Journal of Environmental Management 340 (2023) 117911

Contents lists available at ScienceDirect

Journal of Environmental Management

journal homepage: www.elsevier.com/locate/jenvman

Research article

Realizing the 2050 Paris climate agreement in West Africa: the role of
financial inclusion and green investments
Mohammed Musah a, Bright Akwasi Gyamfi b, Paul Adjei Kwakwa c, *, Divine Q. Agozie d
a
Department of Accounting, Banking and Finance, Business School, Ghana Communication Technology University, Accra, Ghana
b
School of Management, Sir Padampat Singhania University, Bhatewar, Udaipur, Rajasthan, India
c
School of Arts and Social Sciences; University of Energy and Natural Resources, Sunyani, Ghana
d
University of Ghana Business School Department of Operations and Management Information Systems, Ghana

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

Keywords: International organizations have emphasized the importance of global economies supporting efforts to combat
Climate change climate change. The Paris Agreement or Agenda 2050 urges nations to ensure that the increase in global tem­
Environmental degradation perature is limited to 1.5 ◦ C. Studies have analyzed the factors that contribute to harmful emissions, particularly
Financial inclusion
carbon dioxide emissions, in order to limit temperature rise. However, since there are other equally harmful
Green investment
Technological innovation
pollutants, this study evaluates the impact of financial inclusion and green investment on reducing greenhouse
gas emissions. The study uses data from West Africa, where environmental pollution has significantly increased.
The study employed regression analysis while controlling for economic growth, foreign direct investment (FDI),
and energy consumption. The study’s key findings reveal that financial inclusion and green investment have a
monotonic effect on reducing greenhouse gas emissions. Additionally, the study confirms the environmental
Kuznets curve hypothesis and the pollution haven effect for the region. Technological innovation reduces
pollution, but green investment and financial inclusion reinforce this effect. Therefore, the study recommends
that governments in the sub-region commit to supporting green investment and environmentally friendly
technological innovations. It is also crucial to strictly enforce laws regulating the operations of multinational
corporations in the region.

1. Introduction fluorinated gases have significant global warming potentials due to their
radiative efficiency and longevity (EPA, 2021). Methane, for example,
Acknowledging that temperature rise is a threat to environmental has a global warming potential (GWP) of 27–30 years, while nitrous
quality (EQ), the Paris Agreement or 2050 Agenda calls for nations, both oxide has a GWP 273 times greater than CO2 over a 100-year period.
industrialized and developing, to commit to making sure that the rise in This means that methane can persist in the atmosphere for about 30
global temperature does not exceed 1.5 ◦ C. According to experts, years, while nitrous oxide can persist for 100 years or more. High-GWP
ensuring ecological quality will facilitate sustainable development gases like sulfur hexafluoride (SF6) and other carbons trap heat for
(Musah et al., 2022a, b; Li et al., 2020; Avagyan, 2021). Environmental thousands of years longer than CO2 (EPA, 2021).
quality is affected by an assortment of pollution agents [which have both The pursuit of net-zero GHG emissions has primarily been viewed
negative effects on health and the economy (Yousefi et al., 2021), among through a green development perspective, emphasizing access to clean
the most notable are greenhouse gases (GHGs) (Wang & Wang, 2022; energy sources [as stated in the Sustainable Development Goals (SDGs),
Razaq et al., 2021). especially goal 7] and investment in responsible and environmentally
Carbon dioxide (CO2) emissions is the most commonly emitted conscious production. However, GHG emissions have recently reached
greenhouse gas (GHG) and much of the literature focuses on its de­ their highest-ever annual level from energy combustion and industrial
terminants (Kwakwa and Alhassan, 2018; Wang & Wang, 2022; Razaq processes (IEA, 2021; Avagyan, 2018). The 6% rise in GHG emissions
et al., 2021). However, other GHGs like methane, nitrous oxide, and from 2020 to 2021 took the levels to 36.3 gigatons, more than reversing

* Corresponding author.
E-mail addresses: [email protected] (M. Musah), [email protected] (B.A. Gyamfi), [email protected] (P.A. Kwakwa), [email protected]
(D.Q. Agozie).

https://doi.org/10.1016/j.jenvman.2023.117911
Received 2 February 2023; Received in revised form 20 March 2023; Accepted 9 April 2023
Available online 2 May 2023
0301-4797/© 2023 Elsevier Ltd. All rights reserved.
M. Musah et al. Journal of Environmental Management 340 (2023) 117911

the pandemic-induced decline of 1.9 gigatons in 2020 (IEA, 2021). The Nigeria, Ghana, Guinea, Cote d’Ivoire, Sierra Leone, Liberia, and Togo,
increase in GHG emissions also aligns with the global economic output as shown in Fig. 1, while Senegal, Nigeria, Burkina Faso, Mali, and Niger
growth of 5.9%, making the coupling of GHG emissions with GDP have significant solar and wind energy resources (Sterl and Brecha,
expansion an ongoing threat to ecological quality (Li et al., 2022) and 2020). It is unclear how much renewable energy development could
thus the necessity to decouple GHG emissions from economic growth. reduce regional GHG emissions in the sub-region. However, the poten­
There has been a growing demand for increased investment in green tial for clean energy in the region is substantial, and the WAEP and
initiatives to achieve the goal of reducing greenhouse gas (GHG) emis­ WAPP initiatives to tap into national resources can help to recognize this
sions and promoting sustainable development (Avagyan, 2018). Green potential.
investments (GIs), such as renewable energy and energy efficiency Financial inclusion is on the rise in the region, and this can be
projects, provide a means for economic activity to take place without attributed to the liberalization of the financial sector since the late
harming the environment (Liu et al., 2022; Kwakwa, 2021; Adedoyin 1980s. Although the sector is relatively underdeveloped, its impact on
et al., 2023). This has resulted in a surge in global green investment, economic activities is felt across the region (Hammond et al., 2022).
with three times more money spent on renewable energy between 2010 However, little attention has been given to its implications for the re­
and 2019 compared to between 2001 and 2009 (UNEP, 2020). However, gion’s fight against environmental contamination, with conflicting re­
the $750 billion invested in energy efficiency and green technologies in ports (Alhassan et al., 2022). The region’s technological development is
2021 is still considered inadequate to keep temperature rise below 2 ◦ C not advanced, as in many African countries, and many farmers and
(IEA, 2021). In 2021, renewable energy investment dominated the total manufacturing firms still rely on primitive technologies (Alhassan et al.,
amount of energy investment, accounting for 70% of the $820 billion 2019). Nevertheless, efforts are underway in the sub-region to drive
invested in new power generation, primarily from developed and some technological transformation. Between 2020 and 2021, the African
developing countries (IEA, 2021). technological sector experienced a 20.6% increase in capital investment,
Apart from GIs, some argue for financial inclusion (FI) as an alter­ amounting to US $2.15 billion (UNCTAD, 2022). In West Africa, the
native to the negative link between economic growth and GHG emis­ number of tech hubs is increasing, with 9.1% annual growth in e-com­
sions (Yuan et al., 2022). According to researchers, increasing financial merce and 21.2% annual growth in digital services (OECD, 2021).
inclusivity will enhance access to financial services and business finance, Therefore, it is crucial to examine how this development can contribute
promoting sustainable economic development, renewable energy, and to addressing the GHG threat.
low-carbon activities that improve environmental quality (Ullah et al., Despite numerous studies on the determinants of greenhouse gases
2022; Ozturk and Ullah, 2022). However, others caution against the (GHGs), there are still valid reasons for further research in this area. The
potential capital crowding-out effect of financial development, which impact of financial development (FD) on the environment is mixed, and
could discourage research, limit technological innovation, and increase the impact of green investments (GI) on ecological degradation in Af­
carbon emissions (Islam et al., 2021; Yao et al., 2022). Technological rican countries is not well explored. Additionally, the moderating role of
innovation is also emphasized as a means of improving environmental financial inclusion (FI) and GI in the link between technological in­
quality by enhancing industrial process efficiency and reducing waste novations (TI) and environmental contamination has not been
and energy consumption (Obobisa et al., 2022). adequately examined. Chen et al. (2022) suggest that in developing
While it is true that countries in Sub-Saharan Africa emit fewer countries with limited technological levels, greater financial inclusion
greenhouse gases than developed nations, they are among the most can lead to increased technological innovation. Moreover, green in­
affected by their consequences. However, the rate of emissions in the vestments can influence technological innovations to become more
region, notably in the West African sub-region, such as carbon dioxide, environmentally friendly. However, there is a lack of empirical studies
nitrous oxide, and methane, is concerning (World Bank, 2022). There examining the effect of these variables on environmental pollution in
have been few optimistic reports on the region’s progress toward West Africa. Therefore, this study aims to analyze the effect of FI, GI, and
developing its renewable energy potential (Gyimah et al., 2022). The TI on environmental pollution in West Africa to provide. It seeks to
West African Energy Program (WAEP) and the West African Power Pool provide policymakers in the sub-region with empirical evidence to
(WAPP) indicate that by combining their different national resources, design or enhance appropriate frameworks to tackle the menace of
ECOWAS states can increase their clean energy supply by 60% (Sterl and GHGs. West African countries and others on the continent are the most
Brecha, 2020). Countries with sufficient hydropower resources include vulnerable to the effects of climate change. If steps are not taken to curb
the pace of GHG emissions the goals of the Paris Agreement, which they
are parties to, may not be met and member countries in the sub-region
will suffer more. The reason is that an increase in temperature can
greatly reduce agricultural yields, increase water scarcity, climate
migration and conflicts (World Economic Forum, 2021). Therefore, as
mentioned by the World Economic Forum (2021), efforts by countries in
the sub-region to reduce temperature rise is necessary.This study will
offer possible ways of achieving a low-carbon economy in the region to
attain sustainable development.
The study contributes to the existing literature in several ways. First,
it is one of the few studies that examine environmental quality using an
environmental pollution index consisting of multiple GHGs (CO2,
Methane emissions, Nitrous oxide, and PM2.5) in a single assessment,
unlike many previous studies that have focused solely on CO2 (e.g.,
Dogan et al., 2019; Agozie et al., 2022; Fareed et al., 2022; Bekun et al.,
2022). This approach captures a greater proportion of GHGs and pro­
vides fresh evidence that deviates from the trends observed in previous
literature. Moreover, while GHG emissions continue to rise, energy ef­
ficiency gains and increased clean energy utilization can help mitigate
Fig. 1. West African countries potential green energy contribution to a shared environmental challenges like climate change (UNEP, 2017). This
power pool, based on nationally available resources. highlights the potential of policies and financial inclusion strategies to
Source: Sterl and Brecha (2020). improve green investment and mitigate environmental pollution.

