REGIONAL DIFFERENTIATION IN
INFLUENCING FACTORS OF CLEAN
RENEWABLE ENERGY CONSUMPTION
FROM THE PERSPECTIVE OF AIR
POLLUTION PREVENTION AND CONTROL
Jin Zhan*
Intelligent Manufacturing College, Shanxi Vocational University of Engineering
Science and Technology, Taiyuan, Shanxi, 030004, China.
School of Mechanical Engineering, Taiyuan University of Science and Technology,
Taiyuan, Shanxi, 030024, China.
ggh0546@163.com
Reception: 30/04/2023 Acceptance: 23/06/2023 Publication: 14/07/2023
Suggested citation:
Zhan, J. (2023). Regional differentiation in influencing factors of clean
renewable energy consumption from the perspective of air pollution
prevention and control. 3C Tecnología. Glosas de innovación aplicada a la
pyme, 12(2), 331-345. https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
331
ABSTRACT
Global economic growth is now increasingly conflicting with the sustainable
development strategy. In the context of ecological environmental preservation and air
pollution prevention and control, this paper probes into the regional differentiation in
the influencing factors of clean renewable energy consumption. First and foremost, a
brief analysis of the status quo of clean renewable consumption in China was
outputted, grounded on data on the input and output of 30 provinces and cities
nationwide from 2010 to 2020. Then, national and regional models are built
respectively in virtue of differential GMM, systematic GMM, and bias-corrected LSDV
methods. Furthermore, efforts were invested in dissecting the working mechanism of
the influencing factors and verifying the previous prediction resulting in applying the
Tobit regression method. For every 1% increase in the green finance index, the clean
renewable energy consumption rises by 0.882 accordingly, said the regression
analysis results. Last but not least, it was concluded that the development level of
green finance, internet advance, and technological progress significantly positively
affected clean renewable energy consumption. While the industrial structure, the
degree of openness, and the level of urbanization represented by the proportion of the
secondary industry play hardly-seen impact.
KEYWORDS
Offset correction; Influencing factors; Regression method; Clean renewable energy
consumption; Regional differentiation.
INDEX
ABSTRACT
KEYWORDS
1. INTRODUCTION
2. RESEARCH BASIS
2.1. Current situation of clean renewable energy consumption
2.2. Model parameter setting
3. DATA, VARIABLES, AND DESCRIPTIVE STATISTICS
4. RESULTS AND ANALYSIS
5. DISCUSSION
6. CONCLUSION
REFERENCES
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
332
ABSTRACT
Global economic growth is now increasingly conflicting with the sustainable
development strategy. In the context of ecological environmental preservation and air
pollution prevention and control, this paper probes into the regional differentiation in
the influencing factors of clean renewable energy consumption. First and foremost, a
brief analysis of the status quo of clean renewable consumption in China was
outputted, grounded on data on the input and output of 30 provinces and cities
nationwide from 2010 to 2020. Then, national and regional models are built
respectively in virtue of differential GMM, systematic GMM, and bias-corrected LSDV
methods. Furthermore, efforts were invested in dissecting the working mechanism of
the influencing factors and verifying the previous prediction resulting in applying the
Tobit regression method. For every 1% increase in the green finance index, the clean
renewable energy consumption rises by 0.882 accordingly, said the regression
analysis results. Last but not least, it was concluded that the development level of
green finance, internet advance, and technological progress significantly positively
affected clean renewable energy consumption. While the industrial structure, the
degree of openness, and the level of urbanization represented by the proportion of the
secondary industry play hardly-seen impact.
KEYWORDS
Offset correction; Influencing factors; Regression method; Clean renewable energy
consumption; Regional differentiation.
INDEX
ABSTRACT
KEYWORDS
1. INTRODUCTION
2. RESEARCH BASIS
2.1. Current situation of clean renewable energy consumption
2.2. Model parameter setting
3. DATA, VARIABLES, AND DESCRIPTIVE STATISTICS
4. RESULTS AND ANALYSIS
5. DISCUSSION
6. CONCLUSION
REFERENCES
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
1. INTRODUCTION
The construction of ecological civilization is an important aspect of the long-term
development of the Chinese nation and has always been a hot spot of concern for all
countries. The report of the 19th National Congress states that we should
continuously promote green development, strengthen environmental governance, and
increase the protection of ecosystems [1]. However, the current air pollution problem
faced by human society is a serious constraint to the construction of ecological
civilization [2]. Atmospheric pollution can have catastrophic effects on human
economic and social development, such as climate anomalies, extreme weather such
as dust storms, sea level rise, and global warming [3]. Ecological and environmental
issues are related to the health of the people, the harmony and stability of society, and
the healthy and orderly development of the economy [4]. The management of
atmospheric pollution has become a hot issue of close concern to all sectors of
society [5]. And the use of clean renewable energy has a positive effect on air
pollution control as well as ecological environmental protection [6].
Energy consumption is not only an important aspect of national implementation but
also, it is related to the strategic overall situation of national energy sustainable
development [7]. In the context of sustainable development theory and ecological
modernization, people's requirements for living environment and quality of life are
increasing with the increase in income level [8]. The type of energy consumption in
China is shifting from high pollution to cleanliness [9]. In recent years, the rise of a
low-carbon economy has increased under the premise of the sustainable
development concept [10]. To deeply strengthen the management of the atmospheric
environment, improve the quality of urban and rural air environment, and improve the
ecological environment, China has continuously increased the development and use
of clean energy [11]. The state has also introduced relevant laws and regulations, and
clear requirements for the use of high-pollution fuels in some areas of the province
designating a no-burn zone that requires the use of high-pollution fuels in the no-burn
zone [12]. They should be removed or switched to natural gas, LPG, electricity, or
other clean energy sources promptly. Clean renewable energy resources in China
vary from region to region and within regions due to differences in the economic
development of each region leading to their consumption [13].
