Resources Policy 62 (2019) 427–436

Contents lists available at ScienceDirect

Resources Policy
journal homepage: www.elsevier.com/locate/resourpol

The role of stock market and banking sector development, and renewable T
energy consumption in carbon emissions: Insights from G-7 and N-11
countries
Muhammad Wasif Zafara, Syed Anees Haider Zaidib, Avik Sinhac, Ayfer Gediklid, Fujun Houb,∗
a
School of Management and Economics, Beijing Institute of Technology, South-Zhongguancun Street, Beijing, 100081, PR China
b
School of Management and Economics, Department of Management Science & Engineering Beijing Institute of Technology, Beijing, 100081, China
c
Department of General Management and Economics, Goa Institute of Management, India
d
Department of Economics, Istanbul Medeniyet University, Turkey

A R T I C LE I N FO A B S T R A C T

Keywords: This study probes the role of disaggregated financial development and renewable energy in carbon emissions by
Carbon emissions incorporating gross fixed capital formation and economic growth in the function of carbon emissions. The fi-
Stock market nancial development is measured through the stock market and banking sector development. We also examine
Banking development the validity of the EKC hypothesis, using the data of G-7 and N-11 countries spanning from 1990 to 2016. The
Renewable energy
integration properties of the considered variables are examined through second generation unit roots tests. The
Economic growth
Lagrange Multiplier (LM) bootstrap panel cointegration method has confirmed the long-run equilibrium re-
lationship among the variables for all the four models used. The long-run elasticity results suggest that re-
newable energy increases environmental quality by reducing carbon emission intensity for both groups of panel
countries. Banking development index decreases carbon emissions in G-7 countries, while increases carbon
emissions in N-11 countries. Similarly, stock market development index increases carbon emissions in G-7
countries, while decreases in N-11 countries. Overall, economic growth and fixed capital formation impede
environmental quality by accelerating the intensity of carbon emissions. This study suggests policy implications
based on the empirical results for both groups of countries.

1. Introduction and social development along with an increasing level of energy de-
mand. In this regard, economies across the globe have made pertinent
Over the past few decades, environmental degradation and global efforts to accelerate the share of renewable energy to improve con-
warming have become two predominant threatening and debatable servation and efficiency of energy. Currently, fossil fuel-based sources
global issues. The major cause of these two problems is considered represent the majority of energy sources, i.e. is about 80% of total
greenhouse gas emissions. Among other greenhouse gasses emissions, global energy used worldwide (World Bank, 2017). Prior literature
CO2 emissions are most notorious due to negative effects on environ- (Jahangir Alam et al., 2012; Shahbaz et al. 2013, 2015; Wang et al.,
ment and human health. It is reported that about 76.7% of total GHG 2016; Li et al., 2016a, b) concluded that energy-led economic growth
emissions consists of carbon emissions (IPCC, 2013). Mitigating carbon has a role to play in the level of CO2 emissions or environmental quality
emissions has become a major research area for economists and pol- in numerous contexts. However, recent literature (Al-Mulali et al.,
icymakers because these emissions create many obstacles in ways of 2015a,b,c; Franco et al., 2017; Bölük and Mert, 2015; Al-Mulali et al.,
economic growth. The countries, in order to attain higher growth rates, 2016; Sinha et al., 2017, 2018) has agreed upon identifying the varying
consume a large quantity of nonrenewable energy which results in effects of renewable and non-renewable energy sources on environ-
perk-up of carbon emissions. It is imperative for the policymakers to mental quality through having control over CO2 emissions. The re-
know the multiple reasons behind high emissions and environmental newable energy (i.e. wind, biomass, solar, hydro, and geothermal) is
degradation (Gokmenoglu and Taspinar, 2016). supposed to be the key energy sources and a solution to control the
Over the last couple of decades, the world has witnessed economic climate change issue, along with ensuring high energy security

∗
Corresponding author.
E-mail addresses: [email protected] (M.W. Zafar), [email protected] (S.A.H. Zaidi), [email protected] (A. Sinha),
[email protected] (A. Gedikli), [email protected] (F. Hou).

https://doi.org/10.1016/j.resourpol.2019.05.003
Received 15 April 2019; Received in revised form 1 May 2019; Accepted 7 May 2019
0301-4207/ © 2019 Elsevier Ltd. All rights reserved.
M.W. Zafar, et al. Resources Policy 62 (2019) 427–436

(Menyah and Wolde-Rufael, 2010). Most of the countries are trying to relationships between growth and other determinants (e.g. energy
switch towards clean energy sources to diminish energy dependence consumption, CO2 emissions, and financial development) are subject to
and to control global climate change (Bölük and Mert, 2014). The time variations. For example, relationships vary in the development
adoption of renewable energy sources in comparison of non-renewable phase as well as stable conditions. These economies are important to
sources can reduce CO2 emissions to a great level. study because of their growth and emission patterns. Also, these
Along with energy consumption, financial development also con- countries are growth leaders and their actions are being followed by
tributes significantly to environmental degradation. The literature of other emerging and developing countries. The other sample panel of
environmental economics states dichotomous impacts of financial de- our study is Next 11 or N-11 countries. It is a cluster of eleven emerging
velopment on environmental quality. Financial development expands, economies, namely Bangladesh, Egypt, Indonesia, Iran, Mexico, Ni-
when strong financial institutions stimulate the provision of the higher geria, Pakistan, Philippines, Turkey, South Korea, and Vietnam. As per
loan disbursement with lower financing costs to their customers. This Goldman Sachs (2015), these economies possess great potential to grow
easy availability of loans helps the firms to create a demand for their to compete with existing leading economies and may outperform nu-
products or services, and thereby, stimulating industrialization. merous key developed markets. But these N-11 countries are facing
However, this industrialization entails consumption of commercial en- many challenges, which primarily include energy poverty, banking and
ergy, which in turn deteriorates the environmental quality through taxation laws, environmental pollution, and corruption (University of
ambient air pollution. Initial rise in the financial development results in Calgary, 2018). Therefore it is important to study N-11 countries to
rise in the level of emissions. However, further rise in financial devel- suggest policies for attaining sustainable development.
opment might instigate the rise in the energy innovations, which might This study contributes in a way that it includes N-11 and G-7
result in energy generation from alternate and cleaner sources countries in a single study to enlarge the scope. The financial sector and
(Tamazian et al., 2009), and thereafter the emission level might start to environmental concerns are different in these two panels therefore, the
come down. This phenomenon has been identified by several re- determinants and results are also expected to be different. Hence, both,
searchers, following Environmental Kuznets Curve (EKC) hypothesis positive and negative, scenarios are covered in this study. The policy
framework. Zaidi et al. (2019) has analyzed the role of financial de- implications for both types of panels are presented in this study. From
velopment in reducing CO2 emissions in Asia Pacific Economic Co- methodological perspective, (i) This study divides the financial devel-
operation countries, and the similar results were obtained by Zafar et al. opment system into banking and stock market development, and ex-
(2019) for Organization for Economic Co-operation and Development plores the dynamic impact of each component on environment quality
countries. Studies by Sadorsky (2010), Ulusoy and Demiralay (2017), for G-7 and N-11 countries, (ii) renewable energy consumption is added
Nasir et al. (2019), Shahbaz et al. (2019) have given further evidences in the function of CO2 emission to check its impact on carbon emission,
on the role of financial development in shaping the environmental (iii) this study also checks the existence of EKC hypothesis (iii) The
degradation in diverse contexts. Shahbaz and Sinha (2019) have given a updated econometric tests are applied to check unit root, cointegration,
detailed review on this aspect, while reviewing the literature of EKC long-run relationships and directions of relationships among the vari-
hypothesis. Therefore, authors have incorporated financial develop- ables of both panels.
ment as an explanatory variable in their empirical model. Remaining part of this article presents a literature review in section
Existing studies measured financial development by aggregating II, modeling and data in section III, results and discussion in section IV
different proxies like “domestic credit issued to the private sector” and/ and conclusion in section V with policy implications.
or “FDI” (Sadorsky, 2011; Shahbaz et al., 2015, 2019). Few studies
combined the financial intermediation and stock market variables to 2. Literature review
represent financial development (e.g. Abbasi and Riaz, 2016). Sadorsky
(2011) used (a) the ratio of bank deposits to GDP, (b) Percentage of 2.1. Renewable energy-environment quality nexus
deposits to GDP, and (c) Percentage of liquid liabilities to GDP as de-
terminants of financial development. Another group of researchers used Recent studies have disaggregated energy consumption into re-
(a) stock market capitalization to GDP (%), and (b) stocks turnover to newable and nonrenewable energy consumption and examined their
GDP (%) as determinants of financial development. The prior literature impact on environment quality by employing different econometric
presents mixed results about the role of financial development in en- techniques. First time, Richmond and Kaufmann (2006) disaggregated
vironmental degradation. Most of the studies support that credit issued the energy consumption into renewable and non-renewable sources and
to the private sector decreases the level of emissions, whereas financial checked the impact of each source of energy consumption with carbon
development through FDI increases the level of CO2 emissions. By far, emission by considering the level of economic development to de-
we have not come across any study, that has checked the impact of termine the turning point between economic growth and carbon
financial development on carbon emissions through disaggregated emission for thirty-six OCED and Non-OCED countries over the period
analysis, and there lies the role of our study. There might be different from 1973 to 1997 by using panel methodology. Their empirical results
impacts of banking sector development on CO2 emissions as compared showed a negative relation between renewable energy and carbon
to the role of stock market development. In pursuit of empirical vali- emissions and turning point exists for OECD countries. A few years
dation of our argument, we check to the disaggregate effects of banking later, Apergis et al. (2010) checked the linkages among renewable en-
sector development and the stock market on CO2 emissions for N-11 ergy, nuclear energy, GDP, and carbon emissions by employing panel
and G-7 countries. error correction model for nineteen developing and developed countries
The purpose of this study is to examine the role of stock market & spanning the period from 1984 to 2007. Their findings suggested that
banking sector development, and renewable energy consumption in renewable energy increases carbon emissions while nuclear energy
mitigating the carbon emissions in the two panels of G-7 and N-11 decreases. Later, for OECD countries, Shafiei and Salim (2014) checked
countries. Group of seven (G-7) is a cluster of seven highly developed the significance of renewable energy with environment quality by
economies of the world including Canada, France, Germany, Italy, controlling the population, urbanization, and industrialization in the
Japan, the United Kingdom, and the United States. These economies EKC model. The long-run estimation results obtained through Aug-
possess 58% of the global net wealth (IMF, 2018). This is a forum to mented Mean Group (AMG) approach and concluded that renewable
discuss economic and financial issues among the top major industrial energy is very useful in improving environmental quality. However, by
countries. These economies are not free from challenges. They also face using the panel cointegration methods for MENA countries, Farhani and
the problem of environmental degradation, rising energy consumption Shahbaz (2014) concluded that environmental quality decreases with
and trade wars, etc. According to (Balcilar et al., 2018), the the consumption of renewable energy. The similar results found by

