THE JOURNAL OF ENERGY

AND DEVELOPMENT

Paul Adjei Kwakwa,

“The Effects of Natural Resource Extraction
and Renewable Energy Consumption on
Carbon Dioxide Emissions in Sub-Saharan Africa,”
Volume 47, Number 2

Copyright 2022
 THE EFFECTS OF NATURAL RESOURCE
EXTRACTION AND RENEWABLE ENERGY
CONSUMPTION ON CARBON DIOXIDE EMISSIONS
IN SUB-SAHARAN AFRICA

Paul Adjei Kwakwa*

Introduction

T he debate on the effect natural resources and energy usage have on economic
growth and development has dominated academic discussion for a long time.
The abundance of natural resources in a country has the potential to propel eco-
nomic growth and development through exports for foreign exchange earnings, the
promotion of local industrialization, creation of jobs, and attraction of foreign
direct investment. They are also noted for enhancing infrastructural and human
capital development.1 The opposing argument is that resource-rich countries do
not perform well because of high levels of corruption, political instability,
rent-seeking behavior, and lower investment in human capital.2 Energy consump-
tion is also useful for economic growth because it is a crucial input for productive
activities in all economies. Every sector of all economies, namely, the industrial
sector, the transport sector, educational sector, and health sector, among others,

*Paul Adjei Kwakwa, Ph.D., is a senior lecturer in Economics with the University of Energy and
Natural Resources, Sunyani, Ghana. The author is interested in issues related to sustainable
development. His research has focused heavily in the areas of energy conservation, water conservation,
and the energy-economic growth-environment nexus. Dr. Kwakwa’s recent articles have been
published in Journal of Energy and Development, OPEC Energy Review, International Journal of
Energy Economics and Policy, Journal of Rural and Industrial Development, Renewable and
Sustainable Energy Review, and Journal of Environmental Management.

The Journal of Energy and Development, Vol. 47, Nos. 1-2

Copyright # 2022 by the International Research Center for Energy and Economic Development
(ICEED). All rights reserved.
195
196 THE JOURNAL OF ENERGY AND DEVELOPMENT

cannot function effectively without energy. It is for this reason that governments
and policy makers in all countries are eager to ensure there is enough energy sup-
ply for residential and industrial activities. However, there are concerns that
increased energy usage especially fossil fuel is associated with higher carbon diox-
ide (CO2) emissions.3 CO2 concentration in the atmosphere is a key factor contrib-
uting to global warming and climate change.4 Fossil fuel energy is thus regarded
as not environmentally friendly because its combustion generates more CO2 emis-
sions thereby polluting the environment.
Despite the opposing arguments of the growth and development effects of natu-
ral resources, their extraction has been increasing in recent years. According to the
World Resources Forum, there was approximately a 25-percent growth in the
extraction of global natural resources between 1980 (40 billion tons) and 2005 (58
billion tons).5 Also, it has been reported that natural resources consumption saw an
increment from 27 billion tons in 1970 to 92 billion tons in 2017 with a projection
of between 170 billion tons to 184 billion tons by 2050.6 In light of this, research-
ers are drawing the attention of the academic community, policy makers, and envi-
ronmentalists to the effects that natural resource extraction has on environmental
sustainability. The process of extracting natural resources (for economic gains)
involves the destruction of forest resources and land, water pollution, and degrada-
tion of the environment. Moreover, the extraction and transformation of natural
resources into valuable products involve the use of energy. Thus, the more energy
used for natural resource extraction and its processing implies more CO2 emis-
sions.7 If the extraction and processing of natural resources is energy efficient or
environmentally conscious the environmentally destructive effect of natural
resource extraction is expected to be less and vice versa. It suffices to say that if
the extraction and the processing of natural resources uses renewable energy (RE)
chances are the destructive effects in terms of CO2 emissions will be lower. This is
because renewable energies (REs) are environmentally friendly since they are asso-
ciated with lower CO2 emissions.8
Now, let us delve deeper into this issue for our geographical focus area—
Africa. Africa is well endowed with natural resources such as forest resources,
mineral deposits, and oil reserves. While the economic growth effect of natural
resources has been empirically examined for countries on the continent,9 little
research has assessed the environmental destruction effect of natural resources.10
Some studies have reported that natural resource extraction seriously affects the envi-
ronment via increased CO2 emissions.11 What appears to be common in the extant
literature is that the analyses have assumed that natural resource extraction affects
CO2 emissions in a linear manner. There is the possibility that the relationship
between the two may not be linear since earnings from natural resources beyond a
certain level can trigger lower environmental destruction. Thus, the initial destructive
effects of natural resources extraction may be overturned after a certain threshold.
Through the accumulation of revenue from natural resources, players in the sector,
 RESOURCES, RENEWABLE ENERGY, & CO2 EMISSIONS 197

policy makers, and the government may be in a position to correct the damage that
has been created. After many economic gains, players in the industry and policy
makers may be conscious of the environment and, as a result, energy-efficient tech-
nology or stricter regulations may be instituted to ensure that the extraction of natural
resources does not negatively affect the environment. Moreover, beyond certain
gains, extractive firms can acquire energy-efficient equipment for their activities.
Various energy situations on the continent, especially for Sub-Saharan Africa
(SSA), have been explored by researchers including the demand for renewable
energy,12 production of renewable energy,13 and access to clean energy.14 Largely,
analyses on renewable energy by researchers in SSA are undertaken because
renewable energy still forms a smaller percentage of electricity generation in the
region and its consumption and adoption is comparatively lower.15 Environmental
concerns have also led researchers to investigate the CO2 emissions effect of RE in
many countries/regions.16 Nevertheless, such analysis remains limited for African
countries.17 Almost all existing works on the renewable energy-CO2 emissions
nexus have focused on the direct impact. Although the extractive sector is equally
energy-dependent, the interactive effect that renewable energy usage may have on
the natural resources-CO2 emissions nexus is yet to be explored.
Environmental management has become a daunting task for many developing
countries, particularly SSA nations. The environmental problems in the subregion
range from water pollution and deforestation to waste management issues. Air pol-
lution, particularly CO2 emissions, in the subregion has been of concern to
researchers in recent times. Although CO2 emissions from the region are compara-
tively lower than others, the increasing trend over the years makes many question
the region’s readiness to help fight the climate change menace and global warming,
whose effects are worse for countries in the region. Thus, although China, India,
and United States are among the leading countries when it comes to CO2 emis-
sions, countries in Sub-Saharan Africa are disadvantaged since many of the people
in the region survive on agriculture-related activities, which are also greatly
affected by climate variables. Accordingly, both low and high CO2 emitting coun-
tries need to be committed to reducing the level of CO2 emissions as the world
seeks to attain sustainable development. With CO2 emissions in Sub-Saharan
Africa increasing from 399 thousand kilotons (kt) in 1990 to 823 thousand (kt) in
2018 it becomes crucial to identify the propelling factors to help design appropriate
policies to reduce emissions.18 Although there are growing investigations on the
factors of CO2 emissions in the subregion and elsewhere,19 analysis of natural
resources and renewable energy consumption’s influence on CO2 emissions in the
subregion is limited. The analysis also has been limited to a linear relationship.
Thus, this current work looks at the effect of natural resource extraction and
renewable energy (RE) consumption on CO2 emissions in Sub-Saharan Africa
(SSA). The three objectives this study sets to achieve are: (a) to empirically ascer-
tain the linear and non-linear effects of natural resource extraction on CO2
198 THE JOURNAL OF ENERGY AND DEVELOPMENT

emissions in SSA; (b) to assess the effect RE consumption has on emissions of

CO2 in SSA; and (c) to analyze the interactive effect of RE consumption on the
natural resources-CO2 emissions nexus in SSA.
By way of adding to the body of knowledge, this paper examines the interactive
effect between natural resource extraction and RE consumption on CO2 emissions.
Second, it explores the quadratic effect of natural resource extraction on CO2 emis-
sions. Previous studies20 have assumed the link between natural resources and CO2
emissions is linear, which may not be the case. Third, this research employs the
quantile regression technique in addition to the ARDL technique for analysis.
Fourth, it provides empirical evidence from a region that has not received much
attention in the energy and environmental economics literature.

Review of Related Literature

The threat of climate change and global warming to human life and the ecosys-
tem has been well documented in the literature. Climate change has led to
increased temperatures, flooding, droughts, and increased pests and diseases. The
effect is that food production is threatened, and the health of individuals is also
negatively affected.21 Among efforts undertaken to address the danger of climate
change have included the need to reduce the emissions of greenhouse gasses
(GHGs) as enshrined in the United Nation’s sustainable development goal 13. CO2
emissions account for about 65 percent of GHGs emitted. It is for this reason that
studies have investigated the drivers of CO2 emissions.22
Analyses of the drivers of CO2 emissions have been conducted from different
perspectives including the angle of sociologists, geographers, climate scientists,
environmentalists, engineers, and economists. Economic investigations on the sub-
ject matter, in many cases, have aimed at establishing the CO2 emissions effect on
economic variables. Researchers have relied on theories like the environmental Kuz-
nets curve (EKC) hypothesis,23 STIRPAT model,24 and the pollution haven effect25
to assess how variables like income, population pressure, and energy affect CO2
emissions. Recently, researchers have devoted attention to how natural resources
affect CO2 emissions.26
Theoretically, income’s effect on CO2 emissions is ambiguous. A rise in indi-
viduals’ incomes increases their purchasing power which drives greater demand.
Peoples’ appetite for energy-powered products like vehicles, televisions, air condi-
tioners, washing machines, etc., increases with income.27 This increases energy
consumption, which, in turn, leads to more CO2 emissions. In some developing
countries, as P. Kwakwa et al.28 have identified, citizens’ proof of a change in eco-
nomic status when they earn more income is easily recognizable when they acquire
certain items like private vehicles, which they now regard as a necessity, when
prior to their rise in income were previously deemed as discretionary. Such
 RESOURCES, RENEWABLE ENERGY, & CO2 EMISSIONS 199