The eastern region of China accounts for 41.56% of the total national consumption
in 2020, while western regions account for 38.87% and 19.57%, respectively. The
regional differences are large [14]. In terms of energy use consumption, the five
provinces with the lowest energy consumption accounted for 16.54% of the country's
energy consumption but their combined GDP only accounted for 8.28% of the
country's GDP. The five provinces with the highest energy use consumption account
for 15.22% of the country's energy consumption but their combined GDP accounts for
26.47% of the country's GDP. There are significant regional differences in energy
consumption in China [15]. Therefore, while improving energy use consumption, a
correct understanding of the structural characteristics and regional disparities in
energy use consumption is significant to effectively promote the work of air pollution
prevention and control in China as well as to continuously promote sustainable
development strategies [16-17].
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
333
Clean renewable energy has been studied by many scholars. The literature [18]
presents a model used to describe the development. The literature [19] simulates the
consumption of natural gas by randomly combining several influences including GDP,
population, natural gas imports and exports, and employment while predicting natural
gas consumption by selecting the most realistic equation from all simulated equations.
Literature [20] and literature [21] investigate the impact of the national economic
growth rate through the ARDL method [22-24]. Literature [25] and literature [26]
studied the impact of various factors on renewable energy consumption based on a
multivariate framework. Literature [27] conducted research using panel data from
nearly 35 years and obtained that, from a long-term perspective, natural gas energy
consumption has an impact on the GDP growth of GCC countries. Literature [28]
proposes a parameter estimation method that eliminates omitted variable bias due to
unobserved cross-sectional individual effects by differencing. In addition, the literature
[29] used the Granger causality test to obtain the relationship between natural gas
consumption and GDP. Literature [30] argues that marketization facilitates the
formation of a virtuous response between renewable energy and consumption,
guiding the consumption of clean renewable energy through price signals.
The aforementioned studies include the search for clean and renewable
alternatives to fossil energy through data models. The relationship between
consumption has been explored and the factors influencing have been studied. This
study further extends the study of natural gas energy to clean renewable energy
based on the previous study. We analyze the factors influencing the consumption of
clean renewable energy in the context of ecological environmental protection and air
pollution prevention. We also combine the panel data of 30 Chinese provinces to build
a mathematical model and estimate the model using differential GMM, systematic
GMM, and bias-corrected LSDV methods for the national and regional levels,
respectively. Based on this, the Tobit regression method is applied to validate the
results. Based on the data model, the influencing factors of energy consumption were
analyzed, to provide some theoretical references for ecological environmental
protection and air pollution control.
2. RESEARCH BASIS
2.1. CURRENT SITUATION OF CLEAN RENEWABLE
ENERGY CONSUMPTION
To examine the regional differences in energy consumption, the specifics of each
region in 2020 were investigated and plotted in Figure 1.
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
334
Clean renewable energy has been studied by many scholars. The literature [18]
presents a model used to describe the development. The literature [19] simulates the
consumption of natural gas by randomly combining several influences including GDP,
population, natural gas imports and exports, and employment while predicting natural
gas consumption by selecting the most realistic equation from all simulated equations.
Literature [20] and literature [21] investigate the impact of the national economic
growth rate through the ARDL method [22-24]. Literature [25] and literature [26]
studied the impact of various factors on renewable energy consumption based on a
multivariate framework. Literature [27] conducted research using panel data from
nearly 35 years and obtained that, from a long-term perspective, natural gas energy
consumption has an impact on the GDP growth of GCC countries. Literature [28]
proposes a parameter estimation method that eliminates omitted variable bias due to
unobserved cross-sectional individual effects by differencing. In addition, the literature
[29] used the Granger causality test to obtain the relationship between natural gas
consumption and GDP. Literature [30] argues that marketization facilitates the
formation of a virtuous response between renewable energy and consumption,
guiding the consumption of clean renewable energy through price signals.
The aforementioned studies include the search for clean and renewable
alternatives to fossil energy through data models. The relationship between
consumption has been explored and the factors influencing have been studied. This
study further extends the study of natural gas energy to clean renewable energy
based on the previous study. We analyze the factors influencing the consumption of
clean renewable energy in the context of ecological environmental protection and air
pollution prevention. We also combine the panel data of 30 Chinese provinces to build
a mathematical model and estimate the model using differential GMM, systematic
GMM, and bias-corrected LSDV methods for the national and regional levels,
respectively. Based on this, the Tobit regression method is applied to validate the
results. Based on the data model, the influencing factors of energy consumption were
analyzed, to provide some theoretical references for ecological environmental
protection and air pollution control.
2. RESEARCH BASIS
2.1. CURRENT SITUATION OF CLEAN RENEWABLE
ENERGY CONSUMPTION
To examine the regional differences in energy consumption, the specifics of each
region in 2020 were investigated and plotted in Figure 1.
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
Figure 1. Clean Renewable energy consumption by Provinces in 2020
From Figure 1, we can see that the top three clean renewable energy consumption
in 2020 are Shandong Province, Guangdong Province, and Hebei Province
respectively. The highest consumption of clean renewable energy is over 40,000
million tons [31] in Shandong province, while the lowest consumption is only about 20
million tons of standard coal in Hainan, so we can easily see through the above chart
that there are big differences in the consumption data of clean renewable energy in
different regions at the same point of time.
2.2. MODEL PARAMETER SETTING
Panel data is adopted by a wide range of scholars for its advantages of large data
size. Economic theory suggests that the individual's past state determines the current
behavioral state due to inertia, so the lagged values can be included in the panel
model, and this type of data is dynamic.
The following dynamic panel parameters are considered:
(1)
First-order differencing to eliminate individual effects .
(2)
yit =α+ρyi,t−1+xit β+Ziδ+μi+εit,t= 2,3,⋯,T
μi
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
335
However, the DIF-GMM method has some shortcomings, such as the elimination of
non-observed section individual effects and other variables that do not change with
time when differencing. Also, its estimator is often not efficient (minimum variance).