2
M. Musah et al. Journal of Environmental Management 340 (2023) 117911

Secondly, few studies exist on the impacts of green investment on (Gyamfi et al., 2022; Villanthenkodath and Mahalik, 2022). Similarly,
environmental pollution via GHGs, particularly from an African and barriers to TI as well as its enablers have been extensively discussed
West African perspective (Huang and Lei, 2021). This study extends the (Baloch, et al., 2021; Sinha, et al., 2020). The discussion as linked to the
literature by introducing the green investment index (GI) instead of a economic dimension of TI, proposes that strategic decisions should focus
single variable used in previous investigations (Dogan et al., 2019; on issues such as the spillover effects and information asymmetry, which
Chunyu et al., 2021). By employing a Principal Component Analysis are mainly the sources of market failure (Chen et al., 2022). Other de­
(PCA)) this study uses energy efficiency, clean energy consumption and bates related to TI issues also focus on why green TI is not spreading
research development indicators to determine the composition of green evenly across sectors (Sinha, et al., 2020). On the effect it has on
investment index. This will prove essential to unearth the effect of green pollution, authors like Ren et al. (2022) and Zhao et al. (2022) argue TI
investment on GHG emissions in West Africa. Third, this study adds to aids economic agents to tackle environmental problems. Zheng et al.
the literature on the GI, FI, and EQ nexus by trying to investigate various (2021) studied 29 Chinese provinces, and demonstrated that techno­
interactions and introducing novel FI index indicators. Fourth, many of logical progress undertaken by energy service companies reduced
these studies used econometric estimation procedures to determine the pollutant emissions. This by implication, suggests that economies
antecedents of ecological quality. The field’s ever-changing nature and experiencing TI growth can improve EQ. Shahbaz et al. (2012) affirmed
complex interrelationships between variables necessitate the use of innovation as friendly to the ecosystem of China. Similarly, Le (2020)
more modern and robust econometric modelling techniques. The studied 46 emerging countries and concluded that environmental
cross-sectionally augmented autoregressive distributed lag (CS-ARDL) technology development positively impacted sustainable economic
and cross-sectionally augmented distributed lag (CS-DL) estimation growth via lower pollution.
techniques, among others, are used in this study to examine the causal
connection between the variables. Fifth, a nonlinear relationship be­ 2.3. Financial inclusion and environmental quality
tween FI, GI, and environmental contamination is investigated.
Furthermore, the moderating effect of GI and FI in the relationship be­ Theoretically, financial inclusion (FI) is imperative for environ­
tween TI and environmental pollution is investigated. mental quality (Liu et al., 2022b; Ozturk and Ullah, 2022). Scholars
The remainder of this paper proceeds that: section 2 reviews existing perceive that financial inclusiveness increases the access and availability
works; section 3 presents an in-depth description of the method and of finances to firms and individuals, which can drive their investments in
variables; section 4 presents and discusses results and the final section efficient technologies and practices. Therefore, a more inclusive finan­
concludes with recommendations. cial system can promote superior effects on EQ (Li et al., 2020). None­
theless, greater financial inclusivity also promotes industrialization and
2. Review of related studies economic activity which can exacerbate environmental contamination
(Wang et al., 2022). Moreover, individuals will also access more credit
2.1. Green investments and environmental quality due to greater inclusiveness, thus increasing households’ purchasing
power for polluting items (Ahmad et al., 2022). Many studies on
Economic advancement practices and investments that seek to pro­ financial development suggest much consensus on its positive influence
mote resource conservation, green energy consumption and responsible on EQ (Adom et al., 2018; Kwakwa, 2021). Zaidi et al.’s (2021) explo­
development uphold the green investment (GI) assertion (Liu et al., ration of OECD nations and Renzhi and Baek’s (2020) investigation of
2022). Therefore, GI hinges on socially responsible investment activ­ 103 economies reported that FI mitigated the consequential impacts of
ities, with the goal of improving environmental gains (Yannan et al., economic progress on the ecology. However, Le et al. (2020) reported
2022). It is argued in Ahmad et al. (2022) that investments that decouple contradictory results.
unclean energy consumption and economic growth should be regarded In summary, it can be deduced from the above reviews that
as GI. Thus, investments aimed at improving environmental quality (EQ) numerous emphasize on the crucial role of GI, FI and TI on EQ. While the
constitute GI. Studies examining the EQ and GI connection abound. To findings have been mixed, little comes from African countries. Besides,
understand the drivers of financial investments for sustainable devel­ there is no empirical studies on the moderating role of FI and GI on the
opment, GI sources were identified as the main motivation for financial link between TI and ecological contamination. The above issues are
investments for responsible economic development (Dalevska et al., attended to in this study.
2019; Matuszewska-Janica et al., 2021). Aside from this Shabir et al.
(2022) studied 24 developing and developed nations from 2001 to 2019 3. Materials and methods
and reported that GI development enhanced EQ. Ren et al. (2022) also
examined 30 Chinese provinces and confirmed an inverse association 3.1. Data source and summary statistics
between GI and ecological deterioration. Also, a nonlinear association
between GI and ecological pollution reliant on quality institutions was In analysing the connection amidst the series, panel data on 12 West
observed. Evidence from Chile, Australia and Austria studied by Kahia African nations from 1990 to 2018 was used. Since the data for most of
and Ben (2021) showed that GI mitigates CO2 emission growth in in­ the variables before 1990 and after 2018 was not available, the study
dustrial sectors. Zahan and Chuanmin (2021) reported GI mitigated was confined to the period 1990 to 2018. Thus, the sampled nations and
carbon emissions in China. Also, the works of Cazcarro et al. (2022) on the study period were dependent on data availability. Data on energy
EU member states, Ibahiem and Hanafy (2021) on North Ahmad et al. utilization, GDP, and FDI were obtained from the World Bank’s database
(2022) on Japan affirmed GI as harmless to EQ. (WDI), while that on technological innovations (TI) was extracted from
OECD statistics. Following Musah (2022a, b) and Nwani et al. (2021, the
2.2. Technological innovation and the environment principal component analysis (PCA) technique was applied to construct
indexes for environmental pollution, financial inclusion and green in­
Literature generally purports that technological innovation (TI) can vestments. Many studies [Obobisa et al. (2022), Gyamfi et al. (2022) and
help reduce the concentration of pollutant gases in the atmosphere Yildrim et al. (2022] used CO2 as surrogate of ecological contamination.
where they are used in line with specified quality standards. Companies, However, since other GHGs contribute to ecological pollution, we used
industries, and even government institutions are making significant in­ PM2.5, nitrous oxide (N2O) emissions, CO2 emissions, and methane
vestments in research and development (R&D) that can promote a clean (CH4) emissions to compute an environmental pollution index (EPI) for
environment. However, prior studies on the theoretical role of TI and EQ our investigation. Since there are many indicators for financial inclusion
have largely evidenced reports from developed country contexts (Ofori-Abebrese et al., 2020; Nsiah et al., 2021), we constructed a