situations often lead to an upsurge in CO2 emissions due to the associated increase
in energy consumption which is usually dominated by fossil fuels.
To the contrary, some have argued an increase in income enables individuals to
afford energy-efficient products and devices. This argument is based on the premise
that newer versions and models of energy-using products that are more energy effi-
cient can be relatively expensive and, thus, can be afforded by the higher income
earners. Once individuals’ incomes increase, they can replace their energy-intensive
and inefficient devices with energy-efficient ones, thereby reducing CO2 emis-
sions.29 Thus, whether income will increase or reduce CO2 emissions will, among
other things, be dependent on whether the energy consumption reduction effect out-
weighs the increasing effect. The EKC hypothesis also concludes the relationship
between income and CO2 emissions is concave downward (an inverted-U) in shape.
The reason is that at the initial stages of economic growth and development CO2
emissions increase because authorities pay little attention to the quality of the envi-
ronment. However, after a higher level of economic growth and development is
achieved emphasis is placed on environmental quality. Thus, individuals’ desire for
a better environment increases and the countries policy makers and authorities enact
laws to protect the environment and citizens can afford energy-efficient products
leading to lower levels of CO2 emissions.30
Population pressure has become a global phenomenon. The pressure that comes
from population expansion negatively affects environment quality as CO2 emis-
sions rise. Population growth means more forest resources are cleared31 for resi-
dential and commercial usages. It also means an extension of farmlands to produce
enough food. In addition, it is associated with increased vehicular movements. The
clearance of forests and extension of agricultural farmlands reduces carbon seques-
tration. Also, the demand for transport services increases fossil fuel consumption
thereby leading to higher CO2 emissions. However, according to the compact city
theory, as countries become more populous and towns and cities become more
urbanized, the pressure is capable of promoting economies of scale thereby reduc-
ing CO2 emissions.32
Increased energy consumption releases CO2 emissions that pollute the environ-
ment. Because of this, many advocate for reducing reliance on fossil fuel sources
of energy to concentrate on the development and usage of renewable energy, which
is clean and as such its usage for domestic and industrial purposes will be environ-
mentally beneficial.33
Extracting natural resources like mineral deposits, drilling for oil, quarrying
activities, and felling of timber, etc., can negatively affect the environment through
the energy consumption of machines used for carrying out these activities. There is
also a CO2 emissions effect when some of the resources are being processed
through several activities that rely heavily on energy inputs.34 The loss of forests
to mining mineral deposits and also through logging activities reduces the size of
the forest area that could have sequestrated CO2. There is the possibility that
200 THE JOURNAL OF ENERGY AND DEVELOPMENT

financial gains from natural resources can reduce CO2 emissions when firms in the
industry acquire energy-efficient machines and state authorities are enabled to fight
environmental degradation.
Research on CO2 drivers like income, population pressure, renewable energy, and
natural resources for individual countries and groups of countries abound. Different
estimation techniques have been employed for these analyses. However, the results
have not been conclusive. Studies on single countries include F. Amuakwa-Mensah
and P. Adom, K. Li and B. Lin, and M. Koondhar et al.35 Some have used panel
data for regional or bloc analysis for African countries,36 European countries,37 Asian
countries,38 developing and developed countries,39 OECD countries,40 and BRICS
countries.41 The estimation techniques have largely been fully-modified ordinary least
squares (FMOLS), autoregressive distributed lag (ARDL), generalized methods of
movements (GMM), and dynamic ordinary least squares (DOLS) to assess a linear
association between CO2 emissions and its factors42 with few assessing the
non-linear association between CO2 emissions and its factors.43
The empirical findings reported have not been the same. Regarding income, a
positive impact on CO2 emissions was recorded by K. Li and B. Lin,44 P. Kwakwa
et al.,45 and C. Nwani et al.,46 while S. Adams et al.,47 obtained a negative outcome.
The works of S. Aboagye et al.48 and H. Mahmood et al.49 confirmed a concave
connection between CO2 emissions and income while C. Nwani et al.50 reported
contrary outcomes and C. Isik et al.51 had mixed results. K. Li and B. Lin52 and C.
Nwani et al.53 reported urbanization increases CO2 emissions, which is contrary to
the findings of P. Sadorsky,54 H. Ali et al.,55 and S. Adams et al.56 Also, M. Koond-
har et al.57 and P. Sheng and X. Guo58 found population reduces CO2 emissions but
P. Sadorsky59 found population increases CO2 emissions. Generally, empirical
research60 has found a negative relationship between renewable energy on CO2
emissions (meaning as renewable energy increases, emissions decrease) unlike the
case of natural resource development, which is found to increase CO2 emissions.61
What remains lacking in existing empirical studies is that the non-linear rela-
tionship between natural resources and CO2 emissions has not been assessed yet.
Also, the moderating role of RE consumption in the nexus between natural resour-
ces and CO2 emissions is lacking. Therefore, this study bridges these gaps.

Methods

Theoretical Framework and Empirical Modeling: The theory of green eco-

nomics62 underscores the necessity to balance human activities and environmental
sustainability. Thus, the theory emphasizes that efforts to meet societal needs
should not destroy the quality of the environment. This means that economies
should not only be interested in extracting natural resources for their financial gains
without recourse to the environment. Using renewable energy to meet societal
 RESOURCES, RENEWABLE ENERGY, & CO2 EMISSIONS 201

needs is likely to help achieve this agenda of green economics. Based on this, the
study relies on two other related theories: the Environmental Kuznets curve
hypothesis as discussed by G. Grossman and A. Krueger63 and the Stochastic
Impacts by Regression on Population, Affluence, & Technology (STIRPAT) model
by T. Dietz and E. Rosa64 to meet the objectives. The STIRPAT model posits that
the level of environmental degradation impact (I) is a function of population pres-
sure (P), affluence or economic growth (A), and technology (T), which can be
mathematically stated as:

I 5 a : PB : Al : T s : v (1)

The variables a, b, l, s, and v stand for parameters to be estimated. CO2 emis-

sions constitute environmental problems because they contribute to global warming
and changes in the climate. Therefore, in this model CO2 emissions replace Impact
(I). Total population (POP) and the urbanization share in total population (UB)
will represent population pressure (P). Affluence is represented by income (GDP).
Based on the argument by T. Dietz and E. Rosa,65 renewable energy consumption
(RE) and natural resources extraction (NAR) would represent technology. As a
result, we get equation (2):
b l g s
CO2 5 aðPOPÞ ðUBÞ ðGDPÞ ðNARÞ ðREÞd v (2)

where g, d, and e parameters are to be estimated in addition to those already

explained. Transforming equation (2) into natural logarithm for a panel data results in:

LCO2it 5 ` 1 bLPOPit 1 lLUBit 1 gLYit 1 sLNARit 1 dLREit 1 #it 1 eit

(3)

where i and t denote the individual countries and time dimension correspondingly;
# and e represent the county effect and error term, respectively, L is the symbol
for natural logarithm; and the rest remain the same. Based on the argument that
renewable energy consumption can moderate the environmental impact of natural
resource extraction, an interactive term between the two (LNARxLRE) is intro-
duced to equation (4):

LCO2it 5 ` 1 bLPOPit 1 lLUBit 1 gLYit 1 sLNARit 1 dLREit

(4)
1 uðLNARxLREÞit 1 #it 1 eit

When u .0 it means that renewable energy positively moderates the emissions

effect of natural resource extraction by increasing s upwards and vice versa. To
also ascertain if the effect natural resource extraction has on CO2 emissions is
non-linear, the square term of natural resource (LNAR2) is added to equation (4):
202 THE JOURNAL OF ENERGY AND DEVELOPMENT

LCO2it 5 ` 1 bLPOPit 1 lLUBit 1 gLYit 1 sLNARit 1 rLNAR2it

(5)
1 dLREit 1 #it 1 eit

If s .0 and r . it implies natural resource extraction monotonically increases

CO2 emissions and vice versa.
If s .0 and r , it implies natural resource extraction initially increases CO2
emissions up to a certain level beyond which CO2 emissions are reduced with natu-
ral resources extraction. This gives a concave-shaped relationship between the two,
and the opposite holds.
The study factors in the EKC hypothesis and introduces the square of income
into equations (3) through (5) to obtain, respectively:

LCO2it 5 ` 1 bLPOPit 1 lLUBit 1 gLYit 1 g2 LYit2 1 sLNARit

(6)
1 dLREit 1 #it 1 eit

LCO2it 5 ` 1 bLPOPit 1 lLUBit 1 gLYit 1 g2 LYit2 1 sLNARit

(7)
1 dLREit 1 uðLNARxLREÞit 1 #it 1 eit

LCO2it 5 ` 1 bLPOPit 1 lLUBit 1 gLYit 1 g2 LYit2 1 sLNARit

(8)
1 rLNAR2it 1 dLREit 1 #it 1 eit

Data Source, Description, and Estimation Procedures: The study makes use
of panel data from the World Bank66 for seven natural resource-rich countries in
Sub-Saharan African, namely, Ghana, South Africa, Botswana, Sudan, Nigeria,
DR Congo, and Angola. The available period also spanned from 1971-2018. CO2
emissions are measured by emission of carbon dioxide (kt), population was mea-
sured by the total population, and income was denoted by the gross domestic prod-
uct (GDP) at constant price (U.S. dollars). Urbanization was measured using the
urban population (as a percentage of total population), natural resources extraction
was represented by total natural resources rent (as a percentage of GDP), and
renewable energy consumption was represented by combustible renewables and
waste. These measurements follow what has been commonly used by previous
studies.
Time series and panel data are found to contain unit roots, which usually leads
to obtaining spurious regression results. Therefore, it has become almost a ritual
for studies that use such data for analysis to conduct unit root tests. If a unit root is
present a variable is usually differenced to remove the unit root. Once the unit root
is absent, the variable becomes stationary. In a panel study, the Im, Pesaran, and
Shin67 and Maddala and Wu68 unit root tests are traditionally used to determine
variables’ stationarity. Nonetheless, it is argued that interdependency among panel
data renders these tests ineffective. In such instances, the Pesaran69 Panel Unit
 RESOURCES, RENEWABLE ENERGY, & CO2 EMISSIONS 203