Blundell and Bond combined the difference GMM with the level GMM to perform
GMM estimation of the difference and level equations as a system of equations, called
System GMM (SGS).
The methods mentioned above are more suitable for short dynamic panels.
Because while instrumental variable or GMM-based estimators are consistent
estimators, they may be more heavily biased for smaller and larger long panels. After
Monte Carlo simulations, the results show that the LSDV method is significantly better
than the differential GMM or the systematic GMM for smaller long panels. The basic
idea of the LSDV method is to first estimate the dynamic panel model using the LSDV
method, and the estimated coefficient is . Secondly, the bias of the LSDV method is
estimated as Bias; finally, this bias is subtracted from the estimated LSDV coefficient
to obtain a bias-corrected consistent estimate.
In this paper, the Dynamic Panel Model (DPM) with a first-order lag is considered
because the consumption target is expressed using the previous year's renewable
energy generation, so it contains first-order lagged data of the explanatory variables
[32]. Since the national and regional panels are studied separately in this paper, the
bias-corrected LSDV method is used considering the existence of bias in the
differential GMM and the systematic GMM.
The variables are selected according to the validity of the data, and the model is as
follows:
(3)
Where denotes renewable energy generation in year of region ; is a
random disturbance term.
3. DATA, VARIABLES, AND DESCRIPTIVE STATISTICS
The factors influencing the production of renewable energy generation include the
green financial development index (Gfi), government intervention (Gov), openness to
the outside world (Trade), R&D investment intensity (RD), tertiary industry share (TI),
energy consumption structure (ES), urbanization rate (Urban), and Internet
penetration rate (Ipr).
To make a more accurate and comprehensive measurement, this paper quantifies
the green financial development level by constructing a more reasonable index. The
composition of the index system is shown in Table 1. The degree of government
intervention is expressed as the ratio of general budget expenditures to GDP. The
intensity of R&D investment by provinces and municipalities is used to express the
indicator of technological progress. The development level of the tertiary industry can
well represent the development trend. The level of urbanization is represented by the
β
R
Eit =α+β1×R Ei,t−1+
K
∑
k=1
γk×Xkit +
M
∑
m=1
δm×Ymit +
N
∑
n=1
ωn×Znit +
S
∑
s=1
ρs×Vsit +εi
t
REit
t
i
Vsit
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
336
However, the DIF-GMM method has some shortcomings, such as the elimination of
non-observed section individual effects and other variables that do not change with
time when differencing. Also, its estimator is often not efficient (minimum variance).
Blundell and Bond combined the difference GMM with the level GMM to perform
GMM estimation of the difference and level equations as a system of equations, called
System GMM (SGS).
The methods mentioned above are more suitable for short dynamic panels.
Because while instrumental variable or GMM-based estimators are consistent
estimators, they may be more heavily biased for smaller and larger long panels. After
Monte Carlo simulations, the results show that the LSDV method is significantly better
than the differential GMM or the systematic GMM for smaller long panels. The basic
idea of the LSDV method is to first estimate the dynamic panel model using the LSDV
method, and the estimated coefficient is . Secondly, the bias of the LSDV method is
estimated as Bias; finally, this bias is subtracted from the estimated LSDV coefficient
to obtain a bias-corrected consistent estimate.
In this paper, the Dynamic Panel Model (DPM) with a first-order lag is considered
because the consumption target is expressed using the previous year's renewable
energy generation, so it contains first-order lagged data of the explanatory variables
[32]. Since the national and regional panels are studied separately in this paper, the
bias-corrected LSDV method is used considering the existence of bias in the
differential GMM and the systematic GMM.
The variables are selected according to the validity of the data, and the model is as
follows:
(3)
Where denotes renewable energy generation in year of region ; is a
random disturbance term.
3. DATA, VARIABLES, AND DESCRIPTIVE STATISTICS
The factors influencing the production of renewable energy generation include the
green financial development index (Gfi), government intervention (Gov), openness to
the outside world (Trade), R&D investment intensity (RD), tertiary industry share (TI),
energy consumption structure (ES), urbanization rate (Urban), and Internet
penetration rate (Ipr).
To make a more accurate and comprehensive measurement, this paper quantifies
the green financial development level by constructing a more reasonable index. The
composition of the index system is shown in Table 1. The degree of government
intervention is expressed as the ratio of general budget expenditures to GDP. The
intensity of R&D investment by provinces and municipalities is used to express the
indicator of technological progress. The development level of the tertiary industry can
well represent the development trend. The level of urbanization is represented by the
β
R Eit =α+β1×R Ei,t−1+
K
∑
k=1
γk×Xkit +
M
∑
m=1
δm×Ymit +
N
∑
n=1
ωn×Znit +
S
∑
s=1
ρs×Vsit +εit
REit
t
i
Vsit
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
urbanization rate of each province and city. The indicator of Internet penetration rate is
used to measure the degree of Internet development.
Each influencing factor is quantified on the premise that the possible influencing
factors are clarified. The explanatory variables used for the specific quantified
influencing factors are shown in Table 1.
