progress, energy consumption, total population, people’s life, cultural mindset, etc, which are constantly shaped and changed in
the process of urbanization. Therefore, it is a challenging systematic project to study
the mechanism of urbanization on carbon emission. Ehrlich & Holdren (1971) first
put forward the famous IPAT model, which explored the effect of human activities
on the natural environment from the three main dimensions of population, affluence
and technology. Subsequently, Dietz & Rosa (1997) modified the IPAT model, and
took into account the Stochastic Impact of population growth, wealth change and
technological progress on the environment, and proposed the STIRPAT model:
I it = aP
b
it A
c
it T
d
it ε it
The logarithmic form is:
ln I it = a + b(ln P it ) + c(ln A it ) + d(ln T it ) + ε it
Among them, variables P, A and T have the same meaning as before, and I represents the impact of the environment, usually expressed by carbon dioxide or other
environmental indicators; a represents the fixed constant term, while b, c and d
represent the elastic coefficients of the three variables, which are all undetermined
parameters. In addition, i stands for the observed individual, t is the observed year,
and ε is the error term.
The study of York et al. (2003) found that socio-economic factors or other control
factors can join into the STIRPAT model, assuming these are conceptually consistent with the multiple settings of the model. As a result, STIRPAT is widely used in
environmental and economic research due to its adaptability and inclusiveness. For
instance, Wang and Liancheng (2015) used the fixed assets stock per capita and the
added value per capita in manufacturing as proxy variables of population and wealth
in STIRPAT, and analyzed the correlation between the growth of China’s manufacturing output and the level of energy consumption and carbon dioxide emissions.
Taking OECD countries as an example, Emrah & Ulucak (2019) incorporated energy
R&D expenditure into the technical dimension of STIRPAT model, and explored the