3
M. Musah et al. Journal of Environmental Management 340 (2023) 117911

financial Inclusion index (FII) using the indicators illustrated in Table 1. energy efficiency and clean innovations, which reduces energy utiliza­
As reported by Nsiah et al. (2021), these indicators are key measures of tion and ecological contamination (Koçak and Ulucak, 2019). Therefore,
FI as they are related to the usage and accessibility of banks’ products including it in the GII computation was deemed appropriate. The PCA
and services. The Green Investment Index (GII) was constructed using R outputs are reported in Tables 2–4. From Tables 2 and 3, the first three
& D, green energy, and energy efficiency. This approach partly follows components explained 94% and 92% of the estimated EPI and FII
the studies of Ahmad et al. (2021) and Musah et al. (2022a, b). respectively, while the first two components explained 91% of the var­
Renewable energy is one of the technically feasible and cost-effective iabilities in the estimated GII as shown in Table 4. Also, the components’
ways of mitigating emissions. It improves ecological quality via low eigenvalues in the three tables were more than 1 affirming their sig­
emissions and also promotes energy savings (Pata et al., 2022). There­ nificance. Because the other components in the tables were pretty low,
fore, promoting the utilization of green energies could improve they were excluded from the construction process. Moreover, all the
ecological quality, and help build a safe and sustainable environment on analyzed indicators had higher loadings as per the results portrayed in
a global scale (Gyamfi et al., 2022b). Hence, using it as one of the var­ the tables indicating that the variables were good surrogates for
iables in the GII computation is appropriate. Also, West African coun­ computing the various indexes. The variables used for the study were
tries have aimed at transforming into industrial economies. The rate of selected taking into consideration the SDGs of the UN. Further details on
emissions increases due to the considerable energy demand that this the series are displayed in Table 1.
change in the economic development trajectory produces in both the
industrial and residential sectors (Akram et al., 2021). Energy efficiency 3.2. Model specification and theoretical underpinning
is essential in reducing these emissions because it drives up energy
prices, depletes energy resources, and causes environmental degrada­ The theory of green economics, which centers on human activities
tion (Dogan et al., 2019). Due to the numerous benefits associated with and EQ forms the foundation of this study. It argues for the need to be
energy efficiency, nations in WA have made it a crucial part of their concerned about sustainability and that it is important to be aware of the
sustainable development policies. Energy efficiency could be viewed as potential effects that human activities may have on the environment
one of the main avenues for achieving the SDGs because it has the po­ (Cato, 2012). Since ecological contamination is an undesirable result,
tential to close the energy gap, spur rapid industrialization, and mitigate figuring out what causes or prevents it is very pertinent. Accordingly,
the externalities of climate change (Akram et al., 2021). Therefore, using based on the literature and developments/happenings in Africa, the
it as one of the variables for the GI calculation is appropriate. Besides, level of ecological deterioration in the West African region can be said to
new growth theories claim that R&D is the solution to sustainable be influenced by financial inclusion (FI), green investment (GI) and
development (Kihombo et al., 2021; Zafar et al., 2019) as it encourages technological innovation (TI). While accounting for the effect of eco­
nomic growth (GDP) (Baskurt et al., 2022), energy consumption (EC)
Table 1 (Agboola et al., 2022) and foreign direct investment (FDI) (Sattar et al.,
Data description and measurement units. 2022), the baseline model in logarithm form to obtain valid coefficient
Variable Symbol Measurement Source estimates for the connection between FI, GI, TI and EQ in West Africa
(Shahbaz et al., 2012) can be stated as below:
Technological innovations TI Total patents OECD stats
Energy consumption EC kg of oil equivalent World Bank lnEPIit = ⍺0 + β1 ln FIIit + β2 lnGIIit + β3 lnTIit + β4 lnECit + β5 lnGDPit
per capita (2022)
Gross domestic product GDP Per capita (Constant World Bank + β6 lnFDIit + μit (1)
2015 US$) (2022)
Foreign direct investment FDI Net inflows (% of World Bank Where EPI is the response variable Environmental Pollution Index rep­
GDP) (2022)
resenting environmental quality (EQ) while β1 ….,β6 are respectively the
Variables for World Bank
Environmental (2022) coefficients of explanatory variables FII, GII, TI, EC, GDP and FDI. Also,
Pollution Index (EPI) ⍺0 is the constant term while i stands for the investigated nations.
Carbon dioxide emissions CO2 Metric tons per World Bank Finally, t is the timeframe (1990–2018), while μ is the residual term.
capita (2022) Theoretically, EPI is used as a proxy of ecological pollution because
Methane emissions CH4 Kt of CO2 equivalent World Bank
(2022)
its components cover a greater portion of pollution in the environment
Nitrous oxide emissions N2O Thousand metric World Bank (Abid et al., 2022). The link between FI and EQ has been viewed as
tons of CO2 (2022) ambiguous. According to Yoshino and Morgan (2016), FI can aid
equivalent households in achieving financial stability by loosening lending con­
PM2.5 air pollution, mean PM Micrograms per World Bank
straints. Improved access to funding and the opportunity to invest in
annual exposure cubic meter (2022)
Variables for Financial World Bank more lucrative enterprises may be advantageous for industrial and
Inclusion Index (FII) (2022) manufacturing organizations as well. Making financial services
Automated teller machines ATMs Per 100,000 adults World Bank
(2022)
Table 2
Commercial bank branches CBB Per 100,000 adults World Bank
(2022) Principal components analysis on environmental pollution index (EPI).
Borrowers from BCB Per 1000 adults World Bank Component Eigenvalue Difference Proportion Cumulative
commercial banks (2022)
Depositors of commercial DCB Per 1000 adults World Bank Comp1 1.54 0.32 0.39 0.39
banks (2022) Comp2 1.22 0.21 0.31 0.70
Domestic credit to GDP DCG Percentage of GDP World Bank Comp3 1.01 0.78 0.25 0.95
ratio (2022) Comp4 0.23 – 0.05 1.00
Variables for Green World Bank Variable Comp1 Comp2 Comp3
Investments Index (GII) (2022)
Renewable energy REC Percentage of total World Bank Eigenvectors (loadings)
consumption final energy used (2022) Carbon dioxide emissions 0.68m 0.27 0.34k
Energy efficiency EFF Ratio of energy By Authors from Methane emissions − 0.43 0.58n − 0.18
consumption to GDP World Bank Nitrous oxide emissions 0.55m 0.41 0.72k
(2022) PM2.5 0.22 − 0.65n 0.58k
Research and development R&D Percentage of GDP World Bank
Note: m, n, k denote significant loadings under component 1, 2 and 3
(2022)
respectively.