Table 1. Description of explanatory variables
Figure 2. Descriptive statistics of the explanatory variables
Variable Name Variable Definition
Green Finance Index GIF The entropy method is calculated
from the
exponent
Government Intervention Gov General budget expenditure of government
finance
as a percentage of GDP
Degree Of Openness Trade
The ratio of total imports and exports to GDP
R&D investment intensity RD Proportion of R&D investment
in GDP by province
and city
Industrial Structure TI The ratio of the output value of
the tertiary industry
to GDP
Energy Consumption Structure ES Coal consumption as a percentage
of total energy
consumption
Urbanization Rate Urban The proportion of urban population
to total
population
Internet Penetration Ipr
Proportion of Internet users and population
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
337
The results of the descriptive statistics of the explanatory variables for the whole
country show that the maximum value is 1.15 and the minimum value is 0.17, which
proves that the selected observations have a wide range. The selected observations
are somewhat representative. The mean value of the explanatory variables for the
nation as well as for each province and city is 0.47, and since the mean value is
susceptible to extreme values, we use the standard deviation to indicate the degree of
aggregation of the data. From the above figure, it can be seen that all the interpretive
variables for the country as well as for each province and city are less than 0.18,
which proves that the data are more aggregated and less volatile4
4. RESULTS AND ANALYSIS
First of all, to avoid the problem of biased regression results brought about by the
problem of multicollinearity [33], the corresponding test must first be performed on the
collated sample data. The most direct and effective method is to use the variance
inflation factor (VIF) method for the test [34]. The following information can be found:
none of the selected explanatory variables has a VIF above 10 and their mean does
not exceed 5. Therefore, Table 2 indicates that there is no serious multicollinearity
among the eight explanatory variables we have selected and a panel Tobit regression
can be performed.
Table 2. Multicollinearity test for each explanatory variable
In this study, the quantitative value of clean renewable energy is used as the
dependent variable. The independent variables are eight variables: green financial
development index (Gfi), government intervention (Gov), openness to the outside
world (Trade), R&D investment intensity (RD), tertiary industry share (TI), energy
consumption structure (ES), urbanization rate (Urban), and Internet penetration rate
(Ipr). In this regard, we constructed panel data and used Tobit regression models to
conduct sub-sample regressions for the whole country and the eight economic
regions, and obtained two regression analysis results of positive and negative effects,
respectively, as shown in Figure 3 and Figure 4.
Variable VIF 1/VIF
Gif 9.15 0.1355
Urban 8.65 0.1438
RD 7.92 0.1578
TI 5.30 0.2496
Ipr 5.04 0.2654
Trade 4.72 0.2878
ES 3.74 0.3919
Gov 2.02 0.5350
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
338
The results of the descriptive statistics of the explanatory variables for the whole
country show that the maximum value is 1.15 and the minimum value is 0.17, which
proves that the selected observations have a wide range. The selected observations
are somewhat representative. The mean value of the explanatory variables for the
nation as well as for each province and city is 0.47, and since the mean value is
susceptible to extreme values, we use the standard deviation to indicate the degree of
aggregation of the data. From the above figure, it can be seen that all the interpretive
variables for the country as well as for each province and city are less than 0.18,
which proves that the data are more aggregated and less volatile4
4. RESULTS AND ANALYSIS
First of all, to avoid the problem of biased regression results brought about by the
problem of multicollinearity [33], the corresponding test must first be performed on the
collated sample data. The most direct and effective method is to use the variance
inflation factor (VIF) method for the test [34]. The following information can be found:
none of the selected explanatory variables has a VIF above 10 and their mean does
not exceed 5. Therefore, Table 2 indicates that there is no serious multicollinearity
among the eight explanatory variables we have selected and a panel Tobit regression
can be performed.
Table 2. Multicollinearity test for each explanatory variable
In this study, the quantitative value of clean renewable energy is used as the
dependent variable. The independent variables are eight variables: green financial
development index (Gfi), government intervention (Gov), openness to the outside
world (Trade), R&D investment intensity (RD), tertiary industry share (TI), energy
consumption structure (ES), urbanization rate (Urban), and Internet penetration rate
(Ipr). In this regard, we constructed panel data and used Tobit regression models to
conduct sub-sample regressions for the whole country and the eight economic
regions, and obtained two regression analysis results of positive and negative effects,
respectively, as shown in Figure 3 and Figure 4.
Variable
VIF
1/VIF
Gif
9.15
0.1355
Urban
8.65
0.1438
RD
7.92
0.1578
TI
5.30
0.2496
Ipr
5.04
0.2654
Trade
4.72
0.2878
ES
3.74
0.3919
Gov
2.02
0.5350
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
From the Tobit regression results, all the explanatory variables are significant
except for the urbanization rate, the degree of openness to the outside world, and the
share of tertiary industry, which are not significant in the model. The specific analysis
is as follows:
Figure 3. Results of regression analysis of the positive influence of inter-provincial clean
renewable energy in China
Figure 3 shows that the level of green financial development has a significant
contribution to clean renewable energy efficiency. Overall, for every 1% increase in
the green finance index, the consumption of clean renewable energy increases by
0.882. The level of green finance development provides green credit specifically to the
relevant transition enterprises through banks. Financial instruments such as green
bonds offered in the market by the government and other market players can give
financing support to environmental protection companies and green transformation
companies, and policy incentives to motivate them to improve their production
processes. Reduce factor inputs, and thus increase the consumption of clean and
renewable energy. At the same time, the relevant policies introduced together with
green finance have formed financing constraints for the "three high" enterprises,
restraining the continuation of the crude development approach and forcing the "three
high" enterprises to transform and upgrade their energy structures to increase the
consumption of clean renewable energy. By economic zone, the most obvious role of
green financial development level for the promotion of clean renewable energy
consumption is the Yellow River Middle River Economic Zone. Many of the provinces
in the Middle Yellow River Economic Zone have an industrial system that was
previously dominated by secondary industries and is now undergoing an important
journey of industrialization and transformation. A deep policy dividend has been given
to these regions through credits and bonds under the green finance policy. The region
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
339
where the effect is not obvious is the Northwest Economic Zone. Possible reasons for
this are, on the one hand, that there is a lag in the use of more practical green
financial instruments tools because the pilot scope of green finance has not yet been
fully spread. On the other hand, the northwestern region has not been very effective in
bringing the advantages of the financial system into play at this stage due to the
limitations of its own economic and financial development level [35].
