Renewable Energy 215 (2023) 118983

Contents lists available at ScienceDirect

Renewable Energy
journal homepage: www.elsevier.com/locate/renene

Nexus between green finance, renewable energy and carbon emission:

Empirical evidence from selected Asian economies
Lan Xu a, Yang Wu b, *
a
College of Wealth Management, Ningbo University of Finance & Economics, Ningbo City, Zhejiang Province, 315175, China
b
Shenzhen Lvhuan Renewable Resources Development Co., Ltd. Shenzheng City, Guangdong Province, 518125, China

A R T I C L E I N F O A B S T R A C T

Keywords: Using data from selected Asian countries from 1995 to 2018, this new article investigated the dynamic corre­
Non-fossil fuels lation among non-fossil fuel and fossil - based energy utilization, economic growth, and Gross domestic product
Non-renewable energy (GDP). For empirical research, the panel co-integration approach with structural breakdowns is employed. The
Gross domestic product growth
findings indicate that non-fossil fuel and fossil - based energy, economic growth, and economic expansion have a
Selected ASIAN countries
Green financial activities
long-term equilibrium connection. In both the short and long term, the panel causality studies reveal bidirec­
Panel model tional causation between industrial production and both non-renewable energy and renewable energy con­
sumption. The results from the panel error correction model reveals that the positive (negative) elasticities of
nonrenewable energy consumption are 0.96 (1.15), suggesting that a 1% rise (decrease) in nonrenewable energy
is expected to increase (reduce) 0.96% (1.15%) in carban dioxide emissions. These findings suggest that selected
Asian countries’ industrial production and overall economic growth are still energy-dependent. In addition, there
is two directional short run test causality among renewable energy consumption and non-renewable energy
consumption might help to create a more sustainable energy economy.

1. Introduction been fundamentally renovated and most efficient [3]. However, among
the many economic sectors, such as energy industry sector and tech­
Renewable energy efficiency is the main pillar of environmental and nology sectors are dominates economic activity in SELECTED ASIAN
energy policy for climate change to reduce its negative impacts. Energy countries, consuming the most energy and emitting a substantial
consumption increased from 188 quads in 1993 to 257 quads in 2012 in quantity of carbon dioxide. However, their conclusions are disparate,
these countries, despite average production growth remaining at 1.1% and economists are divided on the subject. The majority of past research
per year [1]. The majority of this energy comes from conventional or have been aggregated assessments that look at overall energy use,
traditional sources, such as oil sector, coal sector, and gas sector. Other pollutant emissions, and economic growth. The studies of [4], on
hand, renewable energy is predictable to the clean, cheap and fastest renewable energy consumption have only freshly been examined (see
growing energy global energy source, as well as reduction the corban Fig. 1).
dioxide emissions and other gases. Asian countries’ economies to show According to a current “GHGs released by the World Meteorological
an increasing growth in renewable energy can be qualified in part to Organization on December 1st, the clear trend of increasing Emission of
governments policies such as renewable energy and renewable energy greenhouse gases has become a long-term problem that will worsen over
collection standard. As a result, countries investment in renewable en­ time. All nations are looking at sustainable energy as a basic form of
ergy sources in Asian regions has USD$150 since 2017, and renewable energy to decrease carbon emissions and avoid harmful emissions [5] in
energy now accounts for around 25% of total energy supply in these the face of these issues. Various types of pollution have evolved as a
economies [2]. Because of the worldwide climate change condition heated policy discussion as a result of the current financial development
caused by the non-renewable consumptions, various high pollution trend. When it comes to the SDGs, the globe is seeing an urgent need for
economies have attempted to make structural changes to electricity a change in urban economic developments to reinstate the balance of
generation methods, primarily switching to renewable energy sources nature. When the industrial development pattern is reoriented, the
from fossil-fuel-based non-fossil-fuel-based sources, and this has lately corporate growth drivers will also have to change. This means that price

* Corresponding author.
E-mail addresses: hainshu@163.com (L. Xu), malanmi@163.com (Y. Wu).

https://doi.org/10.1016/j.renene.2023.118983
Received 5 June 2022; Received in revised form 24 June 2023; Accepted 30 June 2023
Available online 10 July 2023
0960-1481/© 2023 Published by Elsevier Ltd.
L. Xu and Y. Wu Renewable Energy 215 (2023) 118983

formation routes will have to change as well. According to a paper from obvious consequences of increasing REI size. It’s also possible to mini­
the Swiss Re Institutes that was presented during the 47th G7 Summit in mize carbon pollution by raising the REI scale, which promotes changes
Cornwall, economization was regarded as crucial. The G7 nations can in water organization and minimizes power intensity [10]. This illus­
lose 8.5% of their income if they don’t take action on climate change, trates that the REI scale, directly and indirectly, influences carbon
and climatic financing has been highlighted as one of the most important pollution. For one thing, a well-structured investment project may help
drives to combat this problem. As a result, the 47th G7 Summit, which improve the efficiency of buildings, expand the demand growth of real
took place at the same time as COP26, was significant in the context of social goods, and open up new capital opportunities for the future [11].
the Paris agreement. As a result, from the standpoint of sustainable REI structure influences REI size, which in turn determines Carbon
development, economic mobilization may be crucial for the G7 econo­ footprints via economy, capacity utilization, and concentration. There
mies. SDG development report 2021 shows that G7 nations are failing to has to be a thorough examination of the impact of REI on CO2 emissions.
meet the goals of climactic action, especially SDG 13 and this supports It’s possible to choose a few subjects worth exploring further in the
the claim [6]. The G7 nations’ incapacity to raise funds to reinforce their future study after looking at the published studies. There are just a few
growth strategy for combating global warming is also reflected in this studies on the impact of REI on CO2 emissions, which is a problem. With
report [7]. An SDG-focused legislative framework that concentrates on all of the money that is spent on sustainable power, this is a contradic­
economic creation could need to emerge in these situations. To begin tion. Another issue is that much of the available research on REI focuses
this investigation, the necessity for a policy readjustment represents the only on the REI scale, ignoring the impact of REI structure on CO2
initial stage in the process. emissions. Third, the majority of research exclusively looks at the direct
The G7 nations’ stock indexes need particular attention in this policy link connecting impacting variables and carbon dioxide pollution. It’s
discussion. Bioscience Driven Targeted organization [8] released a study reasonable to use many moderator models since the interaction amongst
just before G7’s 47th Summit, revealing that the finance sector of these variables is so complicated, with several secondary effects [12]. Another
nations’ economies is not in line with the Sustainable Development problem with large-scale statistical testing is that renewable sources
Goals. Heating routes of more than 4 ◦ Celsius may be found in these have emerged very late in development.
indexes as well. An examination of the industries represented by these This paper main contribution as: Use to evaluate the long term
indexes shows that fossil fuels mostly fuel them. Laissez-faire economic connection between dependent and independent variables, we utilize
mechanisms have forced these nations to follow an expansion that is not Common Correlated Effects Mean Group (CCEMG) estimator. We use a
ecologically friendly, which is critical for politicians. Since these na­ production function framework based on, which accounts for renewable
tions’ energy utilization patterns need to be changed as well, govern­ and non-renewable energy consumption as well as the conventional
mental action is necessary to bring about this change. The G7 Energy inputs of capital and labor. Following we additionally look for structural
Program, launched in 2014, is intended to realign energy policy toward breakdowns in the data and investigate the potential of cross-sectional
developing affordable, renewable energy. Environmental issues have a dependency (CSD). Both renewable and nonrenewable energy have a
complicated connection with REI from a conceptual point of view [9]. favorable influence on GDP and industrial production, according to the
Increased energy use and carbon pollution are two of the most empirical findings. We also discover that renewable energy may be

Fig. 1. Box normal plot of selected variables.