Root test (cross-sectionally augmented IPS, CIPS) is used to test for the stationar-
ity. This makes it crucial, therefore, to test for the cross-sectional dependence
among the variables prior to testing for the unit root.
Following this is the Pedroni70 cointegration test to ascertain the long-run con-
nection between the dependent variable and regressors. Once cointegration is
established among the variables, the autoregressive distributed lag (ARDL) regres-
sion procedure is employed to determine the effects of population, urbanization,
income, natural resources extraction, and renewable energy consumption on CO2
emissions. The ARDL technique, which generates short-run and long-run coeffi-
cients simultaneously, is chosen for this estimation since it deals with endogeneity
problems associated with panel data. It is also useful when variables for analysis
are stationary at levels and/or first difference and also when the sample size is
small as it holds for this study. Following G. Kollie,71 the basic ARDL model is
often specified as follows:

X
p X
q
Yit 5 lij Yi,tj 1 d9ij Xi,tj 1 mi 1 eit (9)
j1 j50

where Yit symbolizes the dependent variable, Xi,t2j stands for the vector of explana-
tory variables, lij denotes the coefficient on the lags of the dependent variable, d9ij
is the coefficients on the current and lags of the explanatory variables, mi is the
fixed effect, and eit is the error term. A precondition to establish the long-term rela-
tionship among the variables is that they must be cointegrated. An error correction
model (ECM) also is introduced to symbolize short-run deviations that are cor-
rected in the long run. Consequently, equations (6)–(8) are re-parameterized to
account for the ECM to arrive at the panel ARDL model given in the following
equation:

X
p1 X
q1
DlnYit 5 f1 ðlnYi,t1  u91 lnXit Þ 1 lij DlnYi,t1 1 dij DlnXi,tj 1 m1 1 eit
j51 i50

(10)

The variables Y and X in equation (10) remained as already explained, u9i is a

vector of long-run coefficients, lij and d are short-run coefficients, and fi captures
the speed of adjustment, which has to be negatively signed and significant.
Also, the quantile regression analysis is executed to establish the heterogeneous
consequence of natural resources extraction and RE consumption on CO2 emis-
sions as formulated in equation (3). The quantile regression technique accounts
for heterogeneity and non-Gaussian distributions. In addition, quantile regression
estimators provide robust results when data contain outliers and also are skewed.
204 THE JOURNAL OF ENERGY AND DEVELOPMENT

The estimators accommodate unobserved heterogeneity, which facilitates the

assessment of differences in the dependent variable among low, medium, and high
changes. Contrary to this, the Ordinary Least Squares methodology yields esti-
mates representing the effect from an explanatory variable on the “average
country.” As a result, they do not reflect the entire distribution, providing partial
representation of the impact detected by the explanatory variables, unlike the quan-
tile regression. The quantile regression for the panel data therefore is given as:

Yit 5 ai 1 bðqÞxit 1 Uit (11)

where i is the individual countries and t is the time dimension. The dependent vari-
able is Y while x represents the explanatory variables, q denotes the quartile (0 , q
, 1) of the conditional distribution, and a denotes the presence of fixed effects.
The impact of the explanatory variables is allowed to be determined by the quantile
q. Following R. Koenker and G. Bassett,72 the estimation of equation (11) for a
number of quantiles simultaneously is obtained through minimization:

X
T X
n X
m
 
minab wkPqk Yij  ai  bðqk Þ xijÞ 1 lrðaÞ (12)
K1 j1 i1

where Pqk 5 u(q – I(u , 0)) is the piecewise linear quantile loss function provided
by R. Koenker and G. Bassett.73 The weights wk control the relative influence of
the t quantiles (q1, … , qt) on the estimation of the ai parameters.
Moreover, the Dumitrescu and Hurlin74 causality method to inspect the causal
interactions between variables is analyzed. The aim is to arrive at the necessary
policies based on the direction of the causalities, that is, uncovering the unidirec-
tional and bidirectional relationships of studied variables.75 Unlike the vector error
correction model (VECM) Granger causality, the Dumitrescu and Hurlin (DH)
causality test accounts for heterogeneity and cross dependence in panel data. It
also produces a robust estimate even when data is small like those used for this
study. The panel data model can be represented as:

X
p
ðpÞ
X
p
ðpÞ
Yi:t 5 ai 1 di yi:tn 1 bi xi:tn 1 Ui:t (13)
i51 i51

where n is the lag length and x and y denote underlying variables for n
cross-section in t time. The di(r) is the autoregressive parameters and bi(r) is the
regression coefficient across countries.
 RESOURCES, RENEWABLE ENERGY, & CO2 EMISSIONS 205

Results and Discussion

Descriptive Statistic and Correlation Matrix: In table 1 the descriptive statis-

tic and correlation matrix of the variables for the investigation are reported. Over
the study period, the average CO2 emissions among the selected countries has been
66,322 kt with the highest emission being 503,112 kt. While the mean consump-
tion of renewable energy is 55 percent of total energy use, which is an indication
that the selected countries seem to use more renewable energy, the mean natural
resource rent is 12 percent of GDP with the maximum rent being 56 percent of
GDP. This suggests that natural resources contribute significantly to the GDP of
the selected countries.
The results show a positive correlation between income and CO2 emissions,
population and CO2 emissions, as well as urbanization and CO2 emissions. A nega-
tive association is observed between RE consumption, natural resources extraction
and CO2 emissions. Among the explanatory variables there is a positive association
between income and population, income and urbanization, and income and natural
resources extraction. The association between income and natural resources rent is
negative and likewise between renewable energy consumption and urbanization.
Cross-Sectional Dependence and Unit Root Test: The cross-sectional depen-
dence (CD) test results reported in table 2 indicate the null hypothesis of the
absence of cross-sectional interdependency among the variables is rejected by the
three tests, namely, Breusch-Pagan LM, Pesaran scaled LM, and Pesaran CD tests.

Table 1
DESCRIPTIVE STATISTIC AND CORRELATION MATRIXa
CO2 Y RE POP NAR UB
Statistic
Mean 66322.30 8.36E110 55.16486 36389496 12.28317 39.11194
Median 5669.182 2.65E110 55.55767 25107931 9.248667 36.44100
Maximum 503112.4 4.50E111 93.90220 1.76E108 56.61044 66.36800
Minimum 788.4050 2.07E109 7.071682 930408.0 0.000000 17.22000
Std. Dev. 124151.3 1.05E111 27.84274 35886957 11.29072 12.64071
CO2 1.000000
Correlation
Y 0.844113 1.000000
RE 20.629112 20.378342 1.000000
POP 0.234474 0.657001 0.264224 1.000000
NAR 20.173064 20.010954 0.226117 0.232667 1.000000
UB 0.515141 0.404535 20.774324 20.057425 0.147667 1.000000

a
CO2 5 carbon dioxide emissions, Y 5 income (GDP), RE 5 renewable energy consumption;
POP 5 total population; NAR 5 natural resources extraction; and UB 5 urbanization.
206 THE JOURNAL OF ENERGY AND DEVELOPMENT

Table 2
CROSS-SECTIONAL DEPENDENCE AND UNIT ROOT TEST RESULTSa
Cross-Sectional Dependence Test
Test Statistic Prob.
Breusch-Pagan LM 137.0033 0.0000
Pesaran scaled LM 16.81957 0.0000
Pesaran CD 2.155309 0.0311
CIPS Unit root test
Variable Levels, Zt-bar First difference, Zt-bar
LCO2 21.671** N/A
LY 4.160 5.6134***
LPOP 24.319*** N/A
LUB 23.819*** N/A
LNAR 22.308** N/A
LRE 21.695** N/A

a
*** 5 P , 0.001; ** 5 P , 0.05; and N/A 5 not applicable.