R&D investment intensity has an impact on consumption. At the national level,
each unit increase in R&D investment intensity increases clean renewable energy
consumption by 6.425 units. The coefficient value is the largest of all explanatory
variables and has the strongest effect, indicating that technological progress is
particularly critical to the increase in clean renewable energy consumption. This
proves that technological progress is the most powerful and sustainable means to
increase the consumption of clean renewable energy. Further, it can be found that the
southern economic zone has the most significant release of dividends from
technological progress due to the strong technological strength of the provinces and
the high degree of integration between industry, academia, and research. RD has the
greatest impact on consumption, while the eastern, northern and middle, and lower
reaches of the Yangtze River economic zones are not significant [36].
The increase in Internet penetration significantly contributes to the increase in clean
renewable energy consumption, in a homogeneous relationship. For every unit
increase in Internet penetration, clean renewable energy consumption can increase
by 0.34 units. This has great policy implications in the context of the strong Internet
development and the emergence of the digital economy. The greatest impact has
been in the Northern Economic Zone, where the Internet has led to the construction of
an online platform on the one hand, which has greatly reduced the intermediate costs
for the energy companies involved. On the other hand, the penetration rate has
improved the quality of the workforce and optimized the consumption of clean and
renewable energy. In contrast, the southwestern and northwestern economic regions
are more in need of a better role in the improvement of energy consumption through
the upgrading of the Internet [37].
The sign of the regression results of the degree of external openness and the share
of tertiary industry is consistent with the theoretical analysis. The deepening of
external openness can improve the learning effect of domestic enterprises through
foreign technology spillover and trade connection, forming a virtuous cycle of "foreign
spillover, domestic learning, and exchange and progress". Specifically, the positive
effects are evident in the northern, southwestern, and northwestern economic regions.
The increase in the share of the tertiary sector indicates that the industrial structure is
in the process of upgrading, which is disadvantageous compared to the secondary
sector with high energy input. The tertiary sector is more conducive to increasing
consumption because it is dominated by the service sector, which has a lower energy
input and a larger share of clean energy use. In particular, the effect of the eastern
economic zone is the most obvious. The eastern economic zone has attracted a large
number of high-quality labor forces through its advantages of high urbanization and
perfect basic public services, which further optimized its industrial structure. The
tertiary industry is more developed and has a higher proportion.
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
340
where the effect is not obvious is the Northwest Economic Zone. Possible reasons for
this are, on the one hand, that there is a lag in the use of more practical green
financial instruments tools because the pilot scope of green finance has not yet been
fully spread. On the other hand, the northwestern region has not been very effective in
bringing the advantages of the financial system into play at this stage due to the
limitations of its own economic and financial development level [35].
R&D investment intensity has an impact on consumption. At the national level,
each unit increase in R&D investment intensity increases clean renewable energy
consumption by 6.425 units. The coefficient value is the largest of all explanatory
variables and has the strongest effect, indicating that technological progress is
particularly critical to the increase in clean renewable energy consumption. This
proves that technological progress is the most powerful and sustainable means to
increase the consumption of clean renewable energy. Further, it can be found that the
southern economic zone has the most significant release of dividends from
technological progress due to the strong technological strength of the provinces and
the high degree of integration between industry, academia, and research. RD has the
greatest impact on consumption, while the eastern, northern and middle, and lower
reaches of the Yangtze River economic zones are not significant [36].
The increase in Internet penetration significantly contributes to the increase in clean
renewable energy consumption, in a homogeneous relationship. For every unit
increase in Internet penetration, clean renewable energy consumption can increase
by 0.34 units. This has great policy implications in the context of the strong Internet
development and the emergence of the digital economy. The greatest impact has
been in the Northern Economic Zone, where the Internet has led to the construction of
an online platform on the one hand, which has greatly reduced the intermediate costs
for the energy companies involved. On the other hand, the penetration rate has
improved the quality of the workforce and optimized the consumption of clean and
renewable energy. In contrast, the southwestern and northwestern economic regions
are more in need of a better role in the improvement of energy consumption through
the upgrading of the Internet [37].
The sign of the regression results of the degree of external openness and the share
of tertiary industry is consistent with the theoretical analysis. The deepening of
external openness can improve the learning effect of domestic enterprises through
foreign technology spillover and trade connection, forming a virtuous cycle of "foreign
spillover, domestic learning, and exchange and progress". Specifically, the positive
effects are evident in the northern, southwestern, and northwestern economic regions.
The increase in the share of the tertiary sector indicates that the industrial structure is
in the process of upgrading, which is disadvantageous compared to the secondary
sector with high energy input. The tertiary sector is more conducive to increasing
consumption because it is dominated by the service sector, which has a lower energy
input and a larger share of clean energy use. In particular, the effect of the eastern
economic zone is the most obvious. The eastern economic zone has attracted a large
number of high-quality labor forces through its advantages of high urbanization and
perfect basic public services, which further optimized its industrial structure. The
tertiary industry is more developed and has a higher proportion.
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
Figure 4. Results of regression analysis of negative influencing factors of inter-provincial
clean renewable energy in China
As seen in Figure 4, the degree of government intervention has a dampening effect
on clean renewable energy consumption in the Tobit regression results. Overall, each
unit increase in the degree of government intervention. Instead of increasing, the
consumption of clean renewable energy decreases by 0.173 units, which has a
significant counter effect. This reflects that the government's intervention in energy-
related aspects will, to a certain extent, restrict the increase of clean renewable
energy consumption and fail to give full play to the regulation and allocation function
of the market for the relevant factors. Further subsampling reveals that the overall
governance capacity of the government in the northwest economic zone is limited by
historical development factors compared to the developed economic zones.
Therefore, government intervention in energy is not conducive to the allocation of
factors. The improved governance capacity of local governments promotes market
circulation and rational allocation of energy and other factors. The results demonstrate
that government intervention is effective in not hindering the progress of energy
consumption.