2
L. Xu and Y. Wu Renewable Energy 215 (2023) 118983

substituted for non-renewable energy. This study is the most about study between the price of oil and clean energy use, as shown by. According to,
on renewable energy consumption and financial growth to 2018, actual crude prices in South American nations have a long-term corre­
covering 10 Asian countries. This contradicts the findings of found lation with alternative energy consumption in the region. Furthermore,
unidirectional connection between GDP and renewable energy utiliza­ show that oil prices stimulate renewable power usage via innovation. To
tion. We also discover bidirectional causation between industrial pro­ put it another way, the rise in real fuel prices has increased renewable
duction and the use of non-renewable energy consumption and energy usage. In the panel dataset of 25 OECD nations, discovered a
renewable energy consumption will be utilized. Section 2 provides a negative correlation between energy consumption and price and mac­
literature review. The methodology and data are discussed in section 3, roeconomic sophistication.
followed by the empirical results in Section 4. Section 5 concluding re­ Some nations and areas, like those in Africa and Latin America,
marks and policy implications. received much attention in the early research, including Malaysia and
Iran. After then, researchers used a worldwide methodology to examine
2. Literature review the link between export growth and sustainable energy usage [23]. In
industrialized nations like the OECD, the connection between export
2.1. Green finance and economic development growth and sustainable energy usage has recently been considered [24].
Research data restrictions meant that these studies on OECD nations did
Studies examining the link connecting clean energy use and eco­ not incorporate the newest developments, such as the influence of the
nomic progress are becoming more prevalent [13]. This field of study implementation of the Climate Accord, in their analysis of OECD states’
has been spurred on by new considerations of the classic green energy economic performance. Electricity production in OECD nations signifi­
nexus, as opposed to the new drivers of sustainable energy use. Re­ cantly impacts the global transition to a low-carbon economy because of
searchers disagree on the precise area and extent of this correlation since the huge number of major companies and global trading powers that
they rely on multiple datasets, periods and methods. In one of the earlier make up Countries worldwide. OECD nations’ trade liberalization and
research [14],uses panel cointegration methods to examine the drivers sustainable energy use should be examined in this context.
of renewable energy in the G7 nations. According to the author’s When evaluating the link involving two and three-party parameters,
research, long-term renewable energy usage is mostly driven by per such as environmental and economic growth, the trade balance is often
capita income GDP and per-person CO2 Emissions. considered a control condition. Smaller studies have examined how
Although crude oil prices have a small but significant influence on export growth affects renewables power usage directly. Energy trade
this usage, they are not the primary factor. Per capita renewable energy models were often utilized in this research to examine this link. A lack of
usage is positively correlated with actual per national spending, ac­ complexity in existing models has resulted in a lack of precision and
cording to the following research by [15]. theory is supported by the dependability in the estimation of outcomes, which may be due to the
data. Put another way, and it asserts that economic development is adoption of basic models. Research has shown that various economic
essential to increase the amount of clean energy used. In Brazil, China, factors (e.g., semi-energy, financial deepening, foreign investment, and
India, Indonesia, the Philippines, and Turkey, researchers [16] exam­ demographic) may impact renewable energy usage. Trade openness and
ined the factors that influence the use of sustainable energy. The study green energy usage cannot be studied separately, at least not in theory.
illustrates that in Brazil, China, Africa, and Jakarta, green energy usage Because of this, more factors must be [25] included to help decrease
is impacted by affluence and contaminant pollution. In the big scheme of errors caused by missing variables from the standard technologies. It’s
things, the desire for solar power and heat and Turkish is solely deter­ important to achieve strong market unification for RE so that they join
mined by per capita. There is a long-term effect of economic develop­ based on their total expenditure, rather than support systems that offer
ment on electricity usage, according to Refs. [17,18]. grants for RE to maximize their energy production regardless of the
value they provide the system. RE utilization is hindered by resource
2.2. CO2 emissions and renewable energy consumption support systems, which are broadly accepted [26]. For a powerful sup­
port mechanism, the market value of production of a reference plant
Researchers Rosales and Values emphasize conducted a thorough with comparable features is used to calculate payouts for a RE producer;
analysis in which they found that CO2, energy demand, per capita a lower selling price means larger compensation payments. Because the
Economic, reauthorization of the Stockholm Convention, and high remuneration funds are no longer tied to the VRE producer’s output,
power bill rates in business are the primary factors of sustainable energy only market mechanisms may determine operational choices.
economic expansion in 38 countries. Per capita, carbon dioxide emis­ Capacity-based support schemes like these will also encourage RE plants
sions are shown to have a favorable effect on the utilization of renewable that optimize the system value of the electricity generated and not the
energy in 64 countries studied by Ref. [19] indicate that carbon pollu­ power production. One example of this is arranging PV panels so that
tion impairs alternative power requirements in 20 developing nations. their output is more closely matched with the demand pattern. For trade
According to, using renewable energy has a good effect on the economy. openness, believe that capacity-based support schemes may be preferred
Additional factors influencing dependence on fossil fuels in China’s re­ and offer a variety of additional adjustments in the trade architecture,
gions include the number of dioxide emissions, percent export, or capita such as nodal pricing and smaller market time intervals. Capacity-based
importation, and urbanization. However [20], show that lowering the assistance programs, rather than resource subsidies, have been proven to
cost of carbon emissions in China promotes the use of renewable energy. be more successful at lowering technological costs.
As have shown, sustainable energy usage in China lowers carbon output.
In emerging countries, observed a negative correlation between sus­ 3. Methodology and data
tainable energy use and Carbon intensity, as did [21] in the UK.
3.1. Theoretical analysis
2.3. Energy prices and renewable energy consumption
Capital, labor, technical advancement, and energy are the essential
Found that oil prices had an inverse relation with the amount of ingredients of economic growth in industrialized countries, according to
electricity used in the United States. The requirement for energy grows recent research on the subject. Established the analytical framework
as the price of petroleum products rises due to technical progress and employed here, which is supported by Ref. [27]. In order to estimate the
efficiency improvements in carbon dioxide emission. There is an abun­ long-run connection between variables, current paper extends the ARDL
dance of studies on the link between actual oil and gas prices and sus­ and NARDL model function by include renewable and nonrenewable
tainable energy use, including [22]. There is a strong correlation energy consumption, as well as fixed capital variable and total labor