As a result, it becomes appropriate to use the CIPS unit root test to test for variables’
stationarity. The results of the unit root test performed with the CIPS are reported in
table 2 and it shows that all the variables are stationary at levels except income (GDP),
which is stationary at first difference. This is an indication that the variables can be
used for regression estimations without obtaining outcomes that may be spurious.
Panel Cointegration Results: The Pedroni test, as well as the Kao test, were
used to determine whether a long-run relationship exists among the variables. The
results reported in table 3 show that five of Pedroni’s multiple test statistics for
cointegration reject the null hypothesis of no cointegrating relationship. Thus, there
is a long-run relationship among the variables. Similarly, the Kao test confirms the
variables are cointegrated meaning that income, population, urbanization, natural
resources extraction, and RE consumption are the long-run causes of CO2 emis-
sions in Sub-Saharan Africa.
ARDL Regression Results: The results of the ARDL short-run and long-run
estimations are presented in table 4. Models 1 to 3 assume income has a linear rela-
tionship with CO2 emissions as captured in equations (3)–(5) while Models 4 to 6
assume income has a curve-linear relationship as captured in equations (6)–(8).
The results from Models 1 to 3 reveal similar outcomes about the effects of
income, population, urbanization, RE consumption, and natural resources extrac-
tion in the long run. A positive effect is noticed from income, population, and natu-
ral resources extraction in all models and a negative effect from RE consumption.
An income rise in the subregion is observed to be associated with an upsurge in
CO2 emissions. Should there be a 1-percent increase in the income level, CO2
emissions are likely to increase by 0.50 percent to 0.62 percent. Economic
 RESOURCES, RENEWABLE ENERGY, & CO2 EMISSIONS 207

Table 3
COINTEGRATION TEST RESULTSa
Pedroni Tests
Within-Dimension Within-Dimension (Weighted)
Statistic Prob. Prob. Statistic
Panel v-Statistic 0.845964 0.1988 1.080022 0.1401
Panel rho-Statistic 21.428586 0.0766 21.670382 0.0474
Panel PP-Statistic 24.330385 0.0000 24.600457 0.0000
Panel ADF-Statistic 0.023962 0.5096 20.396169 0.3460
Between-Dimension
Statistic Prob.
Group rho-Statistic 21.267255 0.1025
Group PP-Statistic 25.268944 0.0000
Group ADF-Statistic 20.307409 0.3793
Kao Test
Statistic Prob.
ADF 21.671093 0.0474

a
PP5 Phillips–Perron; ADF 5 Augmented Dickey-Fuller.

activities in the subregion have been increasing in recent times to the extent that it
has been considered to be the fastest-growing region.
As argued in the literature, as the income of countries increases demand for goods
and service puts much stress on the environment in order to satisfy citizens’ needs.76
The production process to meet the needs of the citizens involves more energy con-
sumption that discharges CO2 emissions. Besides the production effect, the consump-
tion of goods and services sometimes degrades the environment. This usually occurs
when increases in income make people demand more energy-consuming devices and
gadgets, which, in turn, will cause CO2 emissions to increase from their usage. The
waste management practices in the region are also not the best. Moreover, there has
been a rapidly changing lifestyle of many Africans toward the demand for canned or
rubber packaged products that are not properly disposed of after usage. The accumu-
lation of such waste degrades the environment. That notwithstanding, Models 4 to 6
reveal that income may not have such a linear relationship with CO2 emissions but
rather is non-linear because of the significant negative coefficient of the square of
income. Hence, the significant positive and negative coefficients of income and the
square of income, respectively, confirm the EKC hypothesis. This means that a rise
in income will initially increase CO2 emissions but beyond a certain threshold,
income will reduce CO2 emissions. The reason may be as a result of the technique
effect associated with higher income levels.77 The finding corroborates previous stud-
ies including S. Aboagye et al. and H. Mahmood et al.78
208 THE JOURNAL OF ENERGY AND DEVELOPMENT

Table 4
ARDL REGRESSION RESULTSa
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6
Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient
Var. (Std. Error) (Std. Error) (Std. Error) (Std. Error) (Std. Error) (Std. Error)
LONG-RUN ESTIMATES
LY
1.0262*** 0.5026*** 0.5346*** 4.5661*** 3.6403*** 4.5381***
(0.0841) (0.1316) (0.1283) (1.2586) (1.3605) (1.0417)
LY2
20.0804*** 20.0576** 20.0799***
(0.0250) (0.0273) (0.0207)
LPOP
0.5241** 0.4958* 1.5967*** 1.4479*** 0.7549** 1.3079***
(0.1751) (0.2592) (0.3167) (0.2864) (0.2398) (0.2520)
LUB
22.6622*** 21.7550*** 21.8713*** 22.1025*** 21.8416*** 21.7916***
(0.2532) (0.3363) (0.3108) (0.3782) (0.2725) (0.3009)
LNAR
0.0607*** 0.2101** 0.0391* 0.0518 0.2407*** 0.0320***
(0.0144) (0.0823) (0.0103) (0.0257) (0.0725) (0.0158)
LRE
20.5285*** 21.1207*** 20.3015*** 20.2490** 20.4354*** 20.2566***
(0.1172) (0.1390) (0.1285) (0.1646) (0.1281) (0.1312)
LRE x LNAR
20.0527 20.0539
(0.0214) (0.0192)
LNAR2
0.0103*** 0.0105***
(0.0019) (0.0022)
SHORT-RUN ESTIMATES
ECT-1
20.5595*** 20.3056** 20.4904** 20.4519*** 20.5772** 20.5278***
(0.2390) 0.142169 (0.18298) (0.1693) (0.2303) (0.1668)
D(LY)
0.5238 0.6468** 0.5228* 25.491514 232.328*** 26.5623
(0.3228) (0.3087) (0.3149) (22.5946) (9.7263) (19.6393)
D(LY2)
0.109715 0.6833*** 0.1321
(0.4687) (0.2049) (0.4070)
D(LPOP)
4.2632 1.4657 0.0106 21.5393 13.1996 20.4241
(5.2037) (5.7645) (5.8138) (3.6160) (27.9962) (4.5617)

(continued)
 RESOURCES, RENEWABLE ENERGY, & CO2 EMISSIONS 209

Table 4 (continued)
ARDL REGRESSION RESULTSa
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6
Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient
Var. (Std. Error) (Std. Error) (Std. Error) (Std. Error) (Std. Error) (Std. Error)
D(LUB)
26.2996 25.8617 26.3564 27.3235 20.4190 26.3462
(7.1110) (6.9618) (7.6721) (6.3860) (2.6291) (5.7843)
D(LNAR)
20.0025 21.7965 20.2170 20.0087 21.1325** 20.1761
(0.0243) (2.0928) (0.1320) (0.0232) (0.9263) (0.1169)
D(LRE)
20.1067 21.4328 20.1067 20.2601 21.1377 20.2388
(0.4682) (1.5262) (0.4387) (0.5296) (0.9262) (0.5240)
D(LRE x LNAR)
0.3930 0.2604**
(0.4673) (0.1305)
D(LNAR2)
0.0430 0.0314
(0.0245) (0.0233)
CONST
26.3419** 23.4317** 211.1888*** 231.1776 228.0593** 235.759***
(2.5152) (1.7263) (4.1717) (11.6143) (11.1208) (11.3641)

a
*** 5 P,0.001; **5P,0.05; Standard Error in ( ); Var. 5 Variable; Std. Error 5 Standard
Error; CONST 5 Constant term.

Population is found to significantly affect CO2 emissions as observed in

model 2. A 1-percent rise in the population is associated with about 0.49-percent
growth in CO2 emissions in the subregion. According to The Economist,79 popula-
tion growth in the SSA region stands at 2.7 percent a year; a figure higher than the
rates in South Asia (1.2 percent) and Latin America (0.9 percent). Population
growth implies more mouths to feed and more buildings for accommodation. This
requires more lands being cleared for farming and industries to increase their pro-
duction. The former process sometimes reduces forest vegetation and the latter
increases energy consumption. The combined effect will be that CO2 emissions
will increase. Therefore, it is not surprising that population growth is noted to posi-
tively affect the emissions of CO2 in the region.
On the other hand, urbanization mitigates CO2 emissions in all the models.
Based on the compact city theory, urbanization reduces energy consumption and,
hence, CO2 emissions through economies of scale. Thus, as the share of the urban
population grows, the concentration of firms and people trigger efficiency in the
usage of resources, especially energy, thereby leading to lower CO2 emissions.
210 THE JOURNAL OF ENERGY AND DEVELOPMENT

The finding suggests that the subregion may not be suffering from urbanization in
terms of CO2 emissions. This outcome is also not surprising since, in recent times,
efforts have been made to reduce traffic congestion and slum conditions in the sub-
region. Awareness creation on climate change has been intensified among urban
households, which might have made households and managers of firms to be more
environmentally conscious thereby leading to lower CO2 emissions in urban areas.
Earlier works, including H. Ali et al. and S. Adams et al., reported comparable
outcomes.80
Regarding natural resources extraction, the results from all the models point to a
positive effect. What this means is that, although in terms of revenue mobilization,
foreign exchange, and employment opportunities among other impacts, one can
argue that Sub-Saharan African countries have benefitted from the extraction of nat-
ural resources, the same cannot be said for their effect on environmental quality.
Thus, the natural resources in the subregion are being exploited at the expense of the
environment. This suggests that extraction of natural resources—whether oil, min-
eral, or forest-based—has not been done with the environment in mind or has not
been conducted in a sustainable manner. It is possible to attribute three reasons to
explain this phenomenon with the first being the energy-intensive nature of the
machines used for exploration, extraction, and the processing of natural resources.81
The second reason is the weak environmental control from state agencies. The
consequence is that firms that extract or process these natural resources go
scot-free for flouting environmental regulations. Therefore, there is, more often
than not, no urgency on the part of natural resource extractors to use
energy-efficient technologies in their operations or device ways of efficiently man-
aging their waste. This is especially so since foreign firms dominate companies
that extract major natural resources in the subregion and, if the pollution haven
effect is anything to go by, with the weak environmental regulations CO2 emis-
sions will rise. The third reason can be associated with the local citizens who
extract these natural resources, especially mineral resources, on a small scale.
While the governments in the region have permitted this to go on in order to
employ many, the weak supervision, the inadequate education, and the increasing
number of illegal extractors among other factors is a possible contributor to
increasing CO2 emissions from their operations.
To ascertain whether the relationship between natural resources and CO2 emis-
sions is monotonic or otherwise, a quadratic term of the natural resources’ extrac-
tion variable was introduced in the model. The results reported in Model 3 and
Model 6 show a direct association between CO2 emissions and extraction of natu-
ral resources. If the square term had been negative, that would have been a sigh of
relief. But the current results show that the more natural resources are extracted,
the more CO2 will be emitted, which could be due to the previously mentioned rea-
sons. A positive effect of natural resource extraction on CO2 emissions has also
been reported by L. Wang et al. and J. Hussain et al.82
 RESOURCES, RENEWABLE ENERGY, & CO2 EMISSIONS 211