4
M. Musah et al. Journal of Environmental Management 340 (2023) 117911

( )
Table 3 ∂lnEPIit
∂lnECit = β4 > 0 . Economic growth usually is associated with resource
Principal components analysis on financial inclusion index (FII).
extraction and high energy usage which can harm the environment
Component Eigenvalue Difference Proportion Cumulative
(Weili et al. (2022) as a result economic growth could be detrimental to
( )
Comp1 2.25 0.94 0.45 0.45 lnEPIit
EQ β5 = ∂∂lnGDP > 0 . Contrastingly, some economic expansion activ­
Comp2 1.31 0.29 0.26 0.71 it

Comp3 1.02 0.66 0.21 0.92 ities undertaken in nations are harmless to the ecosystem due to their
Comp4 0.37 0.32 0.07 0.99 reliance on clean energies and energy efficiency. Under the above in­
Comp5 0.05 0.01 1.00
stances, the elastic effect of economic progress on EPI could be negative
–

Variable Comp1 Comp2 Comp3

( )
∂lnEPIit
∂lnGDPit = β5 < 0 supporting Liu et al. (2022) and Elfaki et al. (2022).
Eigenvectors (loadings)
Automated teller machines per 100,000 adults − 0.42 0.55q 0.20 The nexus between FDI and ecological quality is ambiguous. Ac­
Commercial bank branches per 100,000 adults 0.63p 0.13 0.12 cording to the Pollution Haven Hypothesis, multinational establish­
Borrowers from commercial banks per 1000 adults 0.65p 0.11 0.10 ments frequently look for locations with cheap resources and labor.
Depositors of commercial banks per 1000 adults − 0.05 − 0.12 0.97r However, the relocation of the establishments to those areas comes at a
Domestic credit to GDP ratio 0.08 0.81q − 0.02
cost to the environment, due to unsavory activities there (Ashraf et al.,
Note: p, q, r denote significant loadings under components 1, 2 and 3 2022). Also, polluting multinational corporations build unclean in­
respectively. dustries in host countries by taking advantage of the absence of strict
ecological legislation. This degrades the environment (Gokmenoglu
et al., 2015). If FDI inflows support polluting activities that could
Table 4
damage the quality of the ecosystem, then, the coefficient of the variable
Principal Components analysis on green investments index (GII). ( )
lnEPIit
could be positive ∂∂lnFDI = β6 > 0 . In contrast, the halo hypothesis
Component Eigenvalue Difference Proportion Cumulative it

Comp1 1.44 0.14 0.48 0.48

postulates that some influxes of FDI support technological innovations,
Comp2 1.30 1.02 0.43 0.91 energy efficiency, human capital development, clean energy consump­
Comp3 0.27 – 0.09 1.00 tion and research and development projects that could advance
Variable Comp1 Comp2 ecological quality and as such its elasticity parameter could be negative
( )
∂lnEPIit
Eigenvectors (loadings) ∂lnFDIit = β6 < 0 .
Renewable energy consumption 0.79w − 0.05 To determine the moderating role of GI and FI in the relationship
Energy efficiency 0.48 0.67v
between TI and ecological contamination, we incorporated the inter­
Research and development − 0.38 0.74v
active terms between TI and FII (lnTI*lnFII) and between TI and GII
Note: w, v denote significant loadings under components 1 and 2 respectively. (lnTI*lnGII) into the baseline model. To avoid the problem of serious
collinearity, the interaction terms are added in separate models below:
accessible to individuals and entities may increase consumption and
ensuing ecological damage (Renzhi and Baek, 2020). Also, inclusive lnEPIit = ⍺0 + β1 ln FIIit + β2 lnGIIit + β3 lnTIit + β4 lnECit + β5 lnGDPit
financial systems promote capital creation, which stimulates economic + β6 lnFDIit +δ1 (lnTI*lnFII)it + μit
activities, and thus, calls for more energy. The excessive energy usage (2a)
may lead to increased ecological contamination (Zaidi et al., 2021). In
addition, capital markets are viewed as essential indicators of economic lnEPIit = ⍺0 + β1 ln FIIit + β2 lnGIIit + β3 lnTIit + β4 lnECit + β5 lnGDPit
development (Ahmad et al., 2022). Due to stock market’s consistent + β6 lnFDIit + δ2 (ln TI*lnGII)it + μit
success, both individual and institutional investors are drawn to it. This
(2b)
encourages consumption and production, which increases pollution, due
to the widespread utilization of dirty fuels (Raghutla and Chittedi, From 2a) and 2 b) the derivative with respect to TI gives:
2020). With reference to the above premise, the elastic effect of FII on
( ) ∂lnEPIit
EPI could be positive ∂∂lnEPI (2c)
lnFIIit = β1 > 0 , aligning the studies of Ullah
it = β3 + δ1 lnFIIit
∂lnTIit
et al. (2022) and Fareed et al. (2022). In contrast, inclusive financial
services can improve the environment by reducing emissions and other ∂lnEPIit
= β3 + δ2 lnGIIit (2d)
pollutants through the provision of funds for green technologies (Usman ∂lnTIit
et al., 2021). Based on the above premise, the coefficient of FII could be Our focus is on the sign and materiality of parameters β3 , δ1 and δ2 .
( )
negative ∂∂lnEPI
lnFIIit = β1 < 0 .
it The interpretation is that if:
β3 > 0 and δ1 (δ2 ) > 0 it means technological innovation increases
Besides, GI advances ecological quality by stimulating green projects
environmental pollution and it is reinforced by financial inclusion
and other related initiatives (Sun et al., 2021; Shen et al., 2020). We
(green investment).
therefore predict a negative affiliation between GII and ecological
( ) β3 < 0 and δ1 (δ2 ) < 0 it means technological innovation decreases
pollution ∂∂lnEPI
lnGIIit = β2 < 0 . Technological innovation improves ecolog­
it
environmental pollution and it is reinforced by financial inclusion
ical quality while enhancing performance and productivity (Guo et al., (green investment).
2022). Through improvements in production efficiency and processes, β3 > 0 and δ1 (δ2 ) < 0 it means technological innovation increases
technological innovation minimizes waste, pollution, and conserves environmental pollution but financial (green investment) reduces the
energy. Therefore, its effect on EPI is expected to be negative said impact.
( )
∂lnEPIit β3 < 0 and δ1 (δ2 ) > 0 it means technological innovation decreases
∂lnTIit = β3 < 0 . Energy use contributes to economic progress while
environmental pollution but it is positively moderated by financial in­
simultaneously contributing to environmental damage (Usman et al.,
clusion (green investment).
2020). When massive amounts of dirty energies are used to propel
To ascertain the possible nonlinear impacts of FII and GII on envi­
industrialization, production processes and other economic activities as
ronmental pollution, the square of FII (FII2) and the square of GII (GII2)
is the case of African countries, the rate of pollution in the ecology surges
were introduced into the baseline model. The augmented specification
(Ahmad et al., 2022). As such, the parameter of EC could be positive
becomes;

5
M. Musah et al. Journal of Environmental Management 340 (2023) 117911

lnEPIit = ⍺0 + β1 ln FIIit + β2 lnGIIit + β3 lnTIit + β4 lnECit + β5 lnGDPit ( )

√̅̅̅̅ − 1
+ β6 lnFDIit + φ1 lnFII 2it + φ2 ln GII 2it + μit (3) ̃ adj = N N √S̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅
Δ
− E(̃ziT )
̅ (6)
Var(̃ziT )
Where FII2 and GII2 are the squares of FII and GII correspondingly.
Where
Specifically, the relationship between FII and EPI is nonlinear if elas­
ticity coefficients β1 and φ1 have different signs and are statistically 2k(T − k − 1)
̃iT ) = k, Var(Z
E(Z ̃iT ) = (7)
significant. An inverted U-shaped relationship is discovered if β1 is T +1
( )
∂lnEPIit
greater than zero ∂lnFIIit = β1 > 0 and φ1 is less than zero This test assumes that the slope parameters are homogeneous. If this
( )
∂lnEPIit assumption is rejected, then, the coefficients are heterogeneous. To
∂lnFII2it
= φ1 < 0 . On the other hand, a U-shaped connection is estab­
( ) avoid the consequences of predicting spurious regression coefficients
lished if β1 is less than zero ∂∂lnEPI lnFIIit = β1 < 0 and φ1 is greater than zero
it
(Musah, 2022a, b; Murshed et al., 2022a; Li et al., 2022a), there is a need
(
∂lnEPIit
) to study the integration order of the series. Given that the nations are
= φ1 > 0 . Between GII and EPI, an inverted U-shaped relation­
∂lnFII2it connected via globalization and trade among others, the adoption of
( )
ship is unfolded if β2 is greater than zero ∂∂lnEPI first-generation tests for stationarity was deemed inappropriate, because
lnGIIit = β2 > 0 and φ2 is less
it

( ) they do not account for CD, heterogeneity and lack of size properties
than zero ∂∂lnEPI it
lnGII2
= φ2 < 0 . In contrast, a U-shaped association is (Habiba et al., 2022). Therefore, the CIPS and CADF tests, that account
it

discovered if β2 is less than zero and φ2 is greater than zero. To examine for CD were performed. The CADF technique is specified as;
whether the EKC hypothesis existed for West Africa or not, the square of p
∑ p
∑
GDP (GDP2) was incorporated into the baseline model resulting in the Δyit = αi + bi yi,t− 1 + ci yt− 1 + dij Δyt− j + δij Δyi,t− j + eit (8)
following specification; j=0 j=1

lnEPIit = ⍺0 + β1 ln FIIit + β2 lnGIIit + β3 lnGTIit + β4 lnECit + β5 lnGDPit In the above equation, Δyt− j and yt− j denote the cross-sectional averages.
+ β6 lnFDIit + π1 ln GDP2it + itμ Taking a simple average of Eq. (8) leads to the test statistic of the CIPS
(4) expressed as;