428
M.W. Zafar, et al. Resources Policy 62 (2019) 427–436

Bölük and Mert (2014) by employing fixed-effect method for European countries by using the time period 1980–2011. The panel methodology
Union countries during the 1990–2008 period. On the other hand, results demonstrated that all indicators of banking development influ-
Bölük and Mert (2015) checked the influence of energy production on ence energy consumption positively. For OECD countries, Ulusoy and
carbon emissions for Turkey by using the ARDL approach. Their find- Demiralay (2017) checked the shock of stock market development on
ings indicated that renewable energy is helpful in Turkey to reduce the oil and electricity by using data from 1996 to 2011. Their results in-
intensity of carbon emission. Later, Bilgili et al. (2016) found the pre- dicated that stock market liquidity and capitalization are important
sence of the EKC hypothesis for OECD countries by using FMOLS and determinants of oil and electricity demand in OECD countries.
confirmed that renewable energy decreases the carbon emissions. Al- Aforementioned studies clearly indicate that financial development
Mulali and Ozturk (2016) used panel methodology and determined EKC by measuring with the stock market and banking development have a
hypothesis in the presence of renewable energy and energy prices for 27 significant positive impact on energy consumption for single and panel
advanced countries. Results of their study stated that renewable energy countries. Tamazian et al. (2009) checked the link between stock
has a negative and significant impact on carbon emissions, and results market and environment degradation along with other control variables
also depict the validity of the EKC hypothesis. Conversely, by using by using panel data methodology for Brazil, Russia, India, and China
FMOLS and DOLS techniques, Bulut (2017) argued that energy con- (BRIC) over the period 1992–2004. This study used the market value of
sumption from clean sources (renewable energy) does not help to re- stocks as a proxy of stock market development. Their results revealed
duce carbon emission in Turkey over the period 1970–2013. Recently, that environment degradation decrease with the increase in stock
Sinha and Shahbaz (2018) determined the link between renewable market development. By using the data of China, Zhang (2011) studied
energy and CO2 emissions by using the EKC framework for India and the influence of stock market development on environmental quality
their finding suggested that CO2 emissions decrease with the increase in along with other indicators of financial development in the augmented
renewable energy and also supports the EKC hypothesis. However, Jin function of carbon emissions. The study results found that carbon
and Kim (2018) found a bidirectional causal relationship between re- emissions Granger cause stock market development in China. Abbasi
newable energy, nuclear energy, and CO2 emissions for 30 countries by and Riaz (2016) also tested this nexus for Pakistan over the period
using Ganger causality test. Also found the same results for 25 devel- 1970–2011 by using ARDL technique. They used a percentage of stock
oping countries by using the VECM Granger causality approach. In market capitalization to economic growth (GDP) and percentage of
contrast, Inglesi-Lotz and Dogan (2018) probed the role of energy stock traded/turnover to GDP to measure the stock market develop-
consumption in environmental quality for top-10 big electricity pro- ment. Their findings indicated that stock market turnover accelerates
ducing countries in Sub-Saharan Africa. They reported a negative link the carbon emission in Pakistan, but stock market capitalization does
between renewable energy and CO2 emissions. By using the quantile not affect the level of carbon emission. By using the data of G20
regression, Chen and Lei (2018) also reached the same conclusion for countries, Paramati et al. (2017) investigated the role of stock market
30 countries over the period of 1980–2014. An empirical study of Chen growth, renewable energy, and FDI on CO2 emissions for the period
et al. (2019) showed the interesting results for China by using the time 1991–2012. Later, they divided the G20 countries into developed and
period 1980–2014. Their results could not establish the EKC hypothesis developing countries. They used Westerlund and Fisher-type Johansen
by excluding renewable energy in the carbon emission function, but cointegration methods and reported the long-run association between
with the inclusion of renewable energy shows the presence of the EKC the variables. They reported that stock market capitalization improves
hypothesis. environment quality in developed countries but reduce the environ-
ment quality in developing countries. Moreover, environment quality
2.2. Financial development-environmental quality nexus improves with energy consumption from renewable sources in devel-
oped and developing countries. Recently, Paramati at el. (2018) ex-
The contribution of financial development in the quality of the amined the relationship between stock market and environment quality
environment has been discussed in many studies. Majority of the studies by incorporating economic growth, energy efficiency, and population
were conducted on emerging economies. For example, Sadorsky (2010) density as an additional determinant for developed and emerging
investigated the link between energy consumption and the stock market economies over the period 1992–2011. Their study used five proxies to
by using the GMM approach for 22 emerging economies over the period measure stock market development. However, their results indicate that
1990–2006. The empirical results revealed that stock market indicators stock market activities reduce carbon emissions in developed countries
increase energy consumption. For Central and Eastern European fron- but increase carbon emission in emerging economies.
tier economies, Sadorsky (2011) examined the connection between
banking related variables, stock markets indicators, and energy con- 2.3. Economic growth-environmental quality nexus
sumption over the period 1996–2006. The results reported that energy
consumption increases of banking related variables, but from stock There are ample studies available in literature which have checked
market indicators, just stock market turnover possess a positive influ- the linkage of economic growth with environmental quality and de-
ence on energy consumption. Çoban and Topcu (2013) checked the picted diverse results. The diverse results represented by different stu-
influence of banking and stock market development on energy con- dies mainly dependents upon the sample countries and choice of
sumption for the European Union (EU) countries by using the system- econometric techniques. The country's economic development stages
GMM approach over the period 1990–2011. The long-run estimation have a different impact on the intensity of carbon emissions. This
results indicated that the stock market and banking development in- phenomenon was also checked by Padilla and Serrano (2006), who
crease energy consumption for old member countries of the EU. How- investigated the impact of income inequality across the countries on
ever, energy consumption decreases with the increase in banking de- carbon emissions over the period 1971–1999. Their results explained
velopment in new member countries of the EU. Zhang et al. (2011) that different per capita income level has a different impact on per
examined the stock market scale and efficiency effect on energy con- capita emissions. Lee (2013) checked the growth-emission nexus for G-
sumption by using the Chinese industries’ data over the period 10 countries through fixed effect methodology. Their panel estimation
1992–2009. They used stock market turnover ratio as an efficiency results explained the negative link between growth and emissions.
effect and stock market capitalization as a scale effect. The Granger Hakimi and Hamdi (2016) examined the link of economic growth and
causality results indicated that stock market scale Granger causes en- environment quality and their long-run estimation results indicated
ergy consumption, however, energy consumption Granger causes stock that environment quality increases with economic growth for Morocco.
market efficiency. Aslan et al. (2014) examined the link between Ahmed et al. (2017) probed the determinants of carbon emissions in the
banking development and energy consumption for Middle Eastern long-run for south Asian countries for 1971–2013 period. Their results