Each unit increase will lead to a 0.0876 unit decrease. And the absolute value of
the coefficient impact is high compared to the rest of the explanatory variables,
indicating that the energy consumption structure with the share of coal can
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
341
significantly and negatively affect the energy consumption enhancement. The results
of the sub-sample regression also corroborate the national regression results.
The increase in the urbanization rate inhibits the increase in the consumption of
clean renewable energy. Specifically, each unit increase in the urbanization rate
decreases clean renewable energy consumption by 0.107 units. A possible
explanation for the insignificant regression is that the income effect created by the
urbanization process leads to an increase in the demand for residential energy
consumption. At the same time, the negative externalities of urban pollutant emissions
inhibit the improvement of energy consumption. The significant negative regressions
in the northern, northeastern, and middle reaches of the Yellow River and
northwestern economic zones also support the analysis of the previous theoretical
mechanism.
5. DISCUSSION
Given the above research deficiencies, it is necessary to investigate, understand
and grasp the situation, characteristics, and patterns of Chinese household energy
consumption from multiple levels and in-depth in the future research process. In
addition, the structure is changing dramatically with the intensive development of
clean and renewable energy sources. Therefore, further research on the factors
influencing the consumption of clean renewable energy is of great significance and
has important reference value for China's future inquiry on the formulation of related
policies.
6. CONCLUSION
This paper adopts the input-output-related data of 30 provinces and cities
nationwide from 2010 to 2020 to establish a mathematical model. The different
influencing factors of clean renewable energy consumption are sorted out. The trends
and characteristics of spatial differences in clean renewable energy consumption in
China are revealed. The spatial effects of the influencing factors are quantitatively
analyzed, which can provide some reference for energy policy formulation and energy
planning in China. It provides some references for developing energy structure
optimization strategies and energy saving and emission reduction measures with
regional characteristics.
1.
In this paper, differential GMM, systematic GMM, and bias-corrected LSDV
methods are used to build the national and regional models. The highest
consumption of clean renewable energy is over 40,000 million tons, while the
lowest consumption is only about 20 million tons. The evolution and spatial
differences in its development are fully recognized and understood.
2.
The absolute values of the results analyzed by the Tobit regression method,
except for the urbanization rate, the degree of openness to the outside world,
and the share of tertiary industry, which are less than 0.15, all the explanatory
variables are greater than 0.15. This result fully demonstrates that all the
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
342
significantly and negatively affect the energy consumption enhancement. The results
of the sub-sample regression also corroborate the national regression results.
The increase in the urbanization rate inhibits the increase in the consumption of
clean renewable energy. Specifically, each unit increase in the urbanization rate
decreases clean renewable energy consumption by 0.107 units. A possible
explanation for the insignificant regression is that the income effect created by the
urbanization process leads to an increase in the demand for residential energy
consumption. At the same time, the negative externalities of urban pollutant emissions
inhibit the improvement of energy consumption. The significant negative regressions
in the northern, northeastern, and middle reaches of the Yellow River and
northwestern economic zones also support the analysis of the previous theoretical
mechanism.
5. DISCUSSION
Given the above research deficiencies, it is necessary to investigate, understand
and grasp the situation, characteristics, and patterns of Chinese household energy
consumption from multiple levels and in-depth in the future research process. In
addition, the structure is changing dramatically with the intensive development of
clean and renewable energy sources. Therefore, further research on the factors
influencing the consumption of clean renewable energy is of great significance and
has important reference value for China's future inquiry on the formulation of related
policies.
6. CONCLUSION
This paper adopts the input-output-related data of 30 provinces and cities
nationwide from 2010 to 2020 to establish a mathematical model. The different
influencing factors of clean renewable energy consumption are sorted out. The trends
and characteristics of spatial differences in clean renewable energy consumption in
China are revealed. The spatial effects of the influencing factors are quantitatively
analyzed, which can provide some reference for energy policy formulation and energy
planning in China. It provides some references for developing energy structure
optimization strategies and energy saving and emission reduction measures with
regional characteristics.
1. In this paper, differential GMM, systematic GMM, and bias-corrected LSDV
methods are used to build the national and regional models. The highest
consumption of clean renewable energy is over 40,000 million tons, while the
lowest consumption is only about 20 million tons. The evolution and spatial
differences in its development are fully recognized and understood.
2. The absolute values of the results analyzed by the Tobit regression method,
except for the urbanization rate, the degree of openness to the outside world,
and the share of tertiary industry, which are less than 0.15, all the explanatory
variables are greater than 0.15. This result fully demonstrates that all the
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
relevant variables have significant effects on the results, except for the
insignificant effects of urbanization rate, degree of openness to the outside
world, and tertiary industry.
3.
The results of the regression analysis are obtained by the Tobit regression
method, such as every 1% increase in the green finance index increases the
consumption of clean renewable energy by 0.796. The level of green financial
development, the degree of Internet development, and technological advances
are analyzed to have a significant positive effect on the consumption of clean
renewable energy. The results that the industrial structure represented by the
share of secondary industry, the degree of openness to the outside world, and
the level of urbanization do not have a significant effect on the consumption of
clean renewable energy are also derived.
REFERENCES
(1) Meng, F., Guo, J., Guo, Z., et al. (2021). Urban ecological transition: The practice
of ecological civilization construction in China. Science of The Total Environment,
755(Pt 2), 142633.
(2) Kuzma, L., Kurasz, A., Dabrowski, E. J., et al. (2021). Association between air
pollution and case-specific mortality in the north-eastern part of Poland. Case
crossover study with 4,500,000 person-years of follow-up. European Heart
Journal, Supplement_1.
(3) Rodrigues, V., Gama, C., Ascenso, A., et al. (2021). Assessing air pollution in
European cities to support a citizen-centered approach to air quality
management. Science of The Total Environment, 799(11), 149311.
(4) Du, X. , & Huang, Z. . (2017). Ecological and environmental effects of land use
change in rapid urbanization: the case of hangzhou, china. Ecological Indicators,
81(oct.), 243-251.