The CO2 emissions effect of RE consumption in the region is reported to be

negative. Thus, increased usage of renewable energy lowers CO2 emissions.
Renewable energies are clean in terms of the rate of CO2 emissions. The implica-
tion is that more of such energy can be used for production purposes in an econ-
omy and still protect the environment. Many researchers have reported renewable
energy enhances environmental sustainability.83 To determine the role renewable
energy may play in the natural resources-carbon dioxide emissions nexus, an inter-
active term is created between renewable energy consumption and natural resour-
ces extraction. Regression results for all shows that the interaction term has a
statistically significant negative effect. The inference is that RE negatively moder-
ates the positive CO2 emissions effect of natural resources. Thus, as the natural
resources extractive firms rely on renewable energy for their extraction and proc-
essing, the lower the CO2 emissions will be. In this case, using renewable energy
reduces the level at which CO2 emissions would have been.
In the short run, the coefficient of the error correction term (ECT-1) for all mod-
els are negative and statically significant. This confirms a long-run relationship
between CO2 emissions and income, urbanization, renewable energy consumption,
and population in SSA. The EKC hypothesis was not confirmed in the short run,
suggesting that, indeed, it is a long-run occurrence. While renewable energy has a
negative coefficient it is not statistically significant in any of the short-run models.
This sends a message that resorting to renewable energy may not yield an immediate
impact on CO2 emissions. This confirms the findings of N. Ridzuan et al.,84 and a
similar argument can be given to the insignificant effect of population and urbaniza-
tion on CO2 emissions in the short run. There is a short-run negative effect of natural
resource extraction on CO2 emissions. This gives the impression that in the short
run, there appears to be efficient utilization of rents gained from natural resources to
combat pollution. The short run effect of the interactive term of renewable energy
and natural resources suggests adopting renewable energy for extracting natural
resources may not necessarily reduce its emission effect in the short run.
Quantile Regression Analysis: From table 5, the quantile regression results
show that income’s effect is significant for all the quartiles. That is, irrespective of
the level of the emissions, income has a positive effect—meaning that as income
rises so do CO2 emissions. Urbanization exerts a reducing effect on CO2 emissions
in the subregion for all quartiles. The effect of natural resource rent is also positive
throughout the quartiles although the effect appears to reduce from 0.08 at the
10-percent quartile to 0.05 at the 50-percent quartile, then it increases from there to
0.11 at the 90-percent quartile. The greater effects of natural resources at the higher
quartiles suggest that natural resource extractions have a greater effect in countries
with higher CO2 emissions. The insight from this result is that countries with lower
CO2 emissions may be those with efficient ways of extracting their natural resour-
ces or channeling some gains from natural resources to protect the environment.
On the other hand, as natural resources in higher CO2-emitting countries are
212 THE JOURNAL OF ENERGY AND DEVELOPMENT

Table 5
REGRESSION RESULTS FROM QUANTILE TECHNIQUEa
10% 25% 50% 75% 90%
Variable Coefficient Coefficient Coefficient Coefficient Coefficient
LY 0.8278*** 0.8160*** 0.9695*** 0.8262*** 0.7581
(0.0474) (0.0496) (0.0549) (0.0842) (0.5388)
LPOP 20.0098 0.0067 20.1301** 0.0360 0.1503
(0.0424) (0.0454) (0.0635) (0.0975) (0.5537)
LUB 21.4041*** 21.4182*** 21.8589*** 21.7071*** 21.8470
(0.1801) (0.1878) (0.1346) (0.1588) (1.1476)
LNAR 0.0845*** 0.0526* 0.0586*** 0.0778*** 0.1122**
(0.0300) (0.0285) (0.0167) (0.0149) (0.0472)
LRE 21.6356*** 21.5448*** 21.4154*** 21.2916*** 21.1281***
(0.0326) (0.0329) (0.0386) (0.0375) (0.0474)
Adj R2 0.67 0.67 0.72 0.76 0.75

a
*** 5 P,0.001; ** 5 P,0.05; and Standard Error in ( ).

extracted for industrial and other purposes, the processes unleash more CO2 into
the atmosphere because of weak environmental regulations or lower efficiency in
the extraction process. Moreover, this can be related to the fact that such countries
have their economies depending greatly on the natural resources available, but the
rent generated from the natural resources and growth effect of natural resources do
not translate into a quality environment in these countries.
Renewable energy consumption decreases CO2 emissions at all quantile levels.
Moreover, the emissions-reducing effect of renewable energy appears to decrease
with the quantile level. It has significant coefficients of -1.63 and -1.54 at lower quar-
tiles of 10 percent and 20 percent, respectively, and -1.29 and -1.12 at higher quartiles
of 75 percent and 90 percent, respectively. This is an indication that the tendency for
renewable energy usage to reduce CO2 emissions is stronger for countries with lower
emissions than countries with higher emissions. This could be that countries with
higher CO2 emissions may have fossil fuels as the main sources of energy that are
environmentally harmful and, as a result, it dampens the effect that renewable energy
exerts on the emissions of CO2. It also means that higher CO2-emitting countries
would need more renewable energy to have a desired quality of the environment. A
similar outcome has been reported by B. Altinoz and E. Dogan.85
Dumitrescu Hurlin Panel Causality Analysis: The Dumitrescu Hurlin (DH)
Panel Causality results (table 6) shows that there is unidirectional causality from
renewable energy to natural resources rent, from natural resources rent to CO2
emissions, and from population to CO2 emissions.
It is again noted that there is a one-way causality from urbanization to natural
resource extraction and from urbanization to income. There is also a bidirectional
 RESOURCES, RENEWABLE ENERGY, & CO2 EMISSIONS 213

Table 6
CAUSALITY RESULTSa
Null Hypothesis W-Stat. Zbar-Stat. Prob. Conclusion
LRE causes not LPOP 5.65167 4.14536 3.E-05 LRE $ LPOP
LPOP causes not LRE 6.30124 4.90798 9.E-07
LNAR causes not LPOP 6.99375 5.80261 7.E-09 LNAR $ LPOP
LPOP causes not LNAR 3.63157 1.80876 0.0705
LUB causes not LPOP 7.62582 6.62207 4.E-11 LUB$ LPOP
LPOP causes not LUB 8.29107 7.41930 1.E-13
LY causes not LPOP 11.3810 11.0713 0.0000 LY $ LPOP
LPOP causes not LY 7.95378 6.98115 3.E-12
LCO2 causes not LPOP 3.35866 1.48569 0.1374 LPOP!LCO2
LPOP causes not LCO2 5.69210 4.25883 2.E-05
LNAR causes not LRE 1.29796 20.96805 0.3330 LRE!LNAR
LRE causes not LNAR 4.45793 2.70914 0.0067
LUB causes not LRE 6.19442 4.78257 2.E-06 LUB $ LRE
LRE causes not LUB 3.62747 1.76887 0.0769
LY causes not LRE 7.59227 6.38125 2.E-10 LY!LRE
LRE causes not LY 1.58978 20.62657 0.5309
LCO2 causes not LRE 9.69971 8.89792 0.0000 LCO2$ LRE
LRE causes not LCO2 4.05903 2.27554 0.0229
LUB causes not LNAR 5.54803 4.08527 4.E-05 LUB!LNAR
LNAR causes not LUB 2.91034 0.95202 0.3411
LY causes not LNAR 5.21614 3.69102 0.0002 LY!LNAR
LNAR causes not LY 1.95454 20.18335 0.8545
LCO2 causes not LNAR 2.84153 0.85489 0.3926 LNAR!LCO2
LNAR causes not LCO2 5.38074 3.84884 0.0001
LY causes not LUB 3.29320 1.41905 0.1559 LUB ! LY
LUB causes not LY 4.66670 3.05824 0.0022
LCO2 causes not LUB 5.52600 4.06143 5.E-05 LCO2 $ LUB
LUB causes not LCO2 5.51672 4.05041 5.E-05
LCO2 causes not LY 3.73252 1.91457 0.0555 LCO2 $ LY
LY causes not LCO2 6.61788 5.32665 1.E-07

a
! and $ stand for one way/unidirectional and two way/bidirectional causality, respectively.

causality between RE and CO2 emissions as well as between income and CO2
emissions. The results from the causality test generally confirm the long-run results
reported in tables 4 and 5.
In particular, a one-way causality from natural resources and population to CO2
emissions shows an expansion in natural resource extraction and population growth
tends to increase the level of CO2 emissions in the subregion. Previous studies,
214 THE JOURNAL OF ENERGY AND DEVELOPMENT

such as those by J. Hussain et al. and D. Ibraheim et al., also found a causality
from population pressure and natural resource to CO2 emissions.86 The two-way
causality established for renewable energy consumption and CO2 emissions is an
indication that while using more renewable energy may reduce CO2 emissions in
the subregion, there is the likelihood the level of CO2 emissions would influence
the pace of renewable energy consumption. This is not surprising since currently
the concern for the rising level of CO2 emissions, which contributes to global
warming and climate change, is driving calls for investing more in renewable
energy development and consumption. The observed two-way causality between
income and CO2 emissions shows that efforts to increase income in the subregion
may affect the level of CO2 emissions, which may also affect the level of income.
Undeniably, the global discussion on rising CO2 emissions has highlighted the
threat to many developing nations that may suffer as a result of the climate change
associated with burgeoning CO2 emissions.