429
M.W. Zafar, et al. Resources Policy 62 (2019) 427–436

found that economic growth is the major cause of environmental de- determinant. The nexus of growth and emissions mostly display an
gradation in Bangladesh and Pakistan, while economic growth in- inverted U-shaped pattern, i.e. the results obtained through the
creases environment quality in Nepal and Sri Lanka. STIRPAT model are identical to the EKC Hypothesis. The link of tech-
The environmental and economic growth relationship has widely nological development with environmental degradation is compara-
been discussed through the phenomenon of Environment Kuznets Curve tively simple and straightforward (Hagemann, 2017), which demon-
(EKC) hypothesis, which is a pioneering effort of Grossman and Krueger strates that the technological progress of a country is favorable to
(1995). They suggest environmental pollution and economic growth reduce its environmental pollution. The STIRPAT model has been em-
exhibit a quadratic and inverted U-shaped long-run association, which ployed by many researchers and analysts to evaluate the environmental
is called the EKC hypothesis. Numerous studies, see for example effects of different determinants, see for example, (Tan and Wang,
(Apergis and Ozturk, 2015; Cho et al., 2014; Shahbaz et al., 2013; 2017; Xu and Lin, 2015). Present study uses the following models:
Gokmenoglu and Taspinar, 2016; Li et al., 2016a, b), checked this hy-
pothesis in various contexts and presented inconclusive and mixed re- Model-1
sults. Many studies probed the EKC hypothesis and also validated the
possible presence of EKC hypothesis (Shahbaz et al., 2014; Katircioǧ;lu lnCO2, it = β0 + β1 ln GDPit + β2 ln GDPSit + β3 ln GCFit + β4 ln REit
2014; Farhani et al., 2014; Javid and Sharif, 2016; Awad and + β5 ln BDIit + εit
Abugamos, 2017; Balaguer and Cantavella, 2016; Li et al., 2016a, b; Al-
Mulali et al., 2015a,b,c), while some researchers also reported non- Model-2
existence of EKC (Ali et al., 2017a, b; Farhani and Ozturk, 2015; Liu
et al., 2017; Shahbaz and Sinha, 2019). The countries validating the 1nCO2, it = β0 + β1 ln GDPit + β2 ln GDPSit + β3 ln GCFit + β4 ln REit
EKC hypothesis infer that there will be an automatic enhancement in
+ β5 ln BDIit + β6 (ln RE ∗ ln BDI ) + εit
the environmental quality because it is hard to achieve the Kyoto tar-
gets (Cho et al., 2014).
In the literature, various studies examined the nexus among fi- Model-3
nancial development (banking and stock market development), popu-
ln CO2, it = β0 + β1 ln GDPit + β2 ln GDPSit + β3 ln GCFit + β4 ln REit
lation, and energy consumption. However, there are very few studies
available which investigated the impact of stock market and banking + β5 SMDIit + εit
development on carbon emissions in the framework of EKC hypothesis
by incorporating the renewable energy consumption, and gross capital Model-4
formation in the augmented function of carbon emission. Moreover,
there is rarely any study available in the literature which uses the index ln CO2,it = β0 + β1 ln GDPit + β2 ln GDPSit + β3 ln GCFit + β4 ln REit
for the stock market and banking development by using Principal + β5 ln SMDIit + β6 (ln RE ∗ ln SMDI) + εit
Component Analysis (PCA) and examined their impact on carbon
emissions for N-11, and G-7 countries. Where slope coefficients are shown by β, countries are represented by i,
the period is shown through t (1990–2016), μ denotes the residual and
CO2, it indicates CO2 emissions per capita. Model 1 and 3 show the de-
3. Model development and data collection terminants of the carbon emissions which include per capita gross do-
mestic product (GDP, square of GDP (GDPS), gross fixed capital for-
3.1. Model development mation (GCF), renewable energy consumption (RE), banking
development index (BDI), and stock market development index (SMDI),
This study aims to probe the role of disaggregated financial devel- respectively. In order to find additional insights, we have developed
opment (stock market and banking sector development) and renewable Model 2 and 4, where we add interaction terms between banking de-
energy on carbon emissions by incorporating gross fixed capital for- velopment index and renewable energy consumption, and stock market
mation and economic growth in the function of carbon emissions. This development index and renewable energy consumption, respectively.
study uses the IPAT theoretical model to achieve the aforementioned Before proceeding with the analysis, we have checked for the mul-
objective. IPAT theoretical model introduced by Ehrlich and Holdren ticollinearity in the dataset using Variance Inflation Factor (VIF), and
(1971) to explain the determinants of CO2 emissions. the results reported in Appendix 1 shows the presence of multi-
I = P. A. T (1) collinearity in the data. Therefore, before proceeding with the analysis,
we carried out orthogonal transformation in the data. The results re-
Herein, ‘I’ is the Pollutant intensity; ‘P’ refers to aggregate population; ported in Appendix 1 show that after carrying out this transformation,
‘A’ denotes the economic development, and ‘T’ stands for technological the VIF values have come down, showing the removal of multi-
advancement. Nonetheless, the IPAT equation harbors some flaws, as it collinearity in the dataset. With this transformed dataset, we have
oversimplifies several environmental issues. The IPAT calculus pre- carried out the analysis.
sumes that the elasticities of each independent variable correspond to 1
(Timma et al., 2016). Thus, Dietz and Rosa projected another model the
3.2. Data collection
STIRPAT (Stochastic Impacts Regression on Population, Affluence, and
Technology) for precise calculation of environment influencing factors
In order to find the causal relationships among variables, we use
(Dietz and Rosa, 1997);
time series data of N-11 and G-7 countries ranging from 1990 to 2016.
It = aPtb Atc Ttd et (2) This multivariable framework for empirical analysis includes environ-
mental quality measured with carbon emissions (in metric tons), energy
The “a” is the intercept term, b, c, and d are the powers of environ- consumption from renewable sources measured with a billion kilowatt
mental effects corresponding to P, A, & T, and et is the random sto- hours, economic growth measured with gross domestic product (GDP)
chastic error. The STIRPAT model presents the three important de- taken in US dollars (constant 2010), gross fixed capital formation (GCF)
terminants of carbon emissions namely population, GDP and measured with US dollars (constant 2010). This study converts carbon
technology level in the country. Particularly, the larger size of the po- emissions, gross domestic product, renewable energy consumption, and
pulation is responsible for more emissions as compared to a smaller gross fixed capital formation into per capita form by dividing them with
size. But, the environmental effects of economic growth is a difficult the population. The data of per capita GDP, carbon emissions, and gross

430
M.W. Zafar, et al. Resources Policy 62 (2019) 427–436

fixed capital indicators were downloaded from the web page of world Table 1
bank (World Bank, 2017). The International Energy Agency (IEA, 2017) Cross-sectional dependence.
database is used to collect the data on renewable energy consumption Variable CD-test p-value
for selected countries.
G-7
lnCO2 8.56a 0.000
3.3. Financial development indicator
lnGDP 19.50a 0.000
lnGDPS 19.47a 0.000
Although, numerous studies have used diverse proxies for mea- lnGCF 2.98a 0.000
suring the financial development they all used such proxies in aggregate lnRE 10.08a 0.000
form. The results of different studies with different proxies are also lnBANK 6.15a 0.000
lnStock 6.84a 0.000
unlike. To our research objective, we consider banking and stock
RE*BANK 7.01a 0.000
market development indicators separately to measure the financial RE*STOCK 5.00a 0.000
development, and each indicator measures with four proxies. First, N-11
banking development measures with (a) deposit money bank asset to lnCO2 3.975a 0.0001
lnGDP 34.92a 0.0000
GDP ratio, (b) financial system deposit to GDP ratio, (c) the ratio of
lnGDPS 34.96a 0.0000
liquid liabilities to GDP, and (d) private credit by deposit money bank lnGCF 15.47a 0.0000
and financial institutions. Similarly, stock market development in- lnRE 20.75a 0.0000
dicator is measured with (a) stock market turnover ratio, (b) stock lnBANK 24.01a 0.0000
market capitalization to GDP ratio, (c) stock market value traded to lnSTOCK 18.83a 0.0000
RE*BANK 10.99a 0.0000
GDP ratio, and (d) number of listed companies per 10000 people. This
RE*STOCK 13.26a 0.0000
study constructs a weighted index for banking and stock market de-
velopment by using the above-mentioned proxies to measure both in- Note: a significant value at 1%; b significant value at 5%; c significant value at
dicators. The weighted bank and stock market development indexes are 10%.
constructed by applying principal component analysis (PCA) for both
groups of countries. The use of this new indicator (index) provides Table 2
appropriate information related to stock and backing development in- Panel unit roots test results.
dicators, and remove multicollinearity in the empirical model. The data Variables CIPS CADF
of banking and stock market development indicators are collected from
the World Bank Financial Structure Database (Beck et al., 2000). Level First-difference Level First-difference