3
L. Xu and Y. Wu Renewable Energy 215 (2023) 118983

variable. As a result, precisely calculating and assessing the links be­ confirm the integrating relationship between the variables. At a level of
tween renewable and non-renewable energy consumption, industrial 5% significance, the measured F-statistic ensures a long-term co-inte­
production function [28], and gross domestic product growth in devel­ grating relationship between the variables. The recently proposed test
oping environmental policies such as pollution and energy taxes. The by Westerlund is used to examine panel co-integration interactions be­
first experimental model created to observe the impact of these financial tween the variables. In comparison two sets of panel co-integration tests
and demographic factors on renewable energy use is as follows: undertaken [32]. The structural breaks type approach to detect the
location of structural breaks endogenously, minimizing the sum of
Yt = TcCaat Laβt Enγt (1)
squared residuals worldwide. Two alternative models are used to
conduct the panel co-integration test. Model A: GDP output variable =
Yt = TcCaat Laβt Reγ1 γ2
t Nrt (2)
function (Fixed Capital (Ca), Total Labor (La), consumption of renew­
The elasticities of production with regard to fixed capital (Ca), total able energy (Re), consumption of non-renewable energy (Nr). Model B:
labor (La), and energy are measured by A, B, and C, respectively. Energy Industrial Output Variable (IV) = Function (Fixed Capital (Ca), Total
is divided into two groups, according to Ref. [29]: consumption of Labor Force (La), consumption of renewable energy (Re), consumption
renewable energy (Re), consumption of non-renewable energy (Nr). The of non-renewable energy (Nr).
output elasticity with regard to consumption of renewable energy,
consumption of non-renewable energy, respectively, is Y1 and Y2. The 3.2.3. Long-run examination
production function’s logarithmic form produces a log-linear form in Model A has cross-section dependency, heteroscedasticity, and serial
Table 1. correlation, Based on diagnostic tests presented in Table 2, while Model
B has heteroscedasticity and serial correlation. When models have these
Ln Yt = LnTc + aLnCat + β Ln Lat + γ1 Ln REt + γ2 Ln Nrt + ut. (3) issues, traditional panel ADF statistics fixed or random effects can lead
Equations. (3), (4) are also evaluated using industrial output as the to erroneous conclusions and even inconsistent estimates. This paper
dependent variable to recognize the variables links between various Common Correlated effects Mean Group estimator works by filtering
forms of energy, industrial and economic variables: cross-section specific repressors using the cross-section averages of the
dependent variable and observed regresses. Cross-section dependency
Ln IVt = LnTc + aLnCat + β Ln Lat + γ1 Ln REt + γ2 Ln Nrt + ut. (4) may be avoided in this way because the unobserved common component
LnIVt Stands for industrial value added. Equations. (3) And (4) are can be well represented by cross-section averages. CCEMG estimators
referred to as Model-A and B are, respectively, in empirical analysis. are consistent even when the unobserved components have a unit root,
according to Ref. [33]. Local and global shocks are not a problem for the
CCEMG estimato.
3.2. Empirical approach
3.2.4. ARDL analysis for co-integration
3.2.1. Unit root test The ARDL (autoregressive distributive lag) approach is used for co-
To avoid erroneous regression, the time series characteristics of integrating analysis. For a shorter and longer period, this approach de­
variables must be investigated in the empirical analysis. The integra­ termines the connection between the variables in the model. When it
tional characteristics of the series are determined in the first stage. This comes to co-integration analysis, the ARDL model outperforms other
work uses three distinct unit root tests, including [30] (LLC), to techniques. In ARDL technique have two valid results, whether
accomplish this paper aims to analysis of sensitivity and robustness (IPS) I(0) or I(1) or co-integrated amongst them [I(0) and I(1)]. It also enables
as well as [14]. Presents a test statistic based on the t-ratio that does not for long- and short-term relationship monitoring of small and big sample
require bias correcting variables. Due to its pooling design, it also serves determinants, as well as non-biased predict coefficients and relevant t-
as a test against the homogeneous alternative. The LLC presupposes statistics, due to the explanatory factors’ endogenous presence [34,35].
homogeneity in the dynamic of the autoregressive (AR) coefficients for The ARDL-bound methodology to co-integration has gained acceptance
all panel members and is based on the augmented Dickey–Fuller (ADF) among researchers in past years. The equations below show the empir­
method. The LLC technique, in particular, assumes that each unit in the ical formulation of the ARDL co-integration test technique.
panel data has the same AR (1) coefficient, but allows for separate ef­
fects, period wise effects, and, finally, a time trend. The panel based ADF
test that allow for instantaneous stationary and non-stationary series
wise assessment. The IPS proposes a novel, more bendable and
computationally very simple unit root testing approach. The [31], test, a
generalized of panel stationary test of and based on the assumption that
the Long-Run Test homogenous variance or heterogonous variance. Table 2
Under the, fisher-ADF and fisher-ADF panel test the null hypotheses is an Model A and B diagnostic estimation test.
unit root. Model A and B RE assessment FE assessment

Model A
3.2.2. Panel Co-integration Cross sectional dependence
Before regressing the ARDL model, a limits test must be used to P-Value (Pesraran) 0.000*** 0.001**
Heteroscedasticity
P-Value (Modified Wald) 0.000***
Table 1
Sequential correlation
Variables and description. P-Value (Wooldridge) 0.001**
Variable Description Model B
Cross sectional dependence
Yt at time t, indicates the total output P-Value (Pesraran) 0.2066*** 0.3842***
Tc Represent the technology parameters Heteroscedasticity
Caat Represent the capital P-Value (Modified Wald) 0.000***
Laβt Represent the labor Sequential correlation
Enγt Represent the energy P-Value (Wooldridge) 0.00***
Reγ1
t
Represent the renewable energy
Represent the non-renewable energy
Note: RE and FE Donets fixed and random effects assessment. *** shows that the
Nrγ2
t
P-value is significant at the 1% level.

4
L. Xu and Y. Wu Renewable Energy 215 (2023) 118983

lnREt = b0 + ¬ p i = Model1 b1ilnCat − i + ¬ p Table 3

Results of descriptive results.
= 0 Model2i + lnGDP1t − i + ¬ p i
Descriptive GDP La Nr Ca Ta Re
= 0 b3iLabr2t − i + ¬ p i = 0 b4ilnGDP3t − i + ¬ p i
Mean 4.862 2.830 2.889 0.019 5.138 0.336
= 0 b5i + lnGDP2 3t − i + ¬ p i Median 4.627 3.422 3.704 0.063 5.973 0.051
= 0 b6i + lnCapital4t − i + ¬ p i Maximum 4.999 0.494 2.787 0.056 2.108 2.698
Minimum 3.736 0.866 0.898 0.068 3.128 0.611
= 0 b7i + lnLa5t − i + b8lnCO2 t − 1 + b9lnREt − 1 Std.Dev. 4.950 3.901 5.017 0.035 3.324 0.183
Jarque-bera 4.710 4.614 4.762 0.044 4.156 0.330
+ b10lnEnergyt − 1 + b11lnGDPt − 1 + b12lnNrPt
− 1 + b13lnGDPt − 1 + b14lnNrt − 1 + bDumTat Note: La: Total Labor, Nr: Non-Renewable Energy consumption, Ca: Fixed
Capital, Terotechnology, Re: Renewable Energy consumption, GDP: Gross Do­
+ εt, mestic Product.
(5)
For stationarity checks on the variables, ADF test and PP test unit effects and random level effects. To account for the impact of time-
root tests are measured as pre-estimation procedures. The LM test, invariant variables, fixed effects models include time-invariant effects.
ARCH test, and Ramsey RESET test are post-estimation tests that are For each table column model with different random variable on the fixed
used to assess autocorrelation, heteroscedasticity and fitted values’ non- capital effects that can be associated with the explanatory variables. The
linear combination issues, respectively. To determine if the researched individual-specific effect is a random variable that is uncorrelated with
model is stable, the panel PP and panel LCC square tests are used. This the explanatory factors in the random effects model. It is worth noting
allows us to agree equations for additional variables and investigate that utilizing these approaches without taking into account diagnostic
short and long-term causation between test. tests like cross-sectional dependency, heteroscedasticity, and serial
correlation might lead to standard error bias and inefficient outcomes.
The diagnostic test findings are used to choose an acceptable approach
3.3. Data description for predicting the long-run results joining between the all selected var­
iables. Table 1 captures results on the diagnostic tests for the model A
Annual statistics on (variable 1) real GDP, (variable2) Industrial and B. Based on the results of PP tests and ADF test that are long run
production, (variable 3) Fixed capital, (variable 4) Total labor force, effects of RE and NRE energy. More specifically, a 5% growth of
(variable 5) renewable energy sources and (variable 6) non-renewable renewable and non-renewable energy sources, and all other variables
energy sources for an Advance poll panel of 900 observations for the constant position. Moreover, model results, at significance level 1%, the
study on the selected ASIAN countries are China, Vietnam, Indonesia, results of the ADF tests for heteroscedasticity and serial correlation show
Singapore, Malaysia, Japan, Bangladesh, Philippines, Nepal and that the models have an issue with heteroscedasticity and serial
Thailand gathered for the period 1995 to 2018. Are among the correlation.
10countries included in the study. In this study, GDP per capita (con­ After establishing the dataset’s normal distribution, unit root tests
stant 2000 U S.$) is utilized as a proxy for economic production using applying ADF test and PP test at the level with first (difference) is used to
purchasing power parities (PPPs). Already-produced durable items are determine the variables’ assimilating command in Table 4. The variables
referred to as capital, which is employed as an input variable function, exhibit a mixed order of integration, suggesting that they are stationary
and output variable industrial capital stock data is difficult to get and at both the level and first difference, based on ADF model results. These
quantity, gross fixed capital creation is frequently employed as a proxy findings underline the fact that the variables in both models have a co-
for total fixed capital stock increase. Specifically, assuming a constant integrating relationship. Furthermore, the need of the ARDL limits
depreciation rate in line with the perpetual inventory technique changes testing technique to co-integration is prompted by the heterogeneous
and renewable energy investment with changes in the industrial capital integrating order among the variables.
stock [36]. As a result, this analysis relies on statistics on real GDP and
FDI in billions of constant values 2000 United States America (USA) 4.1. Panel unit root test
dollars, total labor force (La) calculate in millions and industrial energy
output variables values in (billions of constant 2000 US$) are used. The Table 4 reports the unit root model, which include (LLC) [37], (IPS).
World Bank provided all of the information described above. As a result, The existence of a unit root, suggesting non-stationarity, is treated as the
nonrenewable energy consumption in this study is calculated as the sum null hypothesis indicated all panel model tests, the unit root test results
of all of these sources energy consumption calculated in quads units. for the natural logarithm level of selected variables are statically insig­
Such as geothermal energy, wind energy, photovoltaic energy, and solar nificant expect the other tests. On the other hand, model A-B tests
thermal energy consumption are all quantified in quadrillion Btu units of comprise individual trends and constants. At the 1% level of signifi­
renewable energy consumption. The World Bank (2020) provided all cance, the statistics results for a 5% level of significance (LnGDP 4.945,
statistics on energy use. Prior to running the analysis, all variables are p-value = 1.000).
transformed to natural logarithms so that the model’s parameter esti­ The panel unit root tests indicated that the study variables in the
mations may be understood as elasticity estimates. model are normal for 5% level of significance (Ca = 4.445, p-value =
9.333). On the other hand, qurdic kernel test was performed for the
3.3.1. Descriptive summary homogeneous by the critical values coefficient of the spatial lagged
Table 3 shows the summary statistics for data properties. According explanatory variables.
to analysis, the minimum values of renewable and non-renewable en­
ergy is (0.611 and 0.898, and maximum 2.698 and 2.787, respectively. 4.2. ARDL-bound based estimation
The GDP minimum values is 3.736 and the maximum values is 4.999.
Summary statistics reveal that the time-series dynamics have low stan­ It is necessary to validate the ARDL limits testing strategy to earlier
dard deviations, showing that the data is distributed normally. co-integration tests applying the ARDL-based simulations model, which
is decided by the obtained F-statistic value. Table 5 shows that the
4. Empirical results estimated F-statistic value is greater than the upper bound’s critical
value, indicating that the null hypothesis of no co-integrating relation­
Table 4 shows the results of a panel unit root tests of fixed level ship between the variables is true. It means that for Models A and B,