Conclusion and Recommendations

While natural resources may have some significant contribution to the growth
and development of developing countries, environmental concerns have been
raised in recent times by researchers. Since renewable energy is environmentally
friendly it can be argued that if the extraction of natural resources would use
greater amounts of renewable energy (RE) that may avert the potential negative
impacts associated with natural resources extraction. In this study using data for
1971-2018, three issues are analyzed within the EKC hypothesis: (a) the linear and
non-linear effects of natural resources extraction on CO2 emissions in SSA; (b) the
CO2 emissions effect of RE consumption in SSA; and (c) the interactive influence
of RE in the natural resources-CO2 emissions nexus in SSA.
Long-run analysis from the ARDL methods confirms the presence of the EKC
hypothesis for the region. Thus, an increase in economic growth and development
has the potential to improve the environmental quality in SSA through the reduc-
tion of CO2 emissions. Also, the CO2 emissions effect of natural resources extrac-
tion is found to be continuously positive. On the other hand, the direct effect
renewable energy has on CO2 emissions is negative, which means that renewable
energy reduces carbon dioxide emissions. Indirectly, renewable energy consump-
tion helps to reduce the positive effect of natural resources on CO2 emissions. The
controlled variables—population and urbanization—are found to exert positive and
negative effects, respectively. The short-run analysis failed to support the EKC
hypothesis but found natural resource extraction to reduce CO2 emissions. A quan-
tile estimation also confirmed that natural resources extraction positively affects
CO2 emissions while the CO2 emissions effect of renewable energy is negative.
From a DH causality analysis there exists one-way causation from renewable
 RESOURCES, RENEWABLE ENERGY, & CO2 EMISSIONS 215

energy consumption to natural resources extraction, from natural resources extrac-

tion to CO2 emissions, and from population to CO2 emissions. Moreover, a
two-way causality between renewable energy consumption and CO2 emissions and
between income and CO2 emissions were observed.
Theoretically, the results imply long-run economic growth can enhance the
environmental quality in SSA. Policy makers in the subregion need to aggressively
remove or minimize factors that hinder low carbon economic growth. Conservation
behavior should be promoted among the growing population in the subregion.
Since practicing a sustainable lifestyle will help reduce CO2 emissions, policy
makers need to embark upon intensive education on issues such as energy effi-
ciency, water management, food conservation, and waste management. The find-
ings confirm the argument that renewable energy is environmentally friendly. As
such, the promotion of renewable energy uptake in the subregion will go a long
way in reducing CO2 emissions. It is therefore imperative for policy makers to
incentivize investors to the renewable energy sector. In practical terms, granting of
tax rebates and provision of subsidies on renewables as incentives will increase the
production and consumption of renewable energy. Similarly, imposing higher taxes
on non-renewable energy will reduce its consumption.
The use of renewable energy by firms in the natural resources sector will
be helpful to obtain economic gains without damaging the quality of the environ-
ment. It is important therefore for policy makers to ensure that extraction of
natural resources is done in a more sustainable manner. Thus, efforts such as low-
ering of taxes on technologies that are environmentally friendly can enhance sus-
tainable extraction and processing of forest resources, mineral resources, and oil
resources.