G-7
3.4. Results and discussion
lnCO2 −2.106 −4.229a −2.106 −4.229a
lnGDP −2.060 −3.910a −2.060 −3.910a
3.4.1. Cross-sectional dependence lnGDPS −1.895 −3.897a −1.895 −3.897a
We started our empirical analysis by examining the cross-sectional lnGCF −2.087 −3.803a −2.087 −4.336a
lnRE −1.606 −4.081a −1.451 −4.081a
dependence in the time series panel data because estimated results can
lnBANK −1.969 −4.265a −1.849 −4.265a
be biased without controlling the cross-sectional dependence. Earlier RE*BANK −1.948 −4.627a −1.868 −4.627a
studies on environmental-finance nexus have ignored the cross-sec- lnSTOCK −2.408 −3.568a −2.411 −3.568a
tional dependence while estimating the empirical results. The cross- RE*STOCK −2.400 −3.561a −2.428 −3.561a
sectional dependence among the panel countries is very common due to N-11
lnCO2 −1.588 −4.655a −1.456 −4.655a
sharing the board, spillover effect, trade agreements, and financial
lnGDP −2.172 −3.377a −2.174 −3.443a
crises, among others. For this purpose, this study has employed Cross- lnGDPS −2.532 −3.349a −2.093 −3.413a
sectional Dependence (CD) method introduced by Pesaran (2004). lnGCF −2.473 −4.418a −2.432 −4.303a
Table 1 reflects the existence of cross-sectional dependence at 1 percent lnRE −1.703 −3.508a −1.515 −3.508a
lnBANK −2.097 −2.943a −1.809 −3.071a
significance level.
RE*BANK −2.281 −3.582a −2.158 −3.582a
lnSTOCK −1.845 −3.622a −1.687 −3.622a
3.4.2. Unit root test RE*STOCK −1.870 −3.680a −1.719 −3.680a
Further, this study examines the integration properties of the con-
sidered variables using CIPS and CADF methods of unit root introduced Note: a significant value at 1%; b significant value at 5%; c significant value at
by Pesaran (2007). These methods also consider cross-sectional de- 10%.
pendence and heterogeneity in the data sets. Table 2 shows the results
of CIPS and CADF which indicate unit root problems at level. But at first Table 3
difference, all the underlying data become stationary. LM-Bootstrap cointegration test.
(The null hypothesis is cointegration among the variables).
3.4.3. LM bootstrap cointegration test lm statistic bootst p-val
As unit root tests confirm the integration order of the variables, the
G-7
next step is to scan the long-run equilibrium relationship among the
CO2=GDP+GDPS+GCF+RENEWABLE+BANK 7.656 0.926
variables. In doing so, the LM bootstrap cointegration approach, in- CO2=GDP+GDPS+GCF+RENEWABLE+stock 7.036 0.968
troduced by Westerlund and Edgerton (2007), is applied which takes N-11
into account cross-sectional dependence and the issue of heterogeneity CO2=GDP+GDPS+GCF+RENEWABLE+BANK 10.388 0.936
CO2=GDP+GDPS+GCF+RENEWABLE+stock 10.335 0.914
in the panel data for all the four models. The null hypothesis explains
that variables are cointegrated in long-run while alternative hypothesis
shows the absence of such cointegration.
countries.
The estimated results of LM bootstrap cointegration method are
presented in Table 3. The outcomes support the rejection of the null
hypothesis for all the four models used to test the determinants of the 3.4.4. Long-run estimation
CO2 emissions over the period of 1990–2016 for the G-7 and N-11 panel This study examines long-run output elasticities through CUP-FM

431
M.W. Zafar, et al. Resources Policy 62 (2019) 427–436

Table 4
CUP-FM and CUP-BC results for G-7 countries.
Variables Model 1 Model 2 Model 3 Model 4

CUP-FM CUP-BC CUP-FM CUP-BC CUP-FM CUP-BC CUP-FM CUP-BC

b a b a a a a
lnGDP 0.6707 (2.692) 0.6918 (2.857) 1.4051 (2.575) 1.6100 (10.680) 1.4561 (7.179) 0.4564 (2.867) 0.9041 (7.345) 0.8876a (7.024)
lnGDPS −0.0066a −0.0028a −0.0509a −0.0620a −0.0609a 0.0186a (11.050) −0.0679a (3.343) −0.0688a (3.224)
(8.063) (8.268) (6.116) (21.008) (13.445)
lnGCF 0.1601b (2.3067) 0.1790a (3.0738) 02137a (3.921) 0.1843a (3.459) 0.2188a (6.293) 0.2624a (9.954) 0.2727a (7.010) 0.2907a (7.931)
lnRE −0.0410a −0.0287a 0.4674b (2.694) 0.5450a (13.074) −0.0279a (8.443) −0.0119a −0.3443a −0.3416a
(7.336) (6.745) (8.782) (17.976) (17.823)
lnBANK −0.0213a −0.0731a −2.0862a −2.4563a
(5.527) (7.869) (4.514) (20.154)
lnSTOCK 0.0115a (13.762) 0.0040a (20.016) 1.3005a (10.516) 1.2256a (9.938)
RE*BANK −0.1206a −0.1417a
(7.598) (31.269)
RE*STOCK 0.0815a (2.734) 0.0774a (3.361)

Note: t-statistics are presented in the parenthesis. Moreover, a significant value at 1%; b significant value at 5%; c significant value at 10%.

Table 5
CUP-FM and CUP-BC results for N-11 countries.
Variables Model 1 Model 2 Model 3 Model 4

CUP-FM CUP-BC CUP-FM CUP-BC CUP-FM CUP-BC CUP-FM CUP-BC

a a a a a a a
lnGDP 2.1383 (9.995) 2.0191 (9.588) 1.9143 (5.066) 1.7332 (9.880) 1.9508 (14.373) 1.8503 (13.852) 1.8403 (14.293) 1.8218a (14.003)
lnGDPS −0.0801a −0.0724a −0.0606a −0.0495a −0.0489a −0.0442a −0.0453a −0.0444a
(16.454) (16.138) (13.311) (29.915) (22.073) (21.693) (22.500) (22.088)
lnGCF 0.0022a (4.617) 0.0022a (4.616) 0.0229a (7.247) 0.0093a (12.223) 0.0816a (12.245) 0.0613a (10.922) 0.0801a (12.212) 0.0609a (10.998)
lnRE −0.0523a −0.0454a −0.3841a (8.049) −0.2460a −0.0066a −0.0020a 0.0287a (6.021) 0.0220a (6.482)
(15.484) (14.517) (12.224) (13.422) (12.512)
lnBANK 0.2300a (6.319) 0.2358a (6.516) 1.4560a (4.650) 0.8631a (4.575)
lnSTOCK −0.0249a −0.0256a −0.1670a (6.283) −0.1331a (5.428)
(34.336) (34.582)
RE*BANK 0.0738a (7.649) 0.0380a (18.620)
RE*STOCK −0.0086a −0.0065a
(43.385) (42.028)

Note: t-statistics are shown in the parenthesis. Moreover, a significant value at 1%; b significant value at 5%; c significant value at 10%.