5
L. Xu and Y. Wu Renewable Energy 215 (2023) 118983

Table 4
Model A and B variables with panel unit root tests.
Methods LnGDP LnLa LnTa LnRe LnNr LnCa

Breitung (Level) 4.945 1.756 0.177 1.099 6.950 − 1.089

(1.000) (0.988) (0.877) (0.912) (1.000) (0.140)
st
1 Difference − 2.786 − 3.444 − 3.447 − 6.188 − 9.865 − 8.055
(0.001)*** (0.000)*** (0.000)*** (0.000)*** (0.000)*** (0.000)***
LLC level 12.005 − 1.534 6.188 1.192 3.944 − 0.091
(1.000) (0.400) (1.000) (0.966) (0.677) (0.155)
1st Difference − 3.755 − 7.460 − 3.534 − 5.177 − 17.765 − 18.452
(0.000)*** (0.000)*** (0.000)*** (0.000)*** (0.001)*** (0.000)***
1PS level 6.345 3.588 4.170 3.344 3.944 1.291
(1.000) (0.998) (0.997) (0.999) (0.999) (0.922)
1st difference − 5.744 − 10.435 − 6.540 − 5.443 − 24.866 − 20.633
(0.000)*** (0.000)*** (0.000)*** (0.000)*** (0.000)*** (0.000)***

Note: ***, **, and * denote significance level at 1%, 5%, and 10%, respectively. Source own author calculations.

Table 5 Table 6
ARDL bound test estimation. Regression results on the Asian countries.
Model -A: Dependent Variable LnRe Model -B: Dependent Variable LnNr Variables Model

Test statistic Value Test statistic Value LnGDP 0.3236***(3.65)

F. statistic 12.44561 F. statistic 7.65543 LnLa 4.4455** (3.60)
Critical bound value LnTa − 0.0040*** (− 3.676)
Significance I (0)Bound I (1)Bound LnRe 0.2365*** (− 5.14)
10% 2.22*** 3.24*** LnNr − 0.1451*** (− 4.22)
5% 2.33** 3.77** LnCa 0.0021* (2.21)
3% 2.75 3.98* Sargan test 0.857
1% 3.51* 4.11* AR(1) 0.0002
AR(2) 0.2451
Note: ***, **, and * denote significance level at 1%, 5%, and 10%, respectively.
Source own author calculations. Note: ***, **, and * denote significance level at 1%, 5%, and
10%, respectively. Source own author calculations.
study [38] variables are co-integrated, allowing the dynamic ARDL
simulations model to be used to estimate short and long-run linkages
Table 7
between variables.
Results of causality tests for GDP and renewable energy.
The results of the dynamic ARDL simulations technique for Models A
Nations Hypstehsis Wald bootstrap Critical values
and B are shown in Table 5. Income growth has a long run and short-run
Test
positive and negligible profile to affect renewable energy usage, ac­ Statistics
cording to the analysis. This means that a 1% increase in income in­
1% 5% 10%
creases renewable energy consumption by 0.41% in the end and 3.22% China GDP∕
=RE 0.428*** 2.813 1.345 2.430
in the short term, respectively (Model-A). Asian present economic boom GDP+∕=RE+ 0.025 16.130** 17.788** 22.516**
is the result of a slew of economic operations and governmental pledges. GDP-∕
=RE- 12.002 6.658 8.971 8.524
Asian counties member country China officials began several renewable Malaysia GDP∕
=RE 9.020 1.984 11.512 8.352
GDP+∕=RE+ 13.013 2.398 10.989 9.295
energy initiatives in 2000, with the goal of meeting more than 14% of
GDP-∕
=RE- 8.017 5.346*** 0.371*** 0.346***
the economy’s energy needs through renewable energy projects. Table 5 Indonesia GDP∕
=RE 10.020 6.042 7.036 0.030
shows the results of the panel model test with structural breakdowns. GDP+∕=RE+ 9.022 9.015 6.061** 10.055*
According to these findings, the null hypothesis of stationarity is rejec­ GDP-∕
=RE- 8.162*** 5.031 5.033 6.030
ted for all variables at the 1% level, and for most variables at the 3.4% Thailand GDP∕
=RE 7.062 10.081** 9.130*** 5.188***
GDP+∕=RE+ 5.277*** 1.029 6.028 10.030
and 2% levels, by either the homogeneous or the heterogeneous long- GDP-∕
=RE- 33.084 8.126*** 4.200** 12.095***
run variance. As a result, even when structural breaks are allowed, all Vietnam GDP∕
=RE 6.017 12.031 5.053 14.024
of the variables are non-stationary at their levels. As can be seen in GDP+∕=RE+ 10.032) 11.175 5.180 12.211
Table 7, the homogeneous effects of renewable energy consumption GDP-∕
=RE- 0.188*** 9.181 4.131 6.170
Bangladesh GDP∕
=RE 0.030 3.252*** 2.117*** 2.298***
account approximately 5% total of total effects (20.123) **9.910).
GDP+∕=RE+ 0.083 0.256 0.298 8.305
GDP-∕
=RE- 0.070 1.589*** 3.262*** 3.378***
Singapore GDP∕
=RE 5.444 9.216 0.284 0.315
4.3. Regression results on the renewable and nonrenewable energy GDP+∕=RE+ 6.430 2.813 1.345) 2.430
GDP-∕
=RE- 2.117*** 16.130** 17.788** 22.516**
Nepal GDP∕
=RE 19.084 1.984 9.512 12.352
Table 6 shows the influence of FDI on renewable and non-renewable
GDP+∕=RE+ 21.017 2.398 10.989 13.295
energy throughout two economic eras. The findings revealed that FDI GDP-∕
=RE- 18.032 10.346*** 11.371*** 10.346***
has a negative but minor direct influence on growing renewable energy Philippines GDP∕
=RE 7.036 5.042 5.036 30.030
usage during a recession. FDI is unsuccessful in boosting renewable GDP+∕=RE+ 8.041 6.015 7.061** 10.055*
energy usage during a recession, according to the findings. Similarly, the GDP-∕
=RE- 5.029 4.031 5.033 9.030
Japan GDP∕
=RE 6.118*** 25.081** 11.130*** 8.188***
interaction term of FDI revealed that FDI was ineffective in increasing
GDP+∕=RE+ 9.037 50.029 6.028 7.030
(RE) renewable energy and conventional energy usage during the GDP-∕
=RE- 10.091** 8.126*** 9.200** 6.094***
recession periods. FDI, on the other hand, works hard to enhance
Note: ***, **, and * denote significance level at 1%, 5%, and 10%, respectively.
renewable energy during times of boom. Along the way, FDI raises non-
Source own author calculations.
renewable energy use.
Investigations have stated that the distribution of resources