(5) Malaspina, P. , Modenesi, P. , & Giordani, P. . (2018). Physiological response of
two varieties of the lichen pseudevernia furfuracea to atmospheric pollution.
Ecological Indicators, 86, 27- 34.
(6)
Thapar, S., Sharma, S. , & Verma, A. . (2017). Local community as
shareholders in clean energy projects: innovative strategy for accelerating
renewable energy deployment in india. Renewable Energy, 101(FEB.), 873-885.
(7) Zhang, Y. J. , Bian, X. J. , Tan, W., & Song, J.. (2017). The indirect energy con-
sumption and co2 emission caused by household consumption in china: an
analysis based on the input-output method. Journal of Cleaner Production, 163
(oct.1), 69-83.
(8) Duda, A. M. . (2017). Leadership and political will for groundwater governance:
indispensable for meeting the new sustainable development goals (sdgs).
Brazilian Journal of Microbiolo gy.
(9) Jing-Li, Fan, Yue-Jun, Zhang, Bing, & Wang. (2017). The impact of urbanization
on residential energy consumption in china: an aggregated and disaggregated
analysis. Renewable & Sustainable Energy Reviews.
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
343
(10)
Sagastume, Gutierrez, Alexis, Cabello, Eras, & Juan, et al. (2018). The current
potential of low-carbon economy and biomass-based electricity in cuba. the case
of sugarcane, energy cane and marabu (dichrostachys cinerea) as biomass
sources. Journal o f Cleaner Production.
(11)
Harini, G., Balasurya, S., & Khan, S.S.. (2022). Recent advances on
gadolinium-based nano-photocatalysts for environmental remediation and clean
energy production: properties, fabrication, defect engineering and toxicity.
Journal of Cleaner Production, 345, 131139-.
(12)
Ma, Q. , Zhao, Y. , Ji, C. , Zhang, Y. , & Ming, B.. (2021). Electricity curtailment
cost coupled to operation model facilitates clean energy accommodation in grid-
connected system. Energi es, 14.
(13)
Leenaers, A., Renterghem, W. V., & Van D. (2016). High burn-up structure of
U(Mo) dispersion fuel. Journal of Nuclear Materials, 476, 218-230.
(14)
Wrman, A., Uvo, C. B., Brandimarte, L., et al. (2020). Virtual energy storage gain
resulting from the spatio-temporal coordination of hydropower over Europe.
Applied Energy, 272, 115249.
(15)
Li, Z., Tian, Q., Xu, J., et al. (2021). Easily Fabricated Low-Energy Consumption
Joule-Heated Superhydrophobic Foam for Fast Cleanup of Viscous Crude Oil
Spills. ACS Applied Materials And Interfaces, 13(43), 51652-51660.
(16)
Liu, Y., Sadiq, F., Ali, W., et al. (2022). Does tourism development, energy
consumption, trade openness and economic growth matter for ecological
footprint: Testing the Environmental Kuznets Curve and pollution haven
hypothesis for Pakistan. Energy, 245.
(17)
Shan, S., Cai, X., Li, K., et al. (2021). Spectral energy characteristics of radiation
in oxy-coal combustion for energy utilization. Fuel, 289(3), 119917.
(18)
Wang, Y., Ji, Q., Shi, X., et al. (2020). Regional renewable energy development
in China: A multidimensional assessment. Renewable and Sustainable Energy
Reviews, 124.
(19)
Anelkovi, A. S. , & Bajatovi, D. . (2020). Integration of weather forecast and
artifi-cial intelligence for a short-term city-scale natural gas consumption
prediction. Journal of Cleaner Prod uction, 266(2823), 122096.
(20)
Ergun Uzlu, Murat Kankal, Adem Akpmar, Tayfun Dede. (2017). Estimates of
energy consumption in Turkey using neural networks with the teaching-learning-
based optimization algorithm. Energy, 75.
(21)
Cem I,sik. (2018). Natural gas consumption and economic growth in Turkey: A
bound test approach. Energy Syst, 1.
(22)
Baz, K., Cheng, J., Xu, D., et al. (2021). Asymmetric impact of fossil fuel and
renewable energy consumption on economic growth: A nonlinear technique.
Energy, 2, 120357.
(23)
Ozturk, I., Al-Mulali, U. (2015). Natural gas consumption and economic growth
nexus: Panel data analysis for GCC countries. Renewable and Sustainable
Energy Reviews, 51.
(24)
Mujtaba, A., Jena, P. K., Bekun, F. V., et al. (2022). Symmetric and asymmetric
impact of economic growth, capital formation, renewable and non-renewable
energy consumption on environment in OECD countries. Renewable and
Sustainable Energy Reviews, 160, 112300-.
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
344
(10) Sagastume, Gutierrez, Alexis, Cabello, Eras, & Juan, et al. (2018). The current
potential of low-carbon economy and biomass-based electricity in cuba. the case
of sugarcane, energy cane and marabu (dichrostachys cinerea) as biomass
sources. Journal o f Cleaner Production.
(11) Harini, G., Balasurya, S., & Khan, S.S.. (2022). Recent advances on
gadolinium-based nano-photocatalysts for environmental remediation and clean
energy production: properties, fabrication, defect engineering and toxicity.
Journal of Cleaner Production, 345, 131139-.
(12) Ma, Q. , Zhao, Y. , Ji, C. , Zhang, Y. , & Ming, B.. (2021). Electricity curtailment
cost coupled to operation model facilitates clean energy accommodation in grid-
connected system. Energi es, 14.
(13) Leenaers, A., Renterghem, W. V., & Van D. (2016). High burn-up structure of
U(Mo) dispersion fuel. Journal of Nuclear Materials, 476, 218-230.
(14) Wrman, A., Uvo, C. B., Brandimarte, L., et al. (2020). Virtual energy storage gain
resulting from the spatio-temporal coordination of hydropower over Europe.