2 ∑
N
Where GDP is the square term of economic progress. The EKC proposes CIPS = N − 1
CADFl (9)
that economic progress initially contributes to ecological deterioration. l=1

But at a later stage, ecological pollution declines as the economy ex­ Both tests posit that the residual terms are non-stationary. Therefore,
pands. The hypothesis assumes the connection between economic failure to validate this assumption implies, the variables are stationary
progress and ecological deterioration to be inverted U-shaped. Hence, it (Murshed et al., 2022b). Since the aforestated tests do not account for
( )
lnEPIit
is validated for West Africa if β5 is greater than zero ∂∂lnGDP it
= β5 > 0 structural breaks, the Bai and Carrion-I-Silvestre (2009) test was also
(
lnEPIit
) performed. The test is expressed as;
and π1 is less than zero ∂∂lnGDP 2 = π 1 < 0 . On the other hand, a U-shaped
it
( ) ∑
li ∑
mi
lnEPIit
association is established if β5 is less than zero ∂∂lnGDP 0 and π1 (10)
′

it
= β 5 < yit = αi + Ft πi + θik DUikt + βi t + γik DTikt + εit
( ) j=1 k=1
∂lnEPIit
is greater than zero ∂lnGDP2 = π1 > 0 .
it
In Eq. (10), Ft and πi denote the common factors and factor loadings
respectively, li and mi are the structural breaks, DU and DT epitomize
3.3. Econometric methodologies dummy variables, and j and k are the break dates. Moreover, conven­
tional cointegration methods like the Kao and Chiang (2000) test fail to
The analytical process begins by checking for cross section de­ account for CD (Xue et al., 2021). Therefore, the test of Westerlund
pendencies (CD) or otherwise in the panel. Cross-sectional correlations (2007) which is vigorous to CD was performed. The test is specified as;
arise due to trade, globalization, economic integration and cross-
( ′)
∑
m ∑
m
dependence of nations among others (Habiba et al., 2022; Musah, ′
Δzit = δi di + θi zi(t− 1) + πi + θij Δzi(t− 1) + φij Δyi(1− j) + ωit (11)
2022a, b). According to Abid et al. (2022) and Koseoglu et al. (2022), j=1 j=1

undertaking panel studies without taking into consideration CD might

result in a poor selection of econometric methods, biased estimates, and In the equation above, θi is the adjustment term which measures the
inaccurate inferences. To avoid these issues, the study used the CD test of speediness to equilibrium association. This test is made up of panel and
Pesaran (2004) which is expressed as; group statistics. The specifications of the group statistics are;
√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅
( )̅
√ 1 ∑N
θi
√ 2T ∑N− 1 ∑ N
Gτ = (12a)
CDstat. = √ φ
̂ →N(0, 1) (5) N i=1 SE(̂
θi)
N(N − 1) t=1 j=i+1 ij

Where T and N denote the time and the cross-sectional dimensions 1 ∑N

Tθi
Gα = (12b)
correspondingly. The test assumes that the error terms are not correlated
′
N i=1 θi (1)
cross-sectionally. If this hypothesis is not accepted, it implies, there is CD
Here, the group average statistics are represented by Gτ and Gα , the
in the residuals. For robustness purpose, the Breusch-Pagan LM, Pesaran
error correction coefficient is denoted by θi , the semiparametric kernel
scaled LM, and the Bias-adjusted LM tests were also performed. Also,
despite the strong connections amidst member countries of WA, there estimator of θi is epitomized by ̂θ i , and the conventional standard error
are still some macroeconomic factors that might cause heterogeneity in of ̂θ i is symbolized by SE(̂
θ i ). The expressions of the panel group sta­
the slope parameters. According to Musah (2022a, b) and Murshed et al. tistics are specified as;
(2022a), ignorance of heterogeneity might lead to wrong estimates and ̂
θi
inferences. Hence, following Tackie et al. (2022), Chen et al. (2022), and Pτ = (12c)
Musah (2022a, b), the homogeneity test of Pesaran and Yamaga (2008) SE(̂
θi)
was performed. The test is specified as;

6
M. Musah et al. Journal of Environmental Management 340 (2023) 117911

variables possess a long-run affiliation, it does not guarantee that they

Pα = T ̂
θi (12d)
are causally related. Hence, we explored the causal relations amidst the
Here, the panel mean statistics are epitomized by Pτ and Pα , and the series using the Dumitrescu and Hurlin (2012) approach which accounts
common error correction estimator is denoted by ̂ θ i . The Westerlund for heterogeneity and CD. The test is expressed as;
(2007) test assumes that studied series are not flanked by a cointegration ∑
M ∑
M
affiliation. For robustness purposes, the Durbin-Hausman (DH) test that Yit = γ i + αi (m) Yit− m + δi (m) Xit− m + εit (17)
controls for CD (Pata et al., 2022) was also performed. The DH test m=1 m=1

proposed by Westerlund (2008) also consists of group and panel statis­

Here, X and Y denote the regressors and the regressand respectively,
tics expressed as;
αi (m) and δi (m) represent autoregressive and regression coefficients
∑
n ∑
T
correspondingly. Also, the constant term is symbolized by γ i , while the
DHg = ̂ φi − φi )2
S i (̃ e it−
̂ (13)
i=1 t=2
1
lags are epitomized by M. The hypothesis of the test are specified as;
H0 = βi = 0 ∀i = 1, …, N1
∑
n ∑
T
DHp = ̂ φi − φi )2
S n (̃ ̂e it− (14)
(24)
1
i=1 t=2 H1 = βi = 0 ∀i = 1, …, N 0≤ N1 /N < 1

Whilst the DHp assumes homogeneity, the DHg on the other hand, H1 = β i ∕
= 0 ∀i = 1, …, N + 1, N + 2, …, N
assumes heterogeneity and are all distributed asymptotically. Also,
when CD and heterogeneity exist in a panel, the adoption of conven­ 4. Results and discussions
tional regression techniques would yield erroneous outcomes. There­
fore, following Li et al. (2022) and Musah (2022a, b), the CS-ARDL 4.1. Descriptive statistics
approach was used to estimate the parameters of the regressors. The
CS-ARDL technique was engaged because it is efficient to handle het­ According to Table 5, there were variations in the variables of
erogeneity, endogeneity and CD (Li et al., 2022). It also accounts for concern. The average values of EPI, FII, GII, TI, EC, GDP and FDI were
omitted variable bias and non-stationarity series (Bindi, 2018). It is 16.32, 20.84, 11.65, − 1.46, 2.73, 22.72 and 0.20 respectively. Their
expressed as; corresponding standard deviations were 1.19, 6.56, 7.03, 0.80, 2.92,
yi,t = μi +
∑py
λij yi,t− j +
∑px ′
β xi,t− j +
∑p
vt + μi,j (15) 1.47 and 1.37 indicating that the data points are scattered around the
j=1 j=0 i,j j=0
averages. Also, the distributions of EPI, FII, GII, EC and GDP are posi­
tively skewed, while those of TI and FDI are negatively skewed. Besides,
Wherevt− j = (yit− j, xit− j ).For robustness purpose, the CS-DL technique
the distributions of EPI, GII, GDP and FDI are leptokurtic based on the
was also engaged. According to Chudik et al. (2017) and Herzer et al. kurtosis outcome. Thus, the variables had sharply peaked and heavy-
(2017), this approach is not bothered by lag selection sensitivity. tailed distributions. Energy use, TI and FII on the other hand, have a
Moreover, it accounts for CD, serial correlation, and heterogeneous platykurtic distribution, implying, the distribution of the series are flatly
slopes (Chudik et al., 2017). The estimator is specified as; peaked with lighter tails and more dispersed scores. Finally, FII, EC, GDP
px− 1
∑ py
∑ px
∑ and FDI have a significant and a positive association with EPI. This
(16) implies the variables tend to move in tandem while the correlation be­
′ ′
lnEPIit = wi + θi xit + δit Δxit− j + γyi,j yit− j + γ x,it xit− j + uit
j=1 j=1 j=1
tween GII, TI and EPI is negative.