NOTES
1
J. D. Sachs and A. M. Warner, “The Big Push, Natural Resource Booms and Growth,” Jour-
nal of Development Economics, vol. 59, no. 1 (1999), pp. 43–76, and A. Zeynalov, “The Gravity
of Institutions in a Resource-Rich Country: The Case of Azerbaijan,” International Economics
and Economic Policy, vol. 14 (2017), pp. 239–61.
2
R. A. Aljarallah and A. Angus, “Dilemma of Natural Resource Abundance: A Case Study of
Kuwait,” Sage Open, vol. 10, no. 1 (2020), pp. 1–24.
3
P. A. Kwakwa, “The Carbon Dioxide Emissions Effect of Income Growth, Electricity Con-
sumption and Electricity Power Crisis,” Management of Environmental Quality: An International
Journal, vol. 32, no. 3 (2021), pp. 470–87.
4
F. Cherubini, G. P. Peters, T. Berntsen, A. H. Strømman, and E. Hertwich, “CO2 Emissions
from Biomass Combustion for Bioenergy: Atmospheric Decay and Contribution to Global
Warming,” GDB Bioenergy, vol. 3, no. 5 (2011), pp. 413–26.
5
World Resources Forum (WRF), “Natural Resources, What Are They?” WRF website (2020),
available at https://www.wrforum.org/publications-2/publications/.
216 THE JOURNAL OF ENERGY AND DEVELOPMENT
6
C. Rhodes, “Human Consumption of Natural Resources Exceeds an Annual 100 Billion
Tonnes,” Resilience website (February 28, 2020), available at https://www.resilience.org/stories/
2020-02-28/human-consumption-of-natural-resources-exceeds-an-annual-100-billion-tonnes/.
7
P. A. Kwakwa, H. Alhassan, and G. Adu, “Effect of Natural Resources Extraction on Energy
Consumption and Carbon Dioxide Emission in Ghana,” International Journal of Energy Sector
Management, vol. 14, no. 1 (2020), pp. 20–39.
8
R. Prasad, L. Joseph, and R. C. Deo, “Modeling and Forecasting Renewable Energy Resour-
ces for Sustainable Power Generation: Basic Concepts and Predictive Model Results,” in Translat-
ing the Paris Agreement into Action in the Pacific, ed. A. Singh (Cham, Switzerland: Springer,
2020), pp. 59–79, and P. A. Kwakwa, “What Determines Renewable Energy Consumption? Star-
tling Evidence from Ghana,” International Journal of Energy Sector Management, vol. 15, no. 1
(2020), pp. 101–18.
9
P. A. Kwakwa, W. Adzawla. H. Alhassan, and A. Achaamah, “Natural Resources and Eco-
nomic Growth: Does Political Regime Matter for Tunisia?” Journal of Public Affairs, version
e2707 (2021); O. Zall
e, “Natural Resources and Economic Growth in Africa: The Role of Institu-
tional Quality and Human Capital,” Resources Policy, vol. 62 (2019), pp. 616–24; K. Janda and
G. Quarshie, “Natural Resources, Oil and Economic Growth in Sub-Saharan Africa,” MPRA
Paper no. 767481, University Library of Munich, Munich, Germany, 2017; A. Winter-Nelson,
“Natural Resources, National Income, and Economic Growth in Africa,” World Development, vol.
23, no. 9 (1995), pp.1507–519; and B. N. Epo and D. R. N. Faha, “Natural Resources, Institutional
Quality, and Economic Growth: An African Tale,” The European Journal of Development
Research, vol. 32, no. 1 (2020), pp. 99–128.
10
P. A. Kwakwa et al., “Effect of Natural Resources Extraction on Energy Consumption and
Carbon Dioxide Emission in Ghana”; S. P. Nathaniel, F. Bekun, and A. Faizulayev, “Modelling
the Impact of Energy Consumption, Natural Resources, and Urbanization on Ecological Footprint
in South Africa: Assessing the Moderating Role of Human Capital,” International Journal of
Energy Economics and Policy, vol. 11, no. 3 (2021), pp. 130–39; and I. Khan, F. Hou, H. P. Le,
and S. A. Ali, “Do Natural Resources, Urbanization, and Value-Adding Manufacturing Affect
Environmental Quality? Evidence from the Top Ten Manufacturing Countries,” Resources Policy,
vol. 72, article no. 102109 (2021).
11
B. Muhammad, M. K. Khan, M. I. Khan, and S. Khan, “Impact of Foreign Direct Investment,
Natural Resources, Renewable Energy Consumption, and Economic Growth on Environmental
Degradation: Evidence from BRICS, Developing, Developed and Global Countries,” Environmen-
tal Science and Pollution Research, vol. 28, no. 17 (2021), pp. 21789–1798; L. Wang, X. V. Vo,
M. Shahbaz, and A. Ak, “Globalization and Carbon Emissions: Is There Any Role of Agriculture
Value-Added, Financial Development, and Natural Resource Rent in the Aftermath of COP21?”
Journal of Environmental Management, vol. 268, article no. 110712 (2020); and J. Hussain, A.
Khan, and K. Zhou, “The Impact of Natural Resource Depletion on Energy Use and CO2 Emis-
sion in Belt & Road Initiative Countries: A Cross-Country Analysis,” Energy, vol. 199, issue C
(2020).
12
P. A. Kwakwa, “The Carbon Dioxide Emissions Effect of Income Growth, Electricity Con-
sumption and Electricity Power Crisis,” and S. Oluoch, P. Lal, A, Susaeta, and B. Wolde, “Public
Preferences for Renewable Energy Options: A Choice Experiment in Kenya,” Energy Economics,
vol. 98 (2021).
 RESOURCES, RENEWABLE ENERGY, & CO2 EMISSIONS 217
13
N. Bayale, E. Ali, A. F. Tchagnao, and A. Nakumuryango, “Determinants of Renewable
Energy Production in WAEMU Countries: New Empirical Insights and Policy Implications,”
International Journal of Green Energy, vol. 18, no. 2 (2021), pp. 602–14, and P. P. da Silva, P.
A. Cerqueira, and W. Ogbe, “Determinants of Renewable Energy Growth in Sub-Saharan Africa:
Evidence from Panel ARDL,” Energy, vol. 156, issue C (2018), pp. 45–54.
14
P. A. Kwakwa, F. Adusah-Poku, and K. Adjei-Mantey, “Towards the Attainment of Sustain-
able Development Goal 7: What Determines Clean Energy Accessibility in Sub-Saharan Africa?”
Green Finance, vol. 3, no. 3 (2021), pp. 268–86.
15
World Bank, World Development Indicators (Washington, D.C.: World Bank, 2021), avail-
able at https://databank.worldbank.org/source/world-development-indicators#.
16
T. Fatima, U. Shahzad, and L. Cui, “Renewable and Nonrenewable Energy Consumption,
Trade and CO2 Emissions in High Emitter Countries: Does the Income Level Matter?” Journal of
Environmental Planning and Management, vol. 64, no. 7 (2021), pp. 1227–251; S. Ali, E. Dogan,
F. Chen, and Z. Khan, “International Trade and Environmental Performance in Top Ten-Emitters
Countries: The Role of Eco-Innovation and Renewable Energy Consumption,” Sustainable Devel-
opment, vol. 29, no. 2 (2021), pp. 378–87; M. Musah, Y. Kong, I. A. Mensah, S. K. Antwi, and
M. Donkor, “The Link between Carbon Emissions, Renewable Energy Consumption, and Eco-
nomic Growth: A Heterogeneous Panel Evidence from West Africa,” Environmental Science &
Pollution Research, vol. 27, no. 23 (2020); L. Yuping, M. Ramzan, L. Xincheng, M. Murshed, A.
A. Ayobamiji, S. I. Bah, and T. S. Adebayo, “Determinants of Carbon Emissions in Argentina:
The Roles of Renewable Energy Consumption and Globalization,” Energy Reports, vol. 7 (2021),
pp. 4747–760; and B. Dogan, O. M. Driha, D. Balsalobre Lorente, and U. Shahzad, “The Mitigat-
ing Effects of Economic Complexity and Renewable Energy on Carbon Emissions in Developed
Countries,” Sustainable Development, vol. 1, no. 1 (2020), pp. 1–12.
17
P. A. Kwakwa and H. Alhassan, “The Effect of Energy and Urbanisation on Carbon Dioxide
Emissions: Evidence from Ghana,” OPEC Energy Review, vol. 24, no. 4 (2018), pp. 301–30, and
I. A. Mensah, M. Sun, C. Gao, A. Y. Omari-Sasu, D. Zhu, B. C. Ampimah, and A. Quarcoo,
“Analysis on the Nexus of Economic Growth, Fossil Fuel Energy Consumption, CO2 Emissions
and Oil Price in Africa Based on a PMG Panel ARDL Approach,” Journal of Cleaner Production,
vol. 228, no. 29 (2019), pp. 161–74.
18
World Bank, World Development Indicators (Washington, D.C.: World Bank, 2021), avail-
able at https://databank.worldbank.org/source/world-development-indicators#.
19
F. Amuakwa-Mensah and P. K. Adom, “Quality of Institution and the FEG (Forest, Energy
Intensity, and Globalization)-Environment Relationships in Sub-Saharan Africa,” Environmental
Science and Pollution Research, vol. 24 (2017), pp. 17455–7473; C. Isik, S. Ongan, D. Ozdemir,
M. Ahmad, M. Irfan, R. Alvarado, and A. Ongan, “The Increases and Decreases of the Environ-
ment Kuznets Curve (EKC) for 8 OECD Countries,” Environmental Science and Pollution
Research, vol. 28 (2021), pp. 28535–8543; M. Salahuddin, M. A. Habib, U. Al-Mulali, I. Ozturk,
M. Marshall, and M. I. Ali, “Renewable Energy and Environmental Quality: A Second-Generation
Panel Evidence from the Sub-Saharan Africa (SSA) Countries,” Environmental Research, vol. 191
(2020); and D. K. Nguyen, T. L. D. Huynh, and M. A. Nasir, “Carbon Emissions Determinants
and Forecasting: Evidence from G6 Countries,” Journal of Environmental Management, vol. 285
(2021).
20
L. Wang et al., op. cit.
218 THE JOURNAL OF ENERGY AND DEVELOPMENT
21
H. Alhassan, P. A. Kwakwa, and W. Adzawla, “Farmers Choice of Adaptation Strategies to
Climate Change and Variability in Arid Region of Ghana,” Review of Agricultural and Applied
Economics, vol. 22, no. 1 (2019), pp. 32–40.
22
H. Alhassan, “The Effect of Agricultural Total Factor Productivity on Environmental Degra-
dation in Sub-Saharan Africa,” Scientific African, vol. 12 (2021).
23
C. Isik et al., op. cit.; C. Nwani, E. L. Effiong, S. I. Okpoto, and I. K. Okere, “Breaking the
Carbon Curse: The Role of Financial Development in Facilitating Low-Carbon and Sustainable
Development in Algeria,” African Development Review, vol. 33, no. 2 (2021), pp. 300–15; and S.
Aboagye, P. Appiah-Konadu, and V. Acheampong, “Economic Expansion and Environmental
Degradation in Ghana: A Sector Decomposition Analysis,” African Journal of Economic Review,
vol. 8, no. 1 (2020), pp. 106–24.
24
P. A. Kwakwa, V. Acheampong, and S. Aboagye, “Does Agricultural Development Affect
Environmental Quality? The Case of Carbon Dioxide Emission in Ghana,” Management of Envi-
ronmental Quality, vol. 33, no. 2 (2022), pp. 527–48.
25
J. Shen, S. Wang, W. Liu, and J. Chu, “Does Migration of Pollution-Intensive Industries
Impact Environmental Efficiency? Evidence Supporting ‘Pollution Haven Hypothesis’,” Journal
of Environmental Management, vol. 242, no. 3 (2019), pp. 142–52, and G. C. Bulus and S. Koc,
“The Effects of FDI and Government Expenditures on Environmental Pollution in Korea: The Pol-
lution Haven Hypothesis Revisited,” Environmental Science and Pollution Research, vol. 28
(2021), pp. 38238–8253.
26
P. A. Kwakwa et al., “Effect of Natural Resources Extraction on Energy Consumption and
Carbon Dioxide Emission in Ghana.”

P. A. Kwakwa et al., “Effect of Natural Resources Extraction on Energy Consumption and

27

Carbon Dioxide Emission in Ghana,” and P. K. Adom, P. A. Kwakwa, and A. Amankwaa, “The
Long-Run Effects of Economic, Demographic, and Political Indices on Actual and Potential CO2
Emissions,” Journal of Environmental Management, vol. 218 (2018), pp. 516–26.

P. A. Kwakwa et al., “Effect of Natural Resources Extraction on Energy Consumption and

28

Carbon Dioxide Emission in Ghana.”

29
P. K. Adom et al., op. cit.
30
P. A. Kwakwa and G. Adu, “Effects of Income, Energy Consumption, and Trade Openness
on Carbon Emissions in Sub-Saharan Africa,” The Journal of Energy and Development, vol. 41,
nos. 1 and 2 (2016), pp. 86–117.
31
P. O. Oyetunji, O. S. Ibitoye, G. O Akinyemi, O. A, Fadele, and O. T. Oyediji, “The Effects
of Population Growth on Deforestation in Nigeria: 1991–2016,” Journal of Applied Sciences and
Environmental Management, vol. 24, no. 8 (2020), pp. 1329–334.
32
P. K. Adom et al., op. cit.
33
S. Ali et al., op. cit., and M. Musah et al., op. cit.

P. A. Kwakwa et al., “Effect of Natural Resources Extraction on Energy Consumption and

34

Carbon Dioxide Emission in Ghana.”

 RESOURCES, RENEWABLE ENERGY, & CO2 EMISSIONS 219
35
F. Amuakwa-Mensah and P. K. Adom, op. cit.; K. Li and B. Lin, “Impacts of Urbanization
and Industrialization on Energy Consumption/CO2 Emissions: Does the Level of Development
Matter?” Renewable and Sustainable Energy Reviews, vol. 52 (2015), pp. 1107–122; M. A.
Koondhar, E. N. Udemba, Y. Cheng, Z. A. Khan, M. A. Koondhar, M. Batool, and R. Kong,
“Asymmetric Causality among Carbon Emission from Agriculture, Energy Consumption, Fertil-
izer, and Cereal Food Production—A Nonlinear Analysis for Pakistan,” Sustainable Energy Tech-
nologies and Assessments, vol. 45 (2021).
36
M. Musah et al., op. cit., and P. A. Kwakwa and G. Adu, op. cit.