and CUP-BC methods. Both the methods which were introduced by positively related to carbon emissions. This explains that gross capital
proposed by Bai and Ng (2006) and Bai et al. (2009) are more effective formation increases carbon emissions. Results intimate that the esti-
approaches among the available techniques because these control the mated coefficients of gross capital formation for G-7 countries are sig-
unobserved non-linearity and cross-sectional dependence. Moreover, nificantly greater than the estimated coefficients for N-11 countries.
CUP-FM and CUP-BC provide robust results even if the integration This difference conveys the larger capital formation established in G-7
order is mixed I (1) and I (0). Furthermore, both estimators have good countries. The similar results are suggested by Abid (2017) for Eur-
size and power values for our samples. The output results of long-run opean Union and MEA (Middle East & African) countries, but these
elasticities are shown in Tables 4 and 5 by using the data of G-7 and N- results are different to those reported by Shahbaz et al. (2018) for G-7
11 panel countries, respectively. countries.
The results demonstrate positive coefficient of GDP and the negative The effect of banking sector development on CO2 emissions is ne-
coefficient for squared GDP (GDPS) in relation to carbon emissions for gative for G-7 and positive for N-11 countries. A 1% improvement in
both groups of countries. This result validates the EKC hypothesis for G- baking development index improves the environmental quality by
7 and N-11 countries. The results infer that in the early period of 0.0213% in G-7 countries, while it increases 0.2300% emissions in N-11
growth, the CO2 emissions start rising because the economies consume countries. The negative impact of the coefficient estimates shows the
both the polluted energy resources along with natural resources. fact that banking development has achieved the maturity level in de-
However, when the economic growth of an economy reaches a certain veloping economies. It infers that the banking sector in these countries
threshold level, a further boost in growth initiates a reduction in allocate their financial resources to the advanced energy-efficient
emissions gradually. The same findings were reported by Ali et al. technologies and also to the green and clean environmental friendly
(2017b) for Malaysia and Al-Mulali et al. (2016) for Kenya. projects. This tendency leads to enhance energy efficiency along with
In both the panels, renewable energy consumption possesses nega- production and reduces emissions levels. A strong banking development
tive statistical link with carbon emissions. This shows that environment structure increases the accessibility of banking loan for the develop-
quality improves with the increase in renewable energy consumption ment of energy efficient technology. The cost of financing is also de-
for both groups of countries. Energy policymakers and en- creased with a strong banking development structure. Moreover, in-
vironmentalists have strongly believed that environment quality can be dustries need long-term financing for the development of energy
improved with the increase of renewable energy consumption. Similar efficient technology which in turn may reduce the carbon emission
findings are reported by Shafiei and Salim (2014) for OECD countries, intensity. The similar results are reported Tamazian at al. (2009), who
Sinha and Shahbaz (2018) for India, and Al-Mulali and Ozturk (2016) found a negative and significant relationship between domestic money
for twenty-seven advance economies. bank asset to GDP and carbon emission for BRICS countries. However,
For both groups of countries, gross fixed capital formation is for the case of N-11 countries, the significant positive connection

432
M.W. Zafar, et al. Resources Policy 62 (2019) 427–436

between banking development index and carbon emission implies that interaction term between stock market development index and re-
carbon emission increase with banking development activities in this newable energy consumption on CO2 emissions. The results indicate
region of countries. The results indicate that financial development that renewable energy increases CO2 emissions for G-7 countries, while
through banking sector hurts the environment. The banking sector is it decreases carbon emission for N-11 countries. This positive interac-
highly connected with the provision and allocation of financial re- tion influence on emissions for the panel of G-7 suggests that stock
sources at low costs to their customers, firms, and also to households for market development in these economies are unable to bring innovation
high energy consumption bit-ticket items. Additionally, auto financing which can augment the share of renewable energy. Another possible
is the profitable production of consumer financing and due to the lack explanation is that the listed firms in these nations still largely use fossil
of environmental regulations, the most outdated and old technology fuel energy in the production houses which lead to an increase in
enhances energy consumption and emissions, which boosts environ- carbon emissions. On the other hand, it shows that growing economies
mental degradation. Our results support the finding reached by (Zhang, meet their energy needs by a reduction in the share of renewable energy
2011), who pointed out that most of the Chinese companies are sup- sources in the energy mix. This enhances the energy consumptions of
ported by the bank loans. The bank's loans facilitate the investment non-renewable energy sources and increases CO2 emissions.
scales of the firms which boost economic activities and carbon emis-
sions. Çoban and Topcu (2013) demonstrated that banking develop-
ment index increases energy consumption which results in increase 3.4.5. Granger causality results
carbon emission. The long-run elasticities results provide a one-way relationship be-
Our study also checks the results of the interaction term of renew- tween the variables. For policymakers, it is imperative to know the two-
able energy consumption and banking development index on CO2 way causal relationships among the variables to develop the policy
emissions for both groups of countries. The interaction findings state implications. In this pursuit, this study employs Dumitrescu and Hurlin
that renewable energy leads to curtailing CO2 emission in G-7 countries panel heterogeneous causality method suggested by Dumitrescu and
but raises the level of emissions in N-11 countries. Hurlin (2012), and their results are presented in Tables 6 and 7 for G-7
Next, results show that stock market development index possesses a and N-11 countries, respectively.
significant negative influence on carbon emissions in G-7 countries. On As we can notice that a feedback relationship exists between growth
the other hand, it bears a positive and significant influence on carbon and emissions of CO2 for G-7 countries. For the case of N-11 countries,
emission in N-11 countries. A 0.011% rise in carbon emission is due to a economic growth possesses unidirectional causal effects on CO2 emis-
1% increase stock market development index for G-7 countries, while a sions. For G-7 countries, a unidirectional causality runs from gross fixed
−0.024% decrease in carbon emission is due to a 1% increase stock capital to CO2 emissions but results could not establish any causal re-
market development index for N-11. These results demonstrate that lationship between these two variables for N-11 countries. The Granger
stock market activities cause to increase carbon emission in G-7 coun- causality results indicate that unidirectional causality also runs from
tries. This implies that stock exchange policies toward the betterment of renewable energy to CO2 emissions for G-7 countries; however, there is
environment quality may not be effective in the G-7 countries. The no causal relationship exists between renewable energy and CO2
policymaker and other concerned government authorities should focus emissions for N-11 countries. The connection between banking devel-
on establishing environmental friendly policies related to the stock opment index and CO2 emissions is bidirectional for G-7 countries, but
exchanges by adopting the energy efficient greener technology. for N-11 countries, the unidirectional causal effect is confirmed of
The similar results are concluded by Abbasi and Riaz (2016) and banking development on CO2 emissions. Meanwhile, results indicate
Paramati et al. (2017), they found a positive connection between stock that a feedback relationship exists between stock market development
market capitalization and environmental degradation. For the case of index and CO2 emissions across the panel of countries. Similarly, eco-
N-11 countries, carbon emissions decrease with enhancement in stock nomic growth and gross fixed capital formation possess bidirectional
market development index. The insights of Table 6 infer that stock relationship for both groups of countries. Our results indicate that
market development in the N-11 countries is adversely related to en- unidirectional link exists from renewable energy to economic growth
vironmental quality. This shows that in N-11 countries the stock market for G-7 countries, reverse results exist for N-11 countries. These results
firms initiated environmental green projects policies and ensure its support the growth hypothesis and the conservative hypothesis for G-7
adaptation in the listed stock exchanges firms. This tendency will en- and N-11 countries, respectively. There is unidirectional relationship
hance the adaptation of green technologies in developed economies, exist from stock market development index to economic growth for G-7,
resulting in reduces energy consumption and emissions. Our results are and bidirectional relationship for N-11, respectively. Likewise, a similar
in the line of Paramati et al. (2018) who found that stock market in- relationship exists between banking development index and economic
dicators decrease carbon emission. For the case of N-11 countries, stock growth across the panel. However, gross capital formation Granger
exchange policies are very effective to achieve a sustainable environ- cause renewable energy in G-7 countries. A feedback relationship found
ment. between gross capital formation and banking development index for in
In the same way, this study also checks the outcomes of the G-7. In contrast, unidirectional relationships exist from gross capital
formation to banking development index in G-7 countries. The

Table 6
Dumitrescu and Hurlin heterogeneous panel causality test results for G-7 Countries.
Independent variables

lnCO2 lnGDP lnGDPS lnGCF lnRE lnBANK lnSTOCK

lnCO2 2.749a (0.006) 2.703a (0.006) 1.796c (0.072) 3.646a (0.000) 4.001a (0.000) 3.323a (0.000)
lnGDP 2.448b (0.014) 0.101 (0.919) 2.495b (0.012) 1.719c (0.085) 4.573a (0.000) 4.381a (0.000)
lnGDPS 2.370b (0.017) 0.049 (0.960) 2.487b (0.0129) 1.683c (0.092) 0.824 (0.409) 4.379a (0.000)
lnGCF 0.251 (0.801) 4.287a (0.000) 4.355a (0.000) 1.995b (0.0460) 2.517b (0.0118) 1.941c (0.0521)
lnRE 1.919c (0.055) 1.618 (0.105) 1.606 (0.1082) 0.768 (0.4421) 2.630*** (0.0085) −0.150 (0.880)
lnBANK 4.859a (0.000) 0.827 (0.408) 4.504a (0.000) 4.646a (0.000) 4.923a (0.000) 7.361a (0.000)
lnSTOCK 1.940c (0.052) 0.065 (0.947) 0.097 (0.9220) 1.123 (0.2614) 0.995 (0.3193) 5.576a (0.000)

Note: a significant value at 1%; b significant value at 5%; c significant value at 10%.