Where y and x indicate the averages of the cross-sections. Besides, if

Table 5
Summary statistics and correlational analysis.
Descriptive statistics

Statistic lnEPI lnFII lnGII lnTI lnEC lnGDP lnFDI

Mean 16.32 20.84 11.65 − 1.46 2.73 22.72 0.20

Median 16.42 21.05 10.85 − 1.52 1.35 22.65 0.43
Max. 19.18 38.81 37.82 − 0.27 6.68 26.87 3.48
Min. 13.91 4.56 0.11 − 3.02 4.17 20.25 − 6.09
Std. Dev. 1.19 6.56 7.03 0.80 2.92 1.47 1.37
Skewness 0.16 0.24 0.94 − 0.02 0.16 0.70 − 1.16
Kurtosis 3.13 2.99 3.89 2.24 1.09 3.60 5.40
Observations 370 370 370 370 370 370 370
Correlational analysis

Variable lnEPI lnFII lnGII lnTI lnEC lnGDP lnFDI

lnEPI 1.00
lnFII 0.57 1.00
(0.00)***
lnGII − 0.19 − 0.01 1.00
(0.00)*** (0.87)
lnTI − 0.33 0.15 0.39 1.00
(0.00)*** (0.00)*** (0.00)***
lnEC 0.32 0.13 0.28 0.34 1.00
(0.00)*** (0.01)** (0.00)*** (0.00)***
lnGDP 0.95 0.52 0.27 0.51 0.44 1.00
(0.00)*** (0.00)*** (0.00)*** (0.00)*** (0.00)***
lnFDI 0.11 − 0.04 0.11 0.22 0.18 0.13 1.00
(0.03)** (0.36) (0.03)** (0.00)*** (0.09)* (0.01)**

Note: ***, **, * denote significance at the 1%, 5% and 10% levels correspondingly.

7
M. Musah et al. Journal of Environmental Management 340 (2023) 117911

4.2. Cross-sectional dependence and heterogeneity tests results such countries have weak laws that bind them to protect the environ­
ment from their actions (Gyamfi et al., 2022). So, many multinational
The results presented in Table 6 reveal that there is residual CD in the firms would move from countries with tougher environmental rules to
studied panel. This situation is confirmed in all the models from all five ones with laxer ecological legislations in order to pollute more as argued
tests. This implies a subsequent test for unit root or stationarity that by the pollution haven effect of FDI (Ashraf et al., 2022). On the other
ignores this situation may not yield accurate results. Moreover, the slope hand, through the pollution halo effect FDI promotes environmental
test also confirmed heterogeneity at the 1% level in the models. As a quality via the transfer of efficient technology and competition (Tanveer
result, the appropriate test should be used for the unit root tests. et al., 2022). In deed, many of the countries in the region have laws
meant to regulate the environment. Aside from the fact that these laws
4.3. Unit root test and cointegration results are comparatively weak, there is also the issue of low enforcement on
the part of authorities. Many of the laws do not commit foreign com­
The unit root tests are reported in Table 7. From the table, the hy­ panies to attach maintenance of environmental quality to their opera­
pothesis of nonstationarity is confirmed for all the variables at levels tions, a situation that may be a result of the urgent need on the part of
except TI and TI*GII, but the same hypothesis cannot be validated at first government officials to attract FDI. Consequently, their operations are
difference. Consequently, the variables are appropriate for the analysis often associated with waste generation that are sometimes not properly
in this study. The results of the cointegration tests are reported in disposed of. They also do not use energy-efficient means of production
Table 8. From the results, the hypothesis of no cointegration amongst which translate into higher environmental degradation. Some studies
the series could not be seconded. This is a confirmation that the vari­ (Kwakwa et al., 2022; Gokmenoglu et al., 2015) have found similar
ables are related in the long run. Thus, the predictors are capable of outcomes.
predicting ecological contamination in West Africa. Economic growth in model 1 has a positive effect on environmental
degradation signalling that increasing economic activities worsen the
4.4. Regression and causality test results region’s EQ. Usually, economic expansion requires more production of
goods and services and the utilization of resources to meet the needs of
The regression results of the main estimator (CS-ARDL) are reported citizens. The process leads to the generation of waste and environmental
in Table 9, while that of the robust estimator (CS-DL) are displayed in pollution. However, in model 4 income enters with a positive sign while
Table 10. The estimates from both methods as reported are similar in its square enters with a negative sign. This validates the EKC argument.
terms of significance and direction of effect. It is observed from Tables 9 This implies income initially increases ecological contamination until it
and 10 that FII, TI, GII, GDP, FDI, and EC exerts significant effects on reaches a certain threshold level. Afterward, the rise in income leads to
ecological deterioration in West Africa. A confirmation of this effect for ecological quality. In the sub-region, although economic size is lower
all the models throws some weight behind the results obtained. Specif­ than many other countries the outcome suggests that economic growth
ically, EC positively ecological pollution in West Africa. A 1% surge in has helped in promoting a quality environment. Environmental aware­
the variable is expected to affect environmental pollution by about ness campaigns by civil society organizations and pressure groups have
1.92–3.94%. Energy is a well-known necessity for the expansion and been intensified in these countries. Citizens and firms have been
development of all global economies. However, the associated emission educated on waste management practices as well which might have
makes it an issue of concern for global leaders. This stems from the fact enabled environmental pollution to be lower as economic growth in­
that fossil fuel, an unclean energy source, has dominated energy con­ creases. Previous studies including Alhassan et al. (2019) obtained
sumption over the past years. Moreover, in many developing countries similar outcomes.
including those of the West African region, there is a high number of On the main variables of interest, FI increases environmental
people that depend on fuelwood for cooking and other economic ac­ degradation as seen in models 1, 2, 3 and 4. However, its square term
tivities. The effect is that some amount of pressure is placed on forest captured in model 3 reduces environmental pollution. In other words, as
sources to meet the demand for firewood and charcoal. This then re­ FI increases it initially has some environmental degradation effect.
duces the carbon sequestration function of the forest cover. The outcome Nevertheless, beyond a certain level of FI it is then linked to lower levels
supports previous studies including Kwakwa and Alhassan (2018) and of environmental pollution. There are many possible reasons behind this
Usman et al. (2021). outcome. Financial inclusion facilitates easy access to credit by house­
Foreign direct investment also positively explains environmental holds and firms to acquire energy-intensive goods or equipment which
degradation in WA. Specifically, a 1% increase in FDI escalates the re­ will eventually lead to higher carbon dioxide emissions. Moreover, easy
gion’s ecological deterioration by about 2.96–6.01. The environmental access to credit increases the consumption and production activities of
hazards caused by FDI has been debated over the years. For some au­ households and firms thereby generating more waste. Since the 1980s
thors, FDI reduces environmental quality in developing countries since that countries in the region liberalized their financial sector following
their subscription to the IMF/World Bank structural adjustment/Eco­
Table 6 nomic Recovery Programme, there has been an expansion in the finan­
Cross-sectional dependence and heterogeneity tests results. cial space. Commercial Banks, rural and community banks, credit unions
and other financial houses have increased in these countries. This might
CD test Model 1 Model 2 Model 3 Model 4
have resulted in increased ecological deterioration as explained earlier.
Breusch-Pagan 278.30 184.11 122.73 195.87 The negative effect of the square term of FI suggests that its effect is non-
LM (0.00)*** (0.00)*** (0.00)*** (0.00)***
Pesaran scaled 32.05 47.34 58.41 60.12
linear. Thus, the initial level of FI increases environmental degradation
LM (0.00)*** (0.00)*** (0.00)*** (0.00)*** but at a later stage, it improves EQ because the financial sector has
Pesaran CD 10.80 18.86 24.22 31.03 become efficient at this stage. Efficient financial systems help entities
(0.02)** (0.00)*** (0.00)*** (0.00)*** and citizens to afford energy efficient gadgets. Higher levels of financial
Bias-adjusted 24.56 30.45 42.59 51.24
inclusion can thus be argued to help households and firms access energy-
LM (0.02)*** (0.00)*** (0.00)*** (0.00)***
Heterogeneity test efficient gadgets for production and consumption purposes. This will
Delta tilde (Δ)̃ 9.23 11.57 13.45 15.85 therefore minimize the generation of waste and ecological damage. The
(0.04)** (0.00)*** (0.00)*** (0.00)*** investigations of Zaidi et al. (2021) and Renzhi and Baek (2020) offer
Adjusted delta 14.12 17.21 20.35 25.12 support to this finding.
tilde (Δ
̃ adj ) (0.00)*** (0.06)*** (0.00)*** (0.00)***
The effects of GII and TI on environmental pollution in the sub region
Note: ***,** denote significance at the 1% and the 10% levels respectively. is negative. A 1% increment in GII and TI exerts 2.21–3.66% and