6
L. Xu and Y. Wu Renewable Energy 215 (2023) 118983

significantly influences effectiveness [39]. This term describes the pro­ connection shocks.
cedure of reducing the amount of misallocation of resources and The long run and short-run outcomes for this investigation produced
therefore enhancing energy efficiency during the tariff-free process of utilizing the augmented mean group (AMG) technique are shown in
the pollution marketplace. The [40]lies at the basis of emissions Table-8. The findings of AMG confirm that GDP, La, and IV have a
reduction trading, i.e., the objective of improving resource allocation by detrimental impact on Ta and Nr. The effects of GDP, Re, Ca, and Ta on
clarifying property rights and employing economic mechanisms to renewable energy consumption are positive, but the effect of FDI on
overcome externality concerns. The following operation rules govern renewable energy consumption is negative. The average speed of
trading in emissions: the government determines the region’s allowable adjustment towards equilibrium for the technology and the nonrenew­
maximum greenhouse emissions, divides that amount into several able energy consumption is 0.61% and 0.62%, respectively. In the short
shares to be distributed to each company, and treats it as a special run, HC causes a 0.56% increase in Labor, fixed capital is 0.0012%, FDI
commodity that can be traded on the business. Trading in carbon is 0.019%, and energy usage is 0.0055%. Capital, on the other hand,
emissions allows firms with excess carbon pollution permits to make up produced a 0.021% decrease in renewable energy consumption on
for the shortage of other organizations, and the extra money earned may average. On average, adjusts to equilibrium at a rate of 0.55% every
be used to buy other items. This is called a “carbon emissions swap.” The year.
higher cost of purchasing credits decreases the input of other compo­ In the long term, the average drop induced by Ta for industrial
nents, affecting the market’s distribution of elements. There are several technology and nonrenewable energy is 0.11% and 0.062%, respec­
benefits to implementing this approach, including eliminating strict tively, Ca is 0.0024% and 0.0043%, La is 0.0069% and 0.0088%, and Ca
environmental consistency, giving businesses more freedom, and using is 0.0028 and 0.0043%. GDP, on the other hand, induces a 0.0034% and
liberal economic transactions to remedy resource misalignments. The 0.0042% rise in industrial technology and nonrenewable energy,
term “resource squandering” refers to when the marginal production respectively. The average growth in renewable energy generated by Ta
values of various sectors or businesses depart from the “productive is 0.55%, Ca is 0.0024%, GDP is 0.0164%, and La is 0.0058%. GDP, on
allocation” condition [41]. If the allocation of components is distorted, it the other hand, generates a 0.011% decrease in renewable energy for the
will cause a loss in efficiency, becoming more severe as the imbalance ASIAN countries. In the short run, the average drop induced by Tech­
grows. When the greenhouse gas trading center is up and running, re­ nology for industrial technology and nonrenewable energy is 0.10% and
sources move from businesses with low efficiencies to those with high 0.070%, respectively, La is 0.0039% and 0.0048%, La is 0.0028% and
usage rates. Energy efficiency will increase when the marginal product 0.0052%, and GDP is 0.0076% and 0.0033%. Ta, on the other hand,
of businesses grows, and resources are better allocated [42]. The [43] exhibits an average rise of 0.0046% in industrial technology and
thus conclude that introducing an energy consumption trading mecha­ 0.0054% in nonrenewable energy. The negative association of Tech­
nism will enhance energy savings by maximizing the allocated nology with Industrial technology and nonrenewable energy consump­
resources. tion in the long-run, but a positive link with renewable energy
Table 6 shows the institutional influence on renewables and non- consumption in the short run, shows that human capital development
renewables throughout two economic eras. Emissions reductions facilitates the transition from nonrenewable energy consumption to
credit’s green manufacturing benefit mostly stems from market revenues
introduction. Businesses are almost certainly to accelerate the
advancement of green technology to take advantage of this portion of Table 8
the market’s income [44]. The assistance of the government. The in­ Results using augmented mean group (AMG) technique.
fluence of government in business cannot be overstated. The govern­ Working Industrial Renewable Non-renewable
ments may give policy assistance, such as subsidies or tax cuts, to Variables Technology energy energy
businesses that produce low levels of carbon emissions. Providing GDP − 0.11*** − 0.062*** 0.55**
financial assistance for R&D may help businesses alleviate their financial (0.0343) (0.016) (0.0272)
burdens while also encouraging them to devote more resources to the La − 0.0024*** − 0.0042*** 0.0024**
(0.0009) (0.0017) (0.0012)
invention of green systems to reduce greenhouse gas emissions and
Ta − 0.0028*** − 0.0043* 0.0164**
reduce carbon. (0.0010) (0.0023) (0.0080)
As a result of the diminished market competition, enterprises will Re − 0.0069*** − 0.0088*** 0.057**
have to enhance their energy efficiency to compete. More money spent (0.0037) (0.0021) (0.0030)
on the energy consumption trading market means more pressure on Nr − 0.0033*** − 0.0042* − 0.010*
(0.0015) (0.00243) (0.0056)
corporations’ R&D budgets to come up with new, cutting-edge tech­
Ca − 0.0038*** − 0.0067*** 0.077*
nologies. Process innovation has become an essential basis for busi­ (0.0026) (0.0028) (0.0055)
nesses to save money and improve their performance when the purchase Constant 0.0028*** 0.31*** − 0.050**
price of energy consumption credits exceeds the cost of green product (0.0045) (0.0062) (0.0224)
Short-run test Results
innovation. As a result, these high-emissions businesses will increase
GDP − 0.10*** − 0.070*** 0.57*
their investments in green innovations and gradually bring about tech­ (0.0294) (0.0125) (0.3032)
nologies, energy reformation, and upgrade. La − 0.0039*** − 0.0048* 0.0013***
(0.0015) (0.0029) (0.0005)
4.4. Causality analysis (renewable and GDP) Ta − 0.0076*** − 0.0048* 0.0013***
(0.0041) (0.0029) (0.0005)
Re − 0.0076*** − 0.0033* 0.019**
According to the findings in Table 7, there is no non-asymmetric (0.0041) (0.0017) (0.0097)
causation link between economic development and renewable energy Nr − 0.0028*** − 0.0052** − 0.010*
usage in the China, Philippines, Indonesia, Malaysia, Singapore and (0.0036) (0.0027) (0.0054)
− 0.0038*** − 0.0067*** 0.077*
Vietnam. Asymmetric and symmetric testing both support this claim. Ca
(0.0026) (0.0028) (0.0055)
While there is no non-asymmetric causality link for Thailand and Japan, ECM( − 1) − 0.0050*** 0.32*** − 0.050**
asymmetric causality analysis validates the negative shocks causality (0.0205) (0.0086) (0.012)
relationship from GDP to renewable energy. The outcome of a non- Constant 0.27*** 0.33*** − 0.050**
asymmetric causality analysis for the Philippines, Nepal and (0.0464) 0.0086 0.2348

Bangladesh suggests that there is a causal relationship between GDP and Note: ***, **, and * denote significance level at 1%, 5%, and 10%, respectively.
renewable energy consumption. This is true in the case of positive Source own author calculations.