A. Safi, Y. Chen, S. Wahab, S. Ali, X. Yi, and M. Imran, “Financial Instability and
37

Consumption-Based Carbon Emission in E-7 Countries: The Role of Trade and Economic
Growth,” Sustainable Production and Consumption, vol. 27 (2021), pp. 383–91; R. Radmehr, S.
R. Henneberry, and S. Shayanmehr, “Renewable Energy Consumption, CO2 Emissions, and Eco-
nomic Growth Nexus: A Simultaneity Spatial Modeling Analysis of EU Countries,” Structural
Change and Economic Dynamics, vol. 57 (2021), pp. 13–27; and R. Shahnazi and Z. D. Shabani,
“The Effects of Renewable Energy, Spatial Spillover of CO2 Emissions and Economic Freedom
on CO2 Emissions in the EU,” Renewable Energy, vol. 169, issue C (2021), pp. 293–307.
38
K. Ahmed, M. Shahbaz, and P. Kyophilavong, “Revisiting the Emissions-Energy-Trade
Nexus: Evidence from the Newly Industrializing Countries,” Environmental Science and Pollution
Research, vol. 23, no. 8 (2016), pp. 7676–691; A. F. M. A. Rahman and A. K. Porna, “Growth
Environment Relationship: Evidence from Data on South Asia,” Journal of Accounting, Finance
and Economics, vol. 4, no. 1 (2014), pp. 86–96; N. Sasaki, Y. Y. Myint, I. Abe, and M. Venka-
tappa, “Predicting Carbon Emissions, Emissions Reductions, and Carbon Removal Due to Defor-
estation and Plantation Forests in Southeast Asia,” Journal of Cleaner Production (2021); and A.
Anwar, A. Sinha, A. Sharif, M. Siddique, S. Irshad, W. Anwar, and S. Malik, “The Nexus
between Urbanization, Renewable Energy Consumption, Financial Development, and CO2 Emis-
sions: Evidence from Selected Asian Countries,” Environment, Development and Sustainability
(2021), pp. 1–21.
39
R. Akhbari and M. Nejati, “The Effect of Corruption on Carbon Emissions in Developed and
Developing Countries: Empirical Investigation of a Claim,” Heliyon, vol. 5, no. 9 (2019); B.
Kretschmer, M. H€ubler, and P. Nunnenkamp, “Does Foreign Aid Reduce Energy and Carbon
Intensities of Developing Economies?” Journal of International Development, vol. 25, no. 1
(2013), pp. 67–91; and S. A. Solarin, S. P. Nathaniel, F. V. Bekun, A. M. Okunola, and A. Alhas-
san, “Towards Achieving Environmental Sustainability: Environmental Quality versus Economic
Growth in a Developing Economy on Ecological Footprint via Dynamic Simulations of ARDL,”
Environmental Science and Pollution Research, vol. 28 (2021), pp. 17942–7959.
40
B. Liddle, “Impact of Population, Age Structure, and Urbanization on Carbon Emissions/
Energy Consumption: Evidence from Macro-Level, Cross-Country Analyses,” Population and
Environment, vol. 35, no. 3 (2014), pp. 286–304.
41
S. I. Khattak and M. Ahmad, “The Cyclical Impact of Green and Sustainable Technology
Research on Carbon Dioxide Emissions in BRICS Economies,” Environmental Science and Pollu-
tion Research, vol. 29, no. 15 (2022), pp. 22687–2707, and M. K. Mahalik, H. Mallick, and H.
Padhan, “Do Educational Levels Influence the Environmental Quality? The Role of Renewable
and Non-Renewable Energy Demand in Selected BRICS Countries with a New Policy
Perspective,” Renewable Energy, vol. 164 (2021), pp. 419–32.
220 THE JOURNAL OF ENERGY AND DEVELOPMENT
42
P. K. Adom et al., op. cit.
43
C. Isik et al., op. cit.
44
K. Li and B. Lin, op. cit.

P. A. Kwakwa et al., “Effect of Natural Resources Extraction on Energy Consumption and

45

Carbon Dioxide Emission in Ghana.”

46
C. Nwani et al., op. cit.
47
S. Adams, P. K. Adom, and E. K. M. Klobodu, “Urbanization, Regime Type and Durability,
and Environmental Degradation in Ghana,” Environmental Science and Pollution Research, vol.
23 (2016), pp. 23825–3839.
48
S. Aboagye et al., op. cit.

H. Mahmood, T. T. Y. Alkhateeb, and M. Furqan, “Industrialization, Urbanization and CO2

49

Emissions in Saudi Arabia: Asymmetry Analysis,” Energy Reports, vol. 6 (2020), pp. 1553–560.
50
C. Nwani et al., op. cit.
51
C. Isik et al., op. cit.
52
K. Li and B. Lin, op. cit.
53
C. Nwani et al., op. cit.

P. Sadorsky, “The Effect of Urbanization on CO2 Emissions in Emerging Economies,”

54

Energy Economics, vol. 41, issue C (2014), pp. 147–53.

55
H. S. Ali, A. S. Abdul-Rahim, and M. B. Ribadu, “Urbanization and Carbon Dioxide Emis-
sions in Singapore: Evidence from the ARDL Approach,” Environmental Science and Pollution
Research, vol. 24, no. 2 (2017), pp. 1967–974.
56
S. Adams, E. Boateng, and A. O. Acheampong, “Transport Energy Consumption and Envi-
ronmental Quality: Does Urbanization Matter?” Science of the Total Environment, vol. 744
(2020).
57
M. A. Koondhar et al., op. cit.

P. Sheng and X. Guo, “The Long-Run and Short-Run Impacts of Urbanization on Carbon
58

Dioxide Emissions,” Economic Modelling, vol. 53 (2016), pp. 208–15.

59
P. Sadorsky, op. cit.
60
P. Sheng and X. Guo, op. cit.; P. A. Kwakwa and H. Alhassan, op. cit.; I. A. Mensah et al.,
op. cit.; S. Ali et al., op. cit.; M. Musah et al., op. cit.; and M. K. Mahalik et al., op. cit.

P. A. Kwakwa et al., “Effect of Natural Resources Extraction on Energy Consumption and

61

Carbon Dioxide Emission in Ghana”; L. Wang et al., op. cit.; and J. Hussain et al., op. cit.
62
M. S. Cato, Green Economics: An Introduction to Theory, Policy and Practice (Oxfordshire,
United Kingdom: Routledge, 2012).
 RESOURCES, RENEWABLE ENERGY, & CO2 EMISSIONS 221
63
G. M. Grossman and A. B. Krueger, “Environmental Impacts of a North American Free
Trade Agreement,” NBER Working Paper no. 3914, Cambridge, Massachusetts, National Bureau
of Economic Research (NBER), 1991.
64
T. Dietz and E. A. Rosa, “Rethinking the Environmental Impacts of Population, Affluence
and Technology,” Human Ecology Review, vol. 1 (1994), pp. 277–300.
65
Ibid.
66
World Bank, op. cit.
67
K. S. Im, M. H. Pesaran, and Y. Shin, “Testing for Unit Roots in Heterogeneous Panels,”
Journal of Econometrics, vol. 115, no. 1 (2003), pp. 53–74.
68
J. L. Wu and S. L. Chen, “Are Real Exchange Rates Stationary Based on Panel Unit-Root
Tests? Evidence from Pacific Basin Countries,” International Journal of Finance & Economics,
vol. 4, no. 3 (1999), pp. 243–52.
69
M. H. Pesaran, “A Simple Panel Unit Root Test in the Presence of Cross-Section Depend-
ence,” Journal of Applied Econometrics, vol. 22, no. 2 (2007), pp. 265–312.

P. Pedroni, “Critical Values for Cointegration Tests in Heterogeneous Panels with Multiple
70

Regressors,” Oxford Bulletin of Economics and Statistics, vol. 61, no. 0 special edition (1999), pp.
653–70.
71
G. B. Kollie, “Export-Led Growth Hypothesis in ECOWAS: A Panel Data Analysis,” African
Journal of Economic Review, vol. 8, no. 2 (2020), pp. 258–75.
72
R. Koenker and G. Bassett, Jr., “Regression Quantiles,” Econometrica, vol. 46, no. 1 (1978)
pp. 33–50.
73
Ibid.
74
E.-I. Dumitrescu and C. Hurlin, “Testing for Granger Non-Causality in Heterogeneous Pan-
els,” Economic Modelling, vol. 29, no. 4 (2012), pp. 1450–460.
75
K. H. Nguyen, and M. Kakinaka, “Renewable Energy Consumption, Carbon Emissions, and
Development Stages: Some Evidence from Panel Cointegration Analysis,” Renewable Energy,
vol. 132, issue C (2019), pp. 1049–057.

S. Aboagye, “Economic Expansion and Environmental Sustainability Nexus in Ghana,” Afri-

76

can Development Review, vol. 29, no. 2 (2017), pp. 155–68.

77
P. A. Kwakwa and G. Adu, op. cit.
78
S. Aboagye et al., op. cit., and H. Mahmood et al., op. cit.

The Economist, “The Economist Special Report on Demography,” 2020, available at https://
79

www.economist.com/special-report/2020/03/26/africas-population-will-double-by-2050.
80
H. S. Ali et al., op. cit., and S. Adams et al., op. cit.

P. A. Kwakwa et al., “Effect of Natural Resources Extraction on Energy Consumption and

81

Carbon Dioxide Emission in Ghana.”

222 THE JOURNAL OF ENERGY AND DEVELOPMENT
82
L. Wang et al., op. cit., and J. Hussain et al., op. cit.
83
S. Ali et al., op. cit.; M. Musah et al., op. cit.; and M. K. Mahalik et al., op. cit.
84
N. H. A. M. Ridzuan, N. F. Marwan, N. Khalid, M. H. Ali, and M. L. Tseng, “Effects of
Agriculture, Renewable Energy, and Economic Growth on Carbon Dioxide Emissions: Evidence
of the Environmental Kuznets Curve,” Resources, Conservation and Recycling, vol. 160 (2020).
85
B. Altinoz and E. Dogan, “How Renewable Energy Consumption and Natural Resource
Abundance Impact Environmental Degradation? New Findings and Policy Implications from
Quantile Approach,” Energy Sources, Part B: Economics, Planning, and Policy, vol. 16, no. 4
(2021), pp. 345–56.
86
J. Hussain et al., op. cit., and D. M. Ibrahiem, “Road Energy Consumption, Economic
Growth, Population and Urbanization in Egypt: Cointegration and Causality Analysis,” Environ-
ment, Development and Sustainability, vol. 20, no. 3 (2018), pp. 1053–066.