8
M. Musah et al. Journal of Environmental Management 340 (2023) 117911

Table 7
Unit root tests results.
CIPS and CADF unit root tests

CIPS CADF

Variable Levels 1st Difference Levels 1st Difference

lnEFP − 1.92 − 3.77*** − 2.44 − 3.22***

lnFII − 2.11 − 5.41*** − 1.85 − 4.68***
lnGII − 1.97 − 2.78*** − 2.57 − 3.74***
lnTI − 2.53* − 4.61*** − 2.65** − 3.41***
lnEC − 1.67 − 3.12*** − 1.48 − 4.45***
lnGDP − 2.02 − 2.84*** − 1.66 − 2.81***
lnFDI − 1.45 − 3.55*** − 2.12 − 4.72***
lnTI*lnFII − 1.86 − 2.67** − 1.32 − 2.86***
lnTI*lnGII − 2.62** − 5.72*** − 2.41** − 4.53***
lnFII2 − 2.24 − 4.44*** − 2.68 − 3.01***
lnGII2 − 1.55 − 6.35*** − 2.03 − 3.63***
lnGDP2 − 0.88 − 2.94*** − 1.57 − 2.88***

Levels 1st Difference

Variable Z statistic Pm statistic P statistic Z statistic Pm statistic P statistic

Bai and Carrion-I-Silvestre (2009) unit root test

lnEFP − 0.87 − 0.96 43.41 − 2.45*** 4.23*** 84.44***
lnFII 0.28 − 0.44 41.01 − 2.31** 3.75*** 56.64***
lnGII − 0.34 − 0.95 38.84 − 1.85** 2.44*** 48.96**
lnGTI − 0.61 − 0.56 39.22 − 1.24* 1.68** 42.87*
lnEC − 0.52 − 0.74 42.45 − 2.67*** 2.42*** 50.49**
lnGDP 0.36 − 0.47 35.61 − 2.34*** 3.68*** 38.72*
lnFDI − 0.47 − 0.65 35.34 − 1.53* 2.56*** 58.65***
lnTI*lnFII − 0.54 − 0.83 41.74 − 2.05** 1.89** 70.33***
lnTI*lnGII − 0.33 − 0.54 43.88 − 1.26* 1.57* 47.21*
2
lnFII − 0.65 − 0.71 38.37 − 2.21** 3.41*** 61.45***
lnGII2 0.55 − 0.88 43.46 − 2.42*** 2.63*** 68.46***
lnGDP2 − 0.72 − 0.94 32.12 − 1.88** 1.87** 72.12***

Notes: For Bai and Carrion-I-Silvestre (2009) test, 1, 5 and 10% critical values (CV) for Z and Pm statistics are 2.326, 1.645 and 1.282 while the critical values (CV) for P
are 56.06, 48.60 and 44.90, separately. Also, ***, **, * denote significance at the 1%, 5% and 10% levels respectively.

Table 8 Table 9
Cointegration tests results. CS-ARDL estimation results.
Westerlund (2008) test Variable Model 1 Model 2 Model 3 Model 4

Statistic Model 1 Model 2 Model 3 Model 4 lnFII 4.11 5.02 5.78 7.44 (0.00)***
(0.00)*** (0.00)*** (0.00)***
Gt − 3.45 − 5.44 − 4.12 − 2.76 (0.03)**
lnGII − 3.45 − 3.78 − 4.43 − 4.82
(0.00)*** (0.00)*** (0.00)***
(0.00)*** (0.00)*** (0.00)*** (0.00)***
Ga − 2.98 − 4.11 − 5.04 − 3.44
lnGTI − 2.91 − 2.97 − 3.45 − 5.63
(0.00)*** (0.00)*** (0.00)*** (0.00)***
(0.00)*** (0.00)*** (0.00)*** (0.00)***
Pt − 1.56 (0.02)** − 2.43 − 2.68 − 4.11
lnEC 1.92 3.41 3.17 3.94 (0.00)***
(0.00)*** (0.00)*** (0.00)***
(0.02)*** (0.00)*** (0.00)***
Pa − 4.43 − 1.35 (0.06)* − 3.45 − 3.83
lnGDP 0.98 (0.04)** 1.84 2.65 2.97 (0.00)***
(0.00)*** (0.00)*** (0.00)***
(0.00)*** (0.00)***
Durbin-Hausman test
lnFDI 2.96 3.16 6.01 5.89 (0.00)***
DHg 5.64 (0.00)*** 2.24 (0.03)** 4.11 (0.00)*** 5.11 (0.00)***
(0.00)*** (0.00)*** (0.00)***
DHp 4.35 (0.00)*** 3.67 (0.00)*** 6.35 (0.00)*** 3.33 (0.00)***
lnTI*lnFII – − 4.42 – –
Notes: ***, **, * denote significance at the 1%, 5% and 10% levels respectively. (0.00)***
lnTI*lnGII – − 2.84 – –
(0.00)***
1.97–3.88% reduction in the level of environmental pollution respec­ lnFII2 – – − 3.57 –
tively. Green investment is environmentally friendly and has received (0.00)***
2
global attention in recent years. Countries within the bloc that are lnGII – – − 2.41 –
(0.00)**
touted to be hardest hit by climate change have channelled some
lnGDP2 – – – − 2.47
amount of money into green investment activities including the gener­ (0.00)***
ation of renewable energy and energy-efficient appliances which might
F-statistic 144.21 156.06 167.88 182.45 (0.00)
have helped in reducing the emissions of greenhouse gasses. This affirms (0.00)*** (0.00)*** (0.00)*** ***
the works of Cazcarro et al. (2022) and Ahmad et al. (2021a). Similarly, R-squared 0.73 0.76 0.81 0.86
Technological innovation is associated with a high level of efficiency RMSE 0.06 0.05 0.03 0.02
which therefore reduces waste especially in the production process. It CD- − 7.23 (0.76) − 5.34 (0.55) − 4.12 (0.43) − 3.82 (0.11)
statistic
also helps in managing waste generated from production and con­
sumption. The results indicate that the region has benefitted in terms of Notes: lnEPI is the dependent variable; *** and ** denote significance at the 1%
environmental quality from technological innovation that has taken and the 5% levels respectively.
place. Previous studies by Ren et al. (2022) and Zhao et al. (2022)
affirmed TI as harmless to EQ.
The moderating role of FI and GI on the TI-environmental pollution

9
M. Musah et al. Journal of Environmental Management 340 (2023) 117911

Table 10 Table 11
CS-DL estimation results. Dumitrescu Hurlin panel causality tests results.
Variable Model 1 Model 2 Model 3 Model 4 Null Hypothesis W-stat. Zbar-Stat. Prob. Causality Flow