7
L. Xu and Y. Wu Renewable Energy 215 (2023) 118983

renewable energy consumption. estimation, LnGDP and LnRE consumption in these nations do not create
Energy consumption management is premised on the notion that CO2, even if these two variables provide a more precise diagnostic for
businesses may exchange greenhouse gas credits as a commodity. The achieving a more sustainable environment.
purchasing technique is critical to the success of carbon emissions It’s possible that perhaps the pilot towns for the greenhouse gas
trading. An important consideration is a degree to which the transfer of trading plan were not picked at random when using the big distinction
resources from low-to increased sectors is facilitated by the amount of approach, which may successfully address certain endogenous diffi­
corporatization [45]. To account for regional disparities in financiali­ culties. Opportunities for additional influences on the selection of pilot
zation, the energy consumption trading plan’s impact on energy effi­ towns for greenhouse gas exchange might alter the accuracy of findings.
ciency will vary. Accordingly, in light of this, the [46]performed an As a result, the endemic issue was minimized to the greatest extent
investigation based on the degree of deregulation and privatization of feasible using the dynamic panel technique. This work employed the air
various cities. The Fan et al. marketization index is often used to gauge circulation parameter as the main success parameter as an indirect in­
an organization’s level of corporatization. A city-level measurement is dicator for the carbon dioxide emissions market pilots based on work
missing from the index. As a result, the study [47] was cited, and the [50]. This is because stricter environmental rules are used when the air
degree of corporatization was measured by the percentage of the total flow coefficient is less and the concentration of pollutants detected is
workforce in the government industry to the overall total workforce. higher. Exogenous conditions are also met by the fact that the forced air
Higher the percentage, the greater the level of commercialization. coefficient can only impact energy savings via carbon pollution trading.
Homogeneity analysis is given in columns (1) to (4) of Table 8 for Wind speed and boundary layer height may calculate the air circulation
varying levels of corporatization. This study’s findings show that the coefficient. The article’s measures types may be found using the geo­
greenhouse gas buying and selling result is statistically significant at a metric mean.
significance level of 1% for towns with a large degree of corporatization. According to Table 9, the results of the two-stage regression estimate
In comparison, a constant rate value is lower at a significant 5% for cities of the dependent variable are presented in columns (1) and (2) of
with a low degree of corporatization, with a correlation value of 0.082. Table 9. Because both F-values are more than 10, we may discard the
Coefficients of energy consumption trading in cities with strong mar­ idea that material possessions are inherently weak. It is clear that this
ketization are statistically significant at a significance level of 5%; article’s conclusion that “energy consumption trading schemes may
however, those in areas with low corporatization are not statistically greatly enhance energy efficiency” has a reputation to uphold consid­
relevant, with a coefficient of determination of 0.033. This illustrates ering the regression findings of the two stages.
that the energy consumption trading plan has been pretty obvious in Several Asian nations have made efforts to reduce the negative
advancing energy efficiency in places with a greater extent of market effect of energy-based benefit programs, such as withdrawing
corporatization. fossil fuel subsidies during times with low costs for more than six
consecutive hours [51]. The non-convexity of GHGs issues means that
individual manufacturers may not always have the motivation to pursue
4.5. NARDL estimation the best path in terms of pricing. A timetable of production. There are
times when the RE restriction may not be ideal from a systemic stand­
The findings of the panel NARDL models for a sample of Asian point while no reimbursements are supplied [52,53].
countries are summarized in Table 9. The influence of decomposed For narrow generating, there are other impediments to economic
explanatory factors on renewable energy is seen in Model-1 results. The relations: value and technology. Concerns have been expressed that
coefficient values, which characterize the level of industrial output when support programs for small-scale Nosocomial are rolled away,
through positive change in Nonrenewable energy as (0.05) and negative considerable quantities of energy, such as for slight PV in Bangladesh
change as (0.17), are shown in the decomposed renewable energy with and Sri Lanka, maybe decommissioned as a result. Small-scale in­
positive (negative) effects on environment. These studies suggest that a stallations have difficulty balancing adequate assistance with the need
1% increase (drop) in industrial technology will result in a 0.5% to avoid unnecessary integration expenses and using concerted, system-
reduction in GDP (0.17%). Several studies have found a negative rela­ friendly control systems and aggregation for small-scale RE operations
tionship between renewable energy and nonrenewable energy con­ [54].
sumption. The elasticities of renewable energy increase (reduce) by 0.06 The dynamism of power networks, such as the binary character of
(0.06), suggesting that a 1% increase (decrease) in renewable energy commitments choices and the coupling between various periods, may
would result in a 0.05% reduction in greenhouse gasses. This observa­ result in inaccurate findings when using regression analysis to determine
tion is consistent with the findings of [48,49]. As we saw in the linear

Table 9
Results of Panel nonlinear estimation ARDL test.
coefficient (γ) LnGDP LnLa LnTa LnRe LnNr

Long-run Equation
LnGDP 4.080** 3.316 4.746 4.754 4.593 4.642
LnLa − 1.833 − 2.302 − 3.353 − 84.706 − 42.427 − 78.136
LnTa 2.129*** 2.518 4.287 5.932 6.278 4.711
LnRe − 2.998 − 3.103 − 4.013 − 66.601 − 114.599 − 80.587
LnNr 3.999 2.316 3.882 4.739 5.939 4.720
LnCa − 1.94 − 4.942 − 2.946 − 117.727 − 101.668 − 52.67
Short-run equation
LnGDP 2.261*** 2.787 2.176 5.977 6.243 5.139
LnLa − 4.979 − 6.432 − 2.128 − 103.877 − 154.393 − 46.646
LnTa 2.807*** 2.921 1.922 6.240 6.332 5.809
LnRe − 4.842 − 3.979 − 3.848 − 102.444 − 104.598 − 102.922
LnNr 3.914* 6.100 3.967 4.695 4.681 4.636
LnCa − 1.51 − 1.619 − 3.178 − 109.45 − 103.267 − 46.218
− 3.194 − 2.824 − 1.995 − 193.381 − 120.638 − 103.777
ρ – 0.979 0.845 – 0.778 0.708

Note: ***, **, and * denote significance level at 1%, 5%, and 10%, respectively. Source own author calculations.