433
M.W. Zafar, et al. Resources Policy 62 (2019) 427–436

Table 7
Dumitrescu and Hurlin heterogeneous panel causality test results for N-11 Countries.
Independent variables

lnCO2 lnGDP lnGDPS lnGCF lnRE lnBANK lnSTOCK

b c
lnCO2 1.985 (0.047) 1.828 (0.067) 1.186 (0.235) 1.030 (0.302) 1.110 (0.266) 1.932c (0.053)
lnGDP 0.305 (0.760) 0.158 (0.873) 2.466b (0.013) 0.486 (0.626) 3.307a (0.000) 1.703c (0.088)
lnGDPS 0.153 (0.877) 0.135 (0.892) 2.306b (0.021) 0.371 (0.710) 3.112a (0.001) 1.725c (0.084)
lnGCF 0.242 (0.808) 3.803a (0.000) 3.591a (0.000) 0.619 (0.535) 0.920 (0.357) 3.526a (0.000)
lnRE 0.635 (0.524) 3.143a (0.001) 3.240a (0.001) 0.177 (0.859) 0.674 (0.500) 0.652 (0.514)
lnBANK 2.199b (0.027) 6.587a (0.000) 6.521a (0.000) 8.107a (0.000) 0.829 (0.406) 1.126 (0.259)
lnSTOCK 3.214a (0.001) 2.955a (0.003) 2.993a (0.0028) 1.841c (0.065) 0.883 (0.377) 1.324 (0.185)

Note: a significant value at 1%; b significant value at 5%; c significant value at 10%.

relationship between gross capital formation and stock market devel- stock market and carbon emission for G-7 and N-11 countries. First, for
opment index is unidirectional for G-7 but bidirectional for N-11 the G-7 countries the development of the stock market and CO2 emis-
countries, respectively. Our results indicate that renewable energy sions is significantly positive, refers to the presence of conventional
Granger cause banking development index and in response, banking production methods due to the inefficiency of institutions. The con-
development also Granger cause renewable energy in G-7 countries. cerned authorities along with policymakers in the G-7 authorities
Similarly, banking development index and stock marketed development should focus on strong institutional setup so that to induce and adopt
index Granger cause each other in the short-run for N-11 countries. the green energy efficient technology in the region. The advanced en-
ergy efficient technology along with fresh knowledge reduces energy
consumption and will enhance environmental quality by low emissions.
4. Conclusion and policy implication Second, the financial and non-financial incentives should be provided
by the government authorities and policymaker. The relationship be-
The link between the stock market and banking development in- tween carbon emission and the stock market is negative and significant
dicators with energy consumption is well documented in the energy- for N-11 countries. The authorities should establish and implement
finance literature. However, few studies have investigated the effects of greener technologies in all stock exchange listed firms and should focus
the stock market and banking sector development on carbon emissions. and enhance shares in clean energy i.e. renewable energy sources to
This study aims to check the role of disaggregated financial develop- mitigate emissions. All such initiatives will surely reduce emissions and
ment and renewable energy consumption in CO2 emissions by con- will enhance environmental quality. We can also notice a similar re-
trolling the effect of gross fixed capital and economic growth in the lationship between banking development and carbon emission for both
framework of the EKC hypothesis. To achieve the aforementioned ob- groups of countries. The relationship between banking development
jective, this study uses G-7 and N-11 countries data over the period and carbon emissions is significantly negative for G-7 countries while
1990–2016. For empirical analysis, this present study uses second positive for N-11 countries.
generation unit root test to examine the stationary properties of the Financial development of banking sector may facilitate firms and
analyzed variables. The panel LM bootstrap cointegration method is customers through the provision of low rate funds for the latest energy-
employed to check the long-run relationship between the mentioned efficient technologies. Our empirical finding suggests that banking
variables. Moreover, the long-run elasticities between the variables sector development is closely connected with economic growth and also
estimate with CUP-FM and CUP-BC methods. it mitigates environmental quality. The mechanism of financial re-
The results of the panel cointegration test confirm the long-run sources allocation should be controlled by the authorities through the
linkages between the underlying variables. The long-run elasticities of central bank, and the banking sector should also monitor those firms
CO2 emissions with respect to economic growth and gross capital for- which accept debts so that to ensure the environmental friendly in-
mation activities are positive and significant and these results also va- vestments. The firms which do not abide by environmental standards
lidate the EKC hypothesis for across the panel countries. However, re- and causes environmental degradation should be suited to the court for
newable energy increases the environment quality by reducing the punishment and also by increasing the interest rate-allocated loans. The
intensity of CO2 emissions from the environment. Banking development government should encourage the renewable energy sector so that to
decreases CO2 emissions in G-7 countries while increasing the emissions enhance energy efficiency and environmental quality. Research and
in N-11 countries. We can notice that stock market activities increase development is another step to create a certain lifetime of profit rather
CO2 emissions in G-7 however decreases in N-11 countries. The casual than wastage of financial resources.
results are also divergent between the variables for both groups of In the future, the researchers can also investigate the financial
countries. openness and financial liberalization with environmental quality in the
The existence of the EKC hypothesis suggests that a higher level of framework of the EKC hypothesis. Moreover, future studies can address
economic growth increase the environment quality by reducing the the structural breaks in the panel data and include these structure
carbon emission intensity from the environment. The policy makers of breaks in the empirical analysis because the subprime mortgage crisis
these economies develop such policies which increase economic directly affects the financial sector and stock market development
growth. Renewable energy is shown to have a decreasing effect on across the work. Future studies also compare the high income, middle
carbon emission intensity in both groups of the panel. This implies that income, and lower-income countries results and can be useful to draw
these economies increase the share of renewable energy in the energy an effective policy to control the environment population.
mix policy. The policymakers should develop such policies which in-
crease the investment in the renewable energy sector by giving the tax
relief and financial incentive to investors. Renewable energy con- Declarations of interest
sumption decreases the intensity of carbon emissions and also helps in
achieving sustainable economic goals. Moreover, consuming renewable None.
energy also decrease the dependence on conventional fossil fuel energy.
The results of this study show the diverse relationships between the

434
M.W. Zafar, et al. Resources Policy 62 (2019) 427–436

Acknowledgment of China [No. 71571019].

This study is sponsored by The National Natural Science Foundation

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.resourpol.2019.05.003.

Appendix 1. Multicolinearity statistics

Countries Variables Before transformation After transformation

VIF Tolerance VIF Tolerance

G-7 lnGDP 928.89 0.0011 1.19 0.8421

lnGDPS 675.40 0.0015 1.43 0.6997
lnGCF 38.57 0.0259 1.54 0.6504
lnRE 6.47 0.1545 1.36 0.7340
lnBANK 5.68 0.1760 1.76 0.5696
lnSTOCK 4.83 0.2070 1.16 0.8592
RE*BANK 73.61 0.0136 1.37 0.7301
RE*STOCK 68.17 0.0147 1.05 0.9494
N-11 lnGDP 480.72 0.0021 1.07 0.4837
lnGDPS 376.73 0.0027 1.56 0.6392
lnGCF 32.99 0.0303 1.09 0.9157
lnRE 4.57 0.2187 1.69 0.5910
lnBANK 2.84 0.3527 1.99 0.5036
lnSTOCK 6.90 0.1450 1.79 0.5600
RE*BANK 77.93 0.0128 1.45 0.6914
RE*STOCK 71.73 0.0139 1.20 0.8354