Applied Energy, 272, 115249.
(15) Li, Z., Tian, Q., Xu, J., et al. (2021). Easily Fabricated Low-Energy Consumption
Joule-Heated Superhydrophobic Foam for Fast Cleanup of Viscous Crude Oil
Spills. ACS Applied Materials And Interfaces, 13(43), 51652-51660.
(16) Liu, Y., Sadiq, F., Ali, W., et al. (2022). Does tourism development, energy
consumption, trade openness and economic growth matter for ecological
footprint: Testing the Environmental Kuznets Curve and pollution haven
hypothesis for Pakistan. Energy, 245.
(17) Shan, S., Cai, X., Li, K., et al. (2021). Spectral energy characteristics of radiation
in oxy-coal combustion for energy utilization. Fuel, 289(3), 119917.
(18) Wang, Y., Ji, Q., Shi, X., et al. (2020). Regional renewable energy development
in China: A multidimensional assessment. Renewable and Sustainable Energy
Reviews, 124.
(19) Anelkovi, A. S. , & Bajatovi, D. . (2020). Integration of weather forecast and
artifi-cial intelligence for a short-term city-scale natural gas consumption
prediction. Journal of Cleaner Prod uction, 266(2823), 122096.
(20) Ergun Uzlu, Murat Kankal, Adem Akpmar, Tayfun Dede. (2017). Estimates of
energy consumption in Turkey using neural networks with the teaching-learning-
based optimization algorithm. Energy, 75.
(21) Cem I,sik. (2018). Natural gas consumption and economic growth in Turkey: A
bound test approach. Energy Syst, 1.
(22) Baz, K., Cheng, J., Xu, D., et al. (2021). Asymmetric impact of fossil fuel and
renewable energy consumption on economic growth: A nonlinear technique.
Energy, 2, 120357.
(23) Ozturk, I., Al-Mulali, U. (2015). Natural gas consumption and economic growth
nexus: Panel data analysis for GCC countries. Renewable and Sustainable
Energy Reviews, 51.
(24) Mujtaba, A., Jena, P. K., Bekun, F. V., et al. (2022). Symmetric and asymmetric
impact of economic growth, capital formation, renewable and non-renewable
energy consumption on environment in OECD countries. Renewable and
Sustainable Energy Reviews, 160, 112300-.
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
(25)
Yang, M., Wang, E. Z., Hou, Y. (2021). The relationship between manufacturing
growth and CO2 emissions: Does renewable energy consumption matter?
Energy, 2, 121032.
(26)
Namahoro, J. P., Nzabanita, J., Wu, Q. (2021). The impact of total and
renewable energy consumption on economic growth in lower and middle- and
upper-middle-income groups: Evidence from CS-DL and CCEMG analysis.
Energy, 237.
(27)
Niu, C., Tan, K., Jia, X., et al. (2021). Deep learning based regression for
optically inactive inland water quality parameter estimation using airborne
hyperspectral imagery. Environmental Pollution, 117534.
(28)
Mo, X. B., Zhang, Y. H., Lei, S. F. (2020). Integrative analysis identifies potential
causal methylation-mRNA regulation chains for rheumatoid arthritis. Molecular
Immunology.
(29)
Sun, Z., Li, X., Cui, G., et al. (2021). A Fast Approach for Detection and
Parameter Estimation of Maneuvering Target With Complex Motions in Coherent
Radar System. IEEE Transactions on Vehicular Technology, PP(99), 1-1.
(30)
Estelle, R., Adrien, W., Salamanca-Giron, R. F., et al. (2021). Functional
Segregation within the Dorsal Frontoparietal Network: A Multimodal Dynamic
Causal Modeling Study. Cerebral Cortex.
(31)
Roldan-Fernandez, J. M., Burgos-Payan, M., Riquelme-Santos, J. M., et al.
(2016). Renewable Generation Versus Demand-side Management. A
Comparison for the Spanish Market. Energy Policy, 96(9), 458-470.
(32)
Orndahl, C. M., Perera, R. A., Riddle, D. L. (2020). Associations Between
Physical Therapy Visits and Pain and Physical Function After Knee Arthroplasty:
A Cross-Lagged Panel Analysis of People Who Catastrophize About Pain Prior
to Surgery. Physical Therapy, 1.
(33)
Fan, J., Wang, J., Liu, M., et al. (2022). Scenario simulations of China's natural
gas consumption under the dual-carbon target. Energy, 252.
(34)
Li, R., Zhang, H., Gao, S., et al. (2021). An improved extreme learning machine
algorithm for transient electromagnetic nonlinear inversion. Computers &
Geosciences, 156(20), 104877.
(35)
Richards, K. C., Vallabhaneni, V., Moelter, S., et al. (2020). 0861 Age, Race, And
Continuous Positive Airway Pressure (CPAP) Confidence Score At 1-week
Predict 3-month CPAP Adherence In Older Adults With Amnestic Mild Cognitive
Impairment And Moderate To Severe Obstructive Sleep Apnea. SLEEP,
Supplement_1.
(36)
Mei, Dong. (2022). Reconstruction of multimodal aesthetic critical discourse
analysis framework. Applied Mathematics and Nonlinear Sciences. doi:10.2478/
AMNS.2021.2.00165.
(37)
Wang, Q., Li, S., Pisarenko, Z. (2020). Heterogeneous effects of energy
efficiency, oil price, environmental pressure, R&D investment, and policy on
renewable energy -- evidence from the G20 countries. Energy, 209.
(38)
Ghaffari, A., Askarzadeh, A. (2020). Design optimization of a hybrid system
subject to reliability level and renewable energy penetration. Energy, 193,
116754.
https://doi.org/10.17993/3ctecno.2023.v12n2e44.331-345
3C Tecnología. Glosas de innovación aplicadas a la pyme. ISSN: 2254-4143
Ed.44 | Iss.12 | N.2 April - June 2023
345