8
L. Xu and Y. Wu Renewable Energy 215 (2023) 118983

the influence of RE on CO2 emissions. A study using transmission line of local governments’ renewable energy programmers.
models to simulate the influence on Emissions of carbon dioxide may be
preferred to an economic structure like [1]. Although this study’s con­ Declaration of competing interest
clusions may not apply to more practical experiments, it would be
worthwhile to do investigations using the statistical method for bigger The authors declare that they have no known competing financial
bioassays to see whether comparable findings can be discovered. interests or personal relationships that could have appeared to influence
the work reported in this paper.
5. Conclusion and policy implications
Data availability
Using data from selected ASIAN countries from 1995 to 2018, this
new article investigated the dynamic correlation among non-fossil fuel The authors do not have permission to share data.
and fossil - based energy utilization, economic growth, and Gross do­
mestic product (GDP). For empirical research, the panel co-integration References
approach with structural breakdowns is employed. The findings indi­
cate that non-fossil fuel and fossil - based energy, economic growth, and [1] Q. Quan, W. Liang, D. Yan, J. Lei, Influences of joint action of natural and social
economic expansion have a long-term equilibrium connection. In both factors on atmospheric process of hydrological cycle in Inner Mongolia, China,
Urban Clim.. 41 (2022), 101043, https://doi.org/10.1016/j.uclim.2021.101043.
the short and long term, the panel causality studies reveal bidirectional [2] H. Liu, P. Yao, X. Wang, J. Huang, L. Yu, Research on the peer behavior of local
causation between industrial production and both non-fossil fuel and government green governance based on seci expansion model, Land 10 (5) (2021)
fossil - based energy consumption. The results from the panel error 472, https://doi.org/10.3390/land10050472.
[3] L. Qiu, L. He, H. Lu, D. Liang, Systematic potential analysis on renewable energy
correction model reveals that the positive (negative) elasticities of centralized co-development at high altitude: a case study in Qinghai-Tibet plateau,
nonrenewable energy consumption are 0.96 (1.15), suggesting that a 1% Energy Convers. Manag. 267 (2022), 115879, https://doi.org/10.1016/j.
rise (decrease) in nonrenewable energy is expected to increase (reduce) enconman.2022.115879.
[4] Z. Zhang, J. Tian, W. Huang, L. Yin, W. Zheng, S. Liu, A haze prediction method
0.96% (1.15%) in corban dioxide emissions. These findings suggest that based on one-dimensional convolutional neural network, Atmosphere 12 (2021)
selected ASIAN countries’ industrial production and overall economic 1327, https://doi.org/10.3390/atmos12101327.
growth are still energy-dependent. The results indicate that real GDP [5] X. Wu, Z. Liu, L. Yin, W. Zheng, L. Song, J. Tian, B. Yang, S. Liu, A haze prediction
model in chengdu based on LSTM, Atmosphere 12 (2021), https://doi.org/
resistances with regard to clean or renewable energy and energy from
10.3390/atmos12111479.
conventional sources values at 0.201% and 0.288%, respectively. Ac­ [6] F. Jincheng, K. Gangqiang, Y. Qing, Group performance of energy piles under
cording to these findings, a 5% growth in clean or renewable energy cyclic and variable thermal loading, J. Geotech. Geoenviron. Eng. 148 (2022),
consumption between in 0.145%–0.290% increases in gross domestic 4022060, https://doi.org/10.1061/(ASCE)GT.1943-5606.0002840.
[7] X. Lei, Q. Xu, C. Jin, Nature of property right and the motives for holding cash:
product (GDP), respectively, and 0.088% and 0.289% increases in in­ empirical evidence from Chinese listed companies, Manag. Decis. Econ. 43 (2022)
dustrial power production. 1482–1500, https://doi.org/10.1002/mde.3469.
These data show that both renewable and non-renewable energy [8] Y. Chen, E.K. kumara, V. Sivakumar, Investigation of finance industry on risk
awareness model and digital economic growth, Ann. Oper. Res. (2021), https://
consumption encourage economic growth in selected ASIAN countries, doi.org/10.1007/s10479-021-04287-7.
although non-renewable energy remains the primary driver of industrial [9] Y. Wu, W. Zhu, The role of CSR engagement in customer-company identification
production and real GDP growth. Given the positive and large impact of and behavioral intention during the COVID-19 pandemic, Front. Psychol. (2021),
https://doi.org/10.3389/fpsyg.2021.721410.
renewable energy on industrial output and GDP growth, boosting [10] L. Deng, Y. Zhao, Investment lag, financially constraints and company
renewable energy use might be a realistic option for addressing climate value—evidence from China, Emerg. Mark. Finance Trade (2022), https://doi.org/
change and ensuring long-term economic prosperity. The key causality 10.1080/1540496X.2021.2025047.
[11] A. Dilanchiev, T. Taktakishvili, Macroeconomic determinants of household
results reveal that in both the short and long run, there is bidirectional consumptions in Georgia, Ann. Financ. Econ 16 (2021), 2150020, https://doi.org/
causation between real GDP and nonrenewable energy usage. This data 10.1142/S2010495221500202.
supports the feedback theory, which states that high economic growth [12] K. Batool, Z.-Y. Zhao, F. Atif, A. Dilanchiev, Nexus between energy poverty and
technological innovations: a pathway for addressing energy sustainability, Front.
leads to increased non-renewable energy usage and vice versa. However,
Environ. Sci. 10 (2022), 888080, https://doi.org/10.3389/fenvs.2022.888080.
there is a one-way causal relationship between GDP and renewable [13] D. Bourghelle, F. Jawadi, P. Rozin, Oil price volatility in the context of Covid-19,
energy usage. To this end, the following practical policy implications are Int. Econ. 167 (2021) 39–49, https://doi.org/10.1016/j.inteco.2021.05.001.
recommended. [14] S. Rahman, Oil price volatility and the US stock market, Empir. Econ. 61 (2021)
1461–1489, https://doi.org/10.1007/s00181-020-01906-3.
[15] U. Jarrett, K. Mohaddes, H. Mohtadi, Oil price volatility, financial institutions and
5.1. Policy implication economic growth, Energy Pol. 126 (2019) 131–144, https://doi.org/10.1016/j.
enpol.2018.10.068.
[16] N. Nonejad, Crude oil price volatility and short-term predictability of the real U.S.
The obtained results the policy implications are that the renewable GDP growth rate, Econ. Lett. 186 (2020), https://doi.org/10.1016/j.
energy usage and production must be increased in the Asian countries. econlet.2019.108527.
The renewable energy is alternative source of fossil fuel energy or [17] X. Xiuzhen, W. Zheng, M. Umair, Testing the fluctuations of oil resource price
volatility: a hurdle for economic recovery, Res. Pol. 79 (2022), 102982, https://
traditional energy that address the climate change and emission issues. doi.org/10.1016/j.resourpol.2022.102982.
The government’s policy maker’s obligation income proper inventive­ [18] K. Ullah, I. Rashid, H. Afzal, M.M.W. Iqbal, Y.A. Bangash, H. Abbas, SS7
ness, with important private and public partnerships. The discovery of a vulnerabilities—a survey and implementation of machine learning vs rule based
filtering for detection of SS7 network attacks, IEEE Commun. Surv. Tutor. 22
negative and strong bidirectional causal relationship between renew­ (2020) 1337–1371, https://doi.org/10.1109/COMST.2020.2971757.
able and nonrenewable energy use is intriguing since reducing one type [19] T. Ghosh, S. Sahu, S. Chattopadhyay, Inflation expectations of households in India:
of energy necessitates increasing the other. role of oil prices, economic policy uncertainty, and spillover of global financial
uncertainty, Bull. Econ. Res. 73 (2021) 230–251, https://doi.org/10.1111/
This conclusion suggests that renewable energy might be used to boer.12244.
replace with traditional or conventional energy sources. The ASIAN [20] Y. Arifin, Pengaruh harga minyak dunia, nilai tukar dan inflasi terhadap
nations’ energy use objectives for 2026 are to employ 30% renewable pertumbuhan ekonomi Indonesia, Econ. Dev. Anal. J. 5 (2018) 474–483, https://
doi.org/10.15294/edaj.v5i4.22184.
energy sources. However, each country’s definition should be practical.
[21] Tajuddin Rosnawintang, P. Adam, Y.P. Pasrun, L.O. Saidi, Effects of crude oil prices
Because each of the selected Asian nations has a varied predisposition to volatility, the internet and inflation on economic growth in asean-5 countries: a
employ renewable energy sources. As a result, policymakers should panel autoregressive distributed lag approach, Int. J. Energy Econ. Pol. (2021),
examine the policies of local governments as well as the incentives in https://doi.org/10.32479/ijeep.10395.
[22] R.A. Meese, K. Rogoff, Empirical exchange rate models of the seventies. Do they fit
these nations when developing new energy usage plans. It is also critical out of sample? J. Int. Econ. 14 (1983) 3–24, https://doi.org/10.1016/0022-1996
that companies and individuals participate in the process as supporters (83)90017-X.