References Bai, Jushan, Ng, Serena, 2006. Evaluating latent and observed factors in macroeconomics
and finance. J. Econom. 131, 507–537.
Bai, Jushan, Kao, Chihwa, Ng, Serena, 2009. Panel cointegration with global stochastic
Abbasi, F., Riaz, K., 2016. CO2 emissions and financial development in an emerging trends. J. Econom. 149 (1), 82–99.
economy: an augmented VAR approach. Energy Policy 90, 102–114. Balaguer, Jacint, Cantavella, Manuel, 2016. Estimating the environmental Kuznets Curve
Abid, M., 2017. Does economic, financial and institutional developments matter for en- for Spain by considering fuel oil prices (1874-2011). Ecol. Indicat. 60 Elsevier
vironmental quality? A comparative analysis of EU and MEA countries. J. Environ. Ltd:853–59.
Manag. 188 (2), 183–194. Balcilar, Mehmet, Zeynel Abidin, Ozdemir, Ozdemir, Huseyin, Shahbaz, Muhammad,
Ahmed, K., Rehman, M.U., Ozturk, I., 2017. What drives carbon dioxide emissions in the 2018. The renewable energy consumption and growth in the G-7 countries: evidence
long-run? Evidence from selected south Asian countries. Renew. Sustain. Energy Rev. from historical decomposition method. Renew. Energy 126, 594–604.
70, 1142–1153. Beck, Thorsten, Demirgüç-Kunt, Aslı, Ross, Levine, 2000. A new database on financial
Al-Mulali, U., Ozturk, I., 2016. The investigation of environmental Kuznets Curve hy- development and structure,’ world bank economic review. http://www.worldbank.
pothesis in the advanced economies: the role of energy prices. Renew. Sustain. org/en/publication/gfdr/data/financial-structure-database.
Energy Rev. 54, 1622–1631. Bilgili, Faik, Koçak, Emrah, Bulut, Ümit, 2016. The dynamic impact of renewable energy
Al-Mulali, Usama, Ozturk, Ilhan, Lean, Hooi Hooi, 2015a. The influence of economic consumption on CO2 emissions: a revisited environmental Kuznets Curve approach.
growth, urbanization, trade openness, financial development, and renewable energy Renew. Sustain. Energy Rev. 54, 838–845.
on pollution in europe. Nat. Hazards 79 (1), 621–644. Bölük, Gülden, Mert, Mehmet, 2014. Fossil & renewable energy consumption, GHGs
Al-Mulali, Usama, Saboori, Behnaz, Ozturk, Ilhan, 2015b. Investigating the environ- (greenhouse gases) and economic growth: evidence from a panel of EU (European
mental Kuznets Curve hypothesis in Vietnam. Energy Policy 76, 123–131. Union) countries. Energy 74 (C), 439–446.
Al-mulali, Usama, Weng-wai, Choong, Sheau-ting, Low, Hakim, Abdul, 2015. Bölük, Mert, 2015. The Renewable Energy, Growth and Environmental Kuznets Curve in
Investigating the environmental Kuznets Curve ( EKC ) hypothesis by utilizing the Turkey : an ARDL Approach, vol. 52 Elsevier Ltd 587–95.
ecological footprint as an indicator of environmental degradation. Ecol. Indicat. 48, Bulut, Umit, 2017. The impacts of non-renewable and renewable energy on CO2 emis-
315–323. sions in Turkey. Environ. Sci. Pollut. Control Ser. 24 (18), 15416–15426
Al-Mulali, Usama, Adebola Solarin, Sakiru, Ozturk, Ilhan, 2016. Investigating the pre- Environmental Science and Pollution Research.
sence of the environmental Kuznets Curve (EKC) hypothesis in Kenya: an auto- Chen, Wenhui, Lei, Yalin, 2018. The impacts of renewable energy and technological in-
regressive distributed lag (ARDL) approach. Nat. Hazards 80 (3), 1729–1747. novation on environment-energy-growth nexus: new evidence from a panel quantile
Alam, Jahangir, Mohammad, Ismat Ara Begum, Buysse, Jeroen, Van Huylenbroeck, regression. Renew. Energy 123, 1–14.
Guido, 2012. Energy consumption, carbon emissions and economic growth nexus in Chen, Yulong, Wang, Zheng, Zhong, Zhangqi, 2019. CO2emissions, economic growth,
Bangladesh: cointegration and dynamic causality analysis. Energy Policy 45, renewable, and non-renewable energy production and foreign trade in China. Renew.
217–225. Energy 131, 208–216.
Ali, Ghaffar, Ashraf, Aqdas, Khalid Bashir, Muhammad, Cui, Shenghui, 2017a. Exploring Cho, C.-H., Chu, Y.-P., Yang, H.-Y., 2014. An environment Kuznets Curve for GHG
environmental Kuznets Curve (EKC) in relation to green revolution: a case study of emissions: a panel cointegration analysis. Energy Sources B Energy Econ. Plan. Policy
Pakistan. Environ. Sci. Policy 77, 166–171. 9 (2), 120–129.
Ali, Wajahat, Abdullah, Azrai, Azam, Muhammad, 2017b. Re-visiting the environmental Çoban, Serap, Topcu, Mert, 2013. The nexus between financial development and energy
Kuznets Curve hypothesis for Malaysia : fresh evidence from ARDL bounds testing consumption in the EU: a dynamic panel data analysis. Energy Econ. 39, 81–88.
approach. Renew. Sustain. Energy Rev. 77 (September 2015), 990–1000. Dietz, T., Rosa, E a, 1997. Effects of population and affluence on CO2 emissions. Proc.
Apergis, Nicholas, Ozturk, Ilhan, 2015. Testing environmental Kuznets Curve hypothesis Natl. Acad. Sci. U.S.A. 94 (January), 175–179.
in asian countries. Ecol. Indicat. 52, 16–22. Dumitrescu, Elena Ivona, Hurlin, Christophe, 2012. Testing for granger non-causality in
Apergis, Nicholas, Payne, James E., Menyah, Kojo, Wolde-Rufael, Yemane, 2010. On the heterogeneous panels. Econ. Modell. 29 (4), 1450–1460.
causal dynamics between emissions, nuclear energy, renewable energy, and eco- Ehrlich, Paul R., Holdren, John P., 1971. Impact of population growth. Science 171,
nomic growth. Ecol. Econ. 69 (11), 2255–2260. 1212–1217.
Aslan, Alper, Apergis, Nicholas, Topcu, Mert, 2014. Banking development and energy Farhani, Sahbi, Ozturk, Ilhan, 2015. Causal relationship between CO2 emissions, real
consumption: evidence from a panel of middle eastern countries. Energy 72, GDP, energy consumption, financial development, trade openness, and urbanization
427–433. in Tunisia. Environ. Sci. Pollut. Res. 1–15.
Awad, A., Abugamos, H., 2017. Income-carbon emissions nexus for Middle East and north Farhani, Sahbi, Shahbaz, Muhammad, 2014. What role of renewable and non-renewable
africa countries: a semi-parametric approach. Int. J. Energy Econ. Policy 7 (2), electricity consumption and output is needed to initially mitigate CO2emissions in
152–159. MENA region? Renew. Sustain. Energy Rev. 40, 80–90.