9
L. Xu and Y. Wu Renewable Energy 215 (2023) 118983

[23] N. Antonakakis, J. Cunado, G. Filis, D. Gabauer, F.P. de Gracia, Oil and asset classes [39] T.S. Genc, OPEC and demand response to crude oil prices, Energy Econ. 66 (2017)
implied volatilities: investment strategies and hedging effectiveness, Energy Econ. 238–246, https://doi.org/10.1016/J.ENECO.2017.06.026.
91 (2020), https://doi.org/10.1016/j.eneco.2020.104762. [40] F. Liu, M. Umair, J. Gao, Assessing oil price volatility co-movement with stock
[24] K.A. Srinita, Effendi, Energy policy analysis, monetary and fiscal on inflation market volatility through quantile regression approach, Res. Pol. 81 (2023),
volatility in Indonesia, Int. J. Energy Econ. Pol. 11 (2021) 309–314, https://doi. 103375, https://doi.org/10.1016/j.resourpol.2023.103375.
org/10.32479/ijeep.10841. [41] W. Fang, Z. Liu, A.R. Surya Putra, Role of research and development in green
[25] S. Das, S.S. Roy, Following the leaders? A study of co-movement and volatility economic growth through renewable energy development: empirical evidence
spillover in BRICS currencies, Econ. Syst. (2022), 100980, https://doi.org/ from South Asia, Renew. Energy 194 (2022) 1142–1152, https://doi.org/10.1016/
10.1016/J.ECOSYS.2022.100980. j.renene.2022.04.125.
[26] H.W. Chen, N. Bin Chang, J.C. Chen, S.J. Tsai, Environmental performance [42] W. Pan, H. Cao, Y. Liu, Green” innovation, privacy regulation and environmental
evaluation of large-scale municipal solid waste incinerators using data policy, Renew. Energy 203 (2023) 245–254, https://doi.org/10.1016/j.
envelopment analysis, Waste Manag. (2010), https://doi.org/10.1016/j. renene.2022.12.025.
wasman.2010.02.002. [43] Q. Wu, D. Yan, M. Umair, Assessing the role of competitive intelligence and
[27] Q. Wang, R. Huang, The impact of COVID-19 pandemic on sustainable practices of dynamic capabilities in business accommodation of SMEs, Econ. Anal.
development goals – a survey, Environ. Res. 202 (2021), https://doi.org/10.1016/ Pol. (2022), https://doi.org/10.1016/j.eap.2022.11.024.
j.envres.2021.111637. [44] A.A.M.H. Al Asbahi, F.Z. Gang, W. Iqbal, Q. Abass, M. Mohsin, R. Iram, Novel
[28] Lyulyov Vasylieva, Streimikiene Bilan, Sustainable economic development and approach of principal component analysis method to assess the national energy
greenhouse gas emissions: the dynamic impact of renewable energy consumption, performance via energy trilemma index, Energy Rep. (2019), https://doi.org/
GDP, and corruption, Energies 12 (2019) 3289, https://doi.org/10.3390/ 10.1016/j.egyr.2019.06.009.
en12173289. [45] M. Ikram, A. Mahmoudi, S.Z.A. Shah, M. Mohsin, Forecasting number of ISO 14001
[29] A. Rempel, J. Gupta, Fossil fuels, stranded assets and COVID-19: imagining an certifications of selected countries: application of even GM (1,1), DGM, and NDGM
inclusive & transformative recovery, World Dev. 146 (2021), 105608, https://doi. models, Environ. Sci. Pollut. Res. (2019), https://doi.org/10.1007/s11356-019-
org/10.1016/j.worlddev.2021.105608. 04534-2.
[30] Y. Bai, C. Dahl, Evaluating the management of U.S. Strategic Petroleum Reserve [46] M. Umair, A. Dilanchiev, Economic recovery by developing business starategies:
during oil disruptions, Energy Pol. (2018), https://doi.org/10.1016/j. mediating role of financing and organizational culture in small and medium
enpol.2018.02.034. businesses, Process Biochem. (2022) 683.
[31] L. Neralić, M.G. Kedžo, A survey and analysis of scholarly literature in DEA [47] E.B. Agyekum, F. Amjad, M. Mohsin, M.N.S. Ansah, A bird’s eye view of Ghana’s
published by Croatian researchers: 1978 – 2018, Zagreb Int. Rev. Econ. Bus. renewable energy sector environment: a Multi-Criteria Decision-Making approach,
(2019), https://doi.org/10.2478/zireb-2019-0014. Util. Pol. (2021), https://doi.org/10.1016/j.jup.2021.101219.
[32] M. Gürlek, M. Tuna, Reinforcing competitive advantage through green [48] X Wang, J Huang, H Liu, Can China’s carbon trading policy help achieve Carbon
organizational culture and green innovation, Serv. Ind. J. 38 (2018) 467–491, Neutrality?-A study of policy effects from the Five-sphere Integrated Plan
https://doi.org/10.1080/02642069.2017.1402889. perspective, J. Environ. Manage. 305 (2022) 114357, https://doi.org/10.1016/j.
[33] M. Kaakeh, S.S. Shirazi, K.K. Gokmenoglu, The extended GREEN-A framework: a jenvman.2021.114357.
gender comparison in consumer support for sustainable businesses practices, [49] S.A.A. Shah, P. Zhou, G.D. Walasai, M. Mohsin, Energy security and environmental
J. Environ. Assess. Pol. Manag. 23 (2021), 2250011, https://doi.org/10.1142/ sustainability index of South Asian countries: a composite index approach, Ecol.
S1464333222500119. Indicat. 106 (2019), 105507, https://doi.org/10.1016/j.ecolind.2019.105507.
[34] Z. Xia, Q. Abbas, M. Mohsin, G. Song, Trilemma among energy, economic and [50] L. Yin, L. Wang, W. Huang, S. Liu, B. Yang, W. Zheng, Spatiotemporal analysis of
environmental efficiency: can dilemma of EEE address simultaneously in era of haze in beijing based on the multi-convolution model, Atmosphere 12 (2021),
COP 21? J. Environ. Manag. (2020) https://doi.org/10.1016/j. https://doi.org/10.3390/atmos12111408.
jenvman.2020.111322. [51] X. Xu, Z. Lin, X. Li, C. Shang, Q. Shen, Multi-objective robust optimisation model
[35] H. Liu, H. Zhao, Upgrading models, evolutionary mechanisms and vertical cases of for MDVRPLS in refined oil distribution, Int. J. Prod. Res. 60 (2022) 6772–6792,
service-oriented manufacturing in SVC leading enterprises: Product-development https://doi.org/10.1080/00207543.2021.1887534.
and service-innovation for industry 4.0, Humanit. Soc. Sci. Commun. 9 (2022) 387, [52] Y. Han, S. Tan, C. Zhu, Y. Liu, Research on the emission reduction effects of carbon
https://doi.org/10.1057/s41599-022-01409-9. trading mechanism on power industry: plant-level evidence from China, Int. J.
[36] U.S. Energy Information Administration, Int. Energy Outlook 2017 (2017) Clim. Chang. Strat. Manag. (2022), https://doi.org/10.1108/IJCCSM-06-2022-
forecasts/ieo/pdf/0484(2016).pdf. 0074 ahead-of-.
[37] P. Shahrestani, M. Rafei, The impact of oil price shocks on Tehran Stock Exchange [53] C. Jin, F.S. Tsai, Q. Gu, B. Wu, Does the porter hypothesis work well in the emission
returns: application of the Markov switching vector autoregressive models, Res. trading schema pilot? Exploring moderating effects of institutional settings, Res.
Pol. 65 (2020), https://doi.org/10.1016/j.resourpol.2020.101579. Int. Bus. Fin. 62 (2022), https://doi.org/10.1016/j.ribaf.2022.101732.
[38] C. Li, M. Umair, Does green finance development goals affects renewable energy in [54] H D Liu, F Huang, J Huang, Measuring the coordination decision of renewable
China, Renew. Energy 203 (2023) 898–905, https://doi.org/10.1016/j. energy as a natural resource contracts based on rights structure and corporate
renene.2022.12.066. social responsibility from economic recovery, Resour. Pol. 78 (9) (2022) 102915,
https://doi.org/10.1016/j.resourpol.2022.102915.

10