435
M.W. Zafar, et al. Resources Policy 62 (2019) 427–436

Farhani, Sahbi, Chaibi, Anissa, Rault, Christophe, 2014a. CO2emissions, output, energy emerging economies. Energy Policy 38 (5), 2528–2535.
consumption, and trade in Tunisia. Econ. Modell. 38, 426–434 Elsevier B.V. Sadorsky, Perry, 2011. Financial development and energy consumption in central and
Franco, Sainu, Venkata Ravibabu, Mandla, Ram Mohan Rao, K., 2017. Urbanization, eastern european frontier economies. Energy Policy 39, 999–1006.
energy consumption and emissions in the Indian context a review. Renew. Sustain. Shafiei, Sahar, Salim, Ruhul A., 2014. Non-renewable and renewable energy consumption
Energy Rev. 71, 898–907 (January). Elsevier. and CO2 emissions in OECD countries: a comparative analysis. Energy Policy 66,
Gokmenoglu, Korhan, Taspinar, Nigar, 2016. The relationship between CO2 emissions, 547–556.
energy consumption, economic growth and FDI: the case of Turkey. J. Int. Trade Shahbaz, M., Sinha, A., 2019. Environmental Kuznets Curve for CO2 emission: a literature
Econ. Dev. 25, 706–723. survey. J. Econ. Stud. 46 (1), 106–168.
Goldman Sachs, 2015. N-11 equity portfolio [online]. Available:” 2015. http://www. Shahbaz, Muhammad, Adebola Solarin, Sakiru, Mahmood, Haider, Mohamed, Arouri,
goldmansachs.com/gsam/docs/funds_international/brochures_and_sales_aids/fund_ 2013. Does financial development reduce CO2 emissions in Malaysian economy? A
literature/advisor_brochure_n-11_en.pdf. time series analysis. Econ. Modell. 35, 145–152.
Grossman, Gene M., Krueger, Alan B., 1995. Economic growth and the environment. Q. J. Shahbaz, Muhammad, Khraief, Naceur, Salah Uddin, Gazi, Ozturk, Ilhan, 2014.
Econ. 110 (2), 353–377. Environmental Kuznets Curve in an open economy: a bounds testing and causality
Hagemann, Andreas, 2017. Cluster-robust bootstrap inference in quantile regression analysis for Tunisia. Renew. Sustain. Energy Rev. 34 Elsevier:325–36.
models. J. Am. Stat. Assoc. 112 (517), 446–456. Shahbaz, Muhammad, Nasreen, Samia, Abbas, Faisal, Anis, Omri, 2015. Does foreign
Hakimi, Abdelaziz, Hamdi, Helmi, 2016. Trade liberalization, FDI in fl ows , environ- direct investment impede environmental quality in high-, middle-, and low-income
mental quality and economic Growth : a comparative analysis between Tunisia and countries? Energy Econ. 51, 275–287.
Morocco. Renew. Sustain. Energy Rev. 58, 1445–1456. Shahbaz, Muhammad, Balsalobre, Daniel, Hussain Shahzad, Syed Jawad, 2018. The in-
IEA, International Energy Agency, 2017. Key World Energy Statistics. fluencing factors of CO2 emissions and the role of biomass energy consumption:
IMF, 2018. A Guide to Committees, Groups, and Clubs. 2018. https://www.imf.org/en/ statistical experience from G-7 countries. Environmental Modeling and Assessment.
About/Factsheets/A-Guide-to-Committees-Groups-and-Clubs#G7. Environmental Modeling & Assessment 1–19.
Inglesi-Lotz, Roula, Dogan, Eyup, 2018. The role of renewable versus non-renewable Shahbaz, M., Balsalobre-Lorente, D., Sinha, A., 2019. Foreign direct Investment–CO2
energy to the level of CO2emissions a panel analysis of Sub- saharan africa's ßig 10 emissions nexus in Middle East and North African countries: importance of biomass
electricity generators. Renew. Energy 123, 36–43 Elsevier Ltd. energy consumption. J. Clean. Prod. 217, 603–614.
IPCC, 2013. Climate Change 2013: the Physical Science Basis. The Contribution of Sinha, A., Shahbaz, M., 2018. Estimation of environmental Kuznets Curve for CO2
Working Group I to the Fifth Assessment Report of the IPCC. Cambridge University emission: role of renewable energy generation in India. Renew. Energy 119, 703–711.
Press. Sinha, A., Shahbaz, M., Balsalobre, D., 2017. Exploring the relationship between energy
Javid, Muhammad, Sharif, Fatima, 2016. Environmental Kuznets Curve and financial usage segregation and environmental degradation in N-11 countries. J. Clean. Prod.
development in Pakistan. Renew. Sustain. Energy Rev. 54, 406–414. 168, 1217–1229.
Jin, Taeyoung, Kim, Jinsoo, 2018. What is better for mitigating carbon emissions – re- Sinha, A., Shahbaz, M., Sengupta, T., 2018. Renewable energy policies and contradictions
newable energy or nuclear energy? A panel data analysis. Renew. Sustain. Energy in causality: a case of Next 11 countries. J. Clean. Prod. 197, 73–84.
Rev. 91, 464–471. Tamazian, Artur, Juan, Piñeiro Chousa, Krishna, Chaitanya Vadlamannati, 2009. Does
Katircioǧlu, Salih Turan, 2014. Testing the tourism-induced EKC hypothesis: the case of higher economic and financial development lead to environmental degradation:
Singapore. Econ. Modell. 41, 383–391. evidence from BRIC countries. Energy Policy 37 (1), 246–253.
Lee, Jung Wan, 2013. The contribution of foreign direct investment to clean energy use, Tan, Xue Ping, Wang, Xin Yu, 2017. Dependence changes between the carbon price and
carbon emissions and economic growth. Energy Policy 55, 483–489. its fundamentals: a quantile regression approach. Appl. Energy 190, 306–325.
Li, Tingting, Wang, Yong, Zhao, Dingtao, 2016a. Environmental Kuznets Curve in China: Timma, Lelde, Zoss, Toms, Blumberga, Dagnija, 2016. Life after the financial crisis.
new evidence from dynamic panel analysis. Energy Policy 91, 138–147 Elsevier. Energy intensity and energy use decomposition on sectorial level in Latvia. Appl.
Lin, Boqiang, Omoju, Oluwasola E., Ngozi, M Nwakeze, Jennifer, U Okonkwo, Ebenezer, Energy 162, 1586–1592.
T Megbowon, 2016b. Is the environmental Kuznets Curve hypothesis a sound basis Ulusoy, Veysel, Demiralay, Sercan, 2017. Energy demand and stock market development
for environmental policy in Africa ? J. Clean. Prod. 133 Elsevier Ltd:712–24. in OECD countries: a panel data analysis. Renew. Sustain. Energy Rev. 71, 141–149.
Liu, Xuyi, Zhang, Shun, Bae, Junghan, 2017. The impact of renewable energy and agri- University of Calgary, 2018. N11 Countries. Energy Education 2018. https://
culture on carbon dioxide Emissions : investigating the environmental Kuznets Curve energyeducation.ca/encyclopedia/N11_countries.
in four selected ASEAN countries. J. Clean. Prod. 164, 1239–1247. Wang, Shaojian, Zhou, Chunshan, Li, Guangdong, Feng, Kuishuang, 2016. “CO2, eco-
Menyah, Kojo, Wolde-Rufael, Yemane, 2010. CO2emissions, nuclear energy, renewable nomic growth, and energy consumption in China's provinces: investigating the spa-
energy and economic growth in the US. Energy Policy 38 (6) Elsevier:2911–15. tiotemporal and econometric characteristics of China's CO2emissions. Ecol. Indicat.
Nasir, M.A., Huynh, T.L.D., Tram, H.T.X., 2019. Role of financial development, economic 69, 184–195.
growth & foreign direct investment in driving climate change: a case of emerging Westerlund, Joakim, Edgerton, David L., 2007. A panel bootstrap cointegration test. Econ.
ASEAN. J. Environ. Manag. 242, 131–141. Lett. 97, 185–190.
Padilla, Emilio, Serrano, Alfredo, 2006. Inequality in CO2 emissions across countries and World Bank, 2017. World development indicators. Available at: https://data.worldbank.
its relationship with income inequality: a distributive approach. Energy Policy 34 org/indicator.
(14), 1762–1772. Xu, Bin, Lin, Boqiang, 2015. Carbon dioxide emissions reduction in China's transport
Paramati, S.R., Mo, D., Gupta, R., 2017. The effects of stock market growth and renewable sector: a dynamic VAR (Vector autoregression) approach. Energy 83, 486–495.
energy use on CO2 emissions: evidence from G20 countries. Energy Econ. 66. Zafar, M.W., Saud, S., Hou, F., 2019. The impact of globalization and financial devel-
Paramati, Sudharshan Reddy, Alam, Md Samsul, Apergis, Nicholas, 2018. The role of opment on environmental quality: evidence from selected countries in the
stock markets on environmental degradation: a comparative study of developed and Organization for Economic Co-operation and Development (OECD). Environ. Sci.
emerging market economies across the globe. Emerg. Mark. Rev. 35, 19–30. Pollut. Control Ser. 26 (13), 13246–13262.
Pesaran, M.H., 2004. General diagnostic tests for cross section dependence in pa- Zaidi, S.A.H., Zafar, M.W., Shahbaz, M., Hou, F., 2019. Dynamic linkages between glo-
nelsgeneral diagnostic tests for cross section dependence in panels. In: CESifo balization, financial development and carbon emissions: evidence from Asia Pacific
Working Paper Series No. 1229. University of Cambridge. Economic Cooperation countries. J. Clean. Prod. 228, 533–543.
Pesaran, M. Hashem, 2007. A simple panel unit root test in the presence of cross-section Zhang, Yue-jun, 2011. The impact of financial development on carbon Emissions : an
dependence. J. Appl. Econom. 22 (2), 265–312. empirical analysis in China. Energy Policy 39 (4), 2197–2203.
Richmond, Kaufmann, Robert K., 2006. “Is there a turning point in the relationship be- Zhang, Y.J., Jing, L.F., Hao, R.C., 2011. Impact of China's stock market development on
tween income and energy use and/or carbon Emissions ? 56, 176–189. energy consumption: an empirical analysis. Energy Procedia 5, 1927–1931.
Sadorsky, Perry, 2010. The impact of financial development on energy consumption in

436