lnFII 3.22 4.87 3.83 5.20 (0.00)*** lnFII⇏lnEPI 6.02 4.11 0.00*** Bidirectional
(0.00)*** (0.00)**** (0.00)*** lnEPI⇏lnFII 4.74 2.45 0.00***
lnGII − 2.55 − 3.12 − 2.21 − 3.66 lnGII⇏lnEPI 3.44 1.72 0.02** Unidirectional
(0.00)*** (0.00)*** (0.00)*** (0.00)*** lnEPI⇏lnGII 1.65 − 0.12 0.54
lnTI − 1.97 − 2.13 − 3.75 − 3.88 lnTI⇏lnEPI 5.81 3.67 0.00*** Unidirectional
(0.00)*** (0.00)*** (0.00)*** (0.00)*** lnEPI⇏lnTI 1.46 − 0.52 0.88
lnEC 0.98 (0.02)** 1.67 (0.00)*** 2.17 2.53 (0.00)*** lnEC⇏lnEPI 7.31 5.54 0.00*** Bidirectional
(0.00)*** lnEPI⇏lnEC 3.12 1.15 0.03**
lnGDP 0.45 (0.00)** 0.72 (0.04)** 1.52 1.81 (0.00)*** lnGDP⇏lnEPI 6.26 4.48 0.00*** Bidirectional
(0.00)*** lnEPI⇏lnGDP 3.27 1.18 0.02**
lnFDI 1.55 2.89 (0.00)*** 3.61 4.51 (0.00)*** lnFDI⇏lnEPI 8.23 6.41 0.00*** Unidirectional
(0.00)*** (0.00)*** lnEPI⇏lnFDI 1.54 − 0.48 0.83
lnTI*lnFII – − 2.24 – – lnTI*lnFII⇏lnEPI 4.89 2.47 0.00*** Unidirectional
(0.00)*** lnEPI⇏lnTI*lnFII 1.65 − 0.38 0.67
lnTI*lnGII – − 1.15 – – ln TI*lnGII⇏lnEPI 7.35 5.61 0.00*** Unidirectional
(0.00)*** lnEPI⇏lnTI*lnGII 1.17 − 0.82 0.98
2
lnFII – – − 1.43 – lnFII2⇏lnEPI 3.04 1.01 0.03** Bidirectional
(0.00)*** lnEPI⇏lnFII2 4.21 2.38 0.00***
lnGII2 – – − 0.21 – lnGII2⇏lnEPI 5.03 3.52 0.00*** Unidirectional
(0.04)** lnEPI⇏ lnGII2 1.57 − 0.42 0.78
lnGDP2 – – – − 1.88 lnGDP2⇏lnEPI 3.38 1.68 0.02** Unidirectional
(0.00)*** lnEPI⇏lnGDP2 1.28 0.61 0.94

F-statistic 103.44 116.12 128.24 132.07 (0.00) Notes: lnEPI is the dependent variable; ***, ** denote significance at the 1% and
(0.02)** (0.00)*** (0.00)*** *** the 5% levels respectively.
R-squared 0.65 0.68 0.72 0.77
RMSE 0.05 0.04 0.02 0.01
CD- − 8.81 (0.82) − 6.02 (0.71) − 5.46 (0.67) − 3.71 (0.48) remaining series had one-way causality with the dependent variable.
statistic Various recommendations stem from the results above. Theoreti­
Notes: lnEPI is the dependent variable; ***, ** denote significance at the 1% and cally, the study suggests that WA countries can rely on FI and GI in
the 5% levels respectively. achieving lower environmental pollution. Practically, it is suggested that
entities in the financial sector should shift their support away from the
nexus as reported in model 2 is found to be negative. The results from development of carbon-intensive products and toward eco-friendly al­
model 2 show that FI enhances the ecological deterioration mitigation ternatives. At the moment the energy-intensive businesses are the ones
effects of TI. Also, GI promotes the ecological contamination abatement that easily get financial support because it is more profitable. Authorities
effects of TI. In other words, FI helps TI to further reduce environmental in the sector should be convinced to look at the future benefit of sup­
problems. This could be as a result of the fact that innovation requires porting eco-friendly activities than the current profit from supporting
huge financial commitment which becomes easier to access when there energy-intensive ventures. The interest rate at which financial in­
is FI. As such, more funds will be available to support activities that will stitutions in the region give credits to firms is comparatively higher than
lead to innovative ways of production that are environmentally friendly. their Asian counterparts. The need to have a vibrant financial sector that
It even becomes more prominent when actors in the financial sector are lowers interest rate is needed critically. Also, climate finance is at early
particularly concerned with environmental safety and would therefore stage in the region and attention should be given to that.
be much more ready to push funds into TI that will safeguard the quality Environmental sustainability is noted undermined by FDI in West
of the environment. Also, GI may trigger TI to further reduce environ­ Africa. Authorities in the region need to enforce their environmental
mental pollution because the former can ensure that the outcome of laws and regulations to avoid the situation where the region becomes a
innovations has some green energy to fuel it for operations. carbon sink for foreign companies. The findings of this investigation
The DH causality outcomes are exhibited in Table 11. From the table, imply that it would be beneficial to support firms to engage in green
a bidirectional causality amidst FII, EC, GDP, FII2 and ecological investment activities in the region. Governments in the region should
contamination was observed. Also, one-way causality from GII, TI, FDI, increase financial assistance to firms in this regard. Specifically,
TI*FII, and TI*GII to environmental degradation was documented. reducing import duties on items that will promote green investment
These are in line with the regression results. should be targeted. Provision of tax incentives to firms that go into green
investment will also be helpful. For example, the government may
5. Conclusion and policy recommendations decide to offer tax rebates to businesses that invest in renewable energy
in order to reduce their dependence on dirty fuels. This will equally
Towards the realization of the 2050 agenda of the Paris Accord, this require a commitment on the part of governments in the region to
study examined the role of financial inclusion (FI) and green investment devote financial resources to improve the renewable energy
(GI) in mitigating ecological pollution in 12 West African nations over infrastructure.
the period 1990 to 2018. It focused on assessing their direct effect on The lack of available data was a problem that plagued this investi­
pollution and their moderating effect on technological innovation. Other gation. The data for some of the factors were missing during some of the
variables namely energy consumption (EC), economic growth/income time periods. All gaps in the dataset were thus filled using a combination
(GDP) and FDI were controlled for. From the regression results, the EKC of interpolation and extrapolation techniques. Therefore, comparable
hypothesis was validated for the sub-region. Also, FI, EC and FDI in­ analyses outside region and in the region could be undertaken in the
crease pollution while GI and TI mitigate pollution in WA. The squares of future when such data become available or accessible to confirm the
FI and GI enhance environmental quality (EQ) by reducing pollution. validity of the study’s findings.
Also, FI and GI were found to enhance the pollution mitigation influence
of TI. The results also show there is among others a two-way causation
amidst FI and pollution, and between income and pollution, while the

10
M. Musah et al. Journal of Environmental Management 340 (2023) 117911

Credit author statement Cato, M.S., 2012. Green Economics: an Introduction to Theory, Policy and Practice.
Routledge.
Cazcarro, I., García-Gusano, D., Iribarren, D., Linares, P., Romero, J.C., Arocena, P.,
Mohammed Musah: Conceptualization, Methodology, Data curation, et al., 2022. Energy-socio-economic-environmental modelling for the EU energy and
writing of draft, Bright Akwasi Gyamfi: Conceptualization; Formal post-COVID-19 transitions. Sci. Total Environ. 805, 150329.
analysis, Supervision and writing of draft, Paul Adjei Kwakwa: Chen, H., Tackie, E.A., Ahakwa, I., Musah, M., et al., 2022. Does energy consumption,
economic growth, urbanization, and population growth influence carbon emissions
Conceptualization; Methodology, writing, Supervision and correspond­ in the BRICS? Evidence from panel models robust to cross-sectional dependence and
ing, Divine Q. Agozie: Literature review, Conceptualization, Visualiza­ slope heterogeneity. Environ. Sci. Pollut. Res. https://doi.org/10.1007/s11356-021-
tion, and writing of draft 17671-4.
Chudik, A., Mohaddes, K., Pesaran, M.H., Raissi, M., 2017. Is there a debt-threshold
effect on output growth? Rev. Econ. Stat. 99 (1), 135–150.
Dalevska, N., Khobta, V., Kwilinski, A., Kravchenko, S., 2019. A model for estimating
Declaration of competing interest social and economic indicators of sustainable development. Entrepren. Sustain.
Issues 6 (4), 1839.
Dogan, E., Taspinar, N., Gokmenoglu, K.K., 2019. Determinants of ecological footprint in
The authors declare that they have no known competing financial MINT countries. Energy Environ. 30 (6), 1065–1086.
interests or personal relationships that could have appeared to influence Dumitrescu, E.I., Hurlin, C., 2012. Testing for Granger non-causality in heterogeneous
the work reported in this paper. panels. Econ. Modell. 29 (4), 1450–1460. https://doi.org/10.1016/j.
econmod.2012.02.014.
Elfaki, K.E., Khan, Z., Kirikkaleli, D., et al., 2022. On the nexus between industrialization
Data availability and carbon emissions: evidence from ASEAN + 3 economies. Environ. Sci. Pollut.
Res. 29, 31476–31485. https://doi.org/10.1007/s11356-022-18560-0.
Data will be made available on request. EPA, 2021. National Overview: Facts and Figures on Materials, Wastes and Recycling,
2021 April 15 [cited 2023 March 29].
Fareed, Z., Rehman, M.A., Adebayo, T.S., et al., 2022. Financial inclusion and the
References environmental deterioration in Eurozone: the moderating role of innovation activity.
Technol. Soc. https://doi.org/10.1016/j.techsoc.2022.101961.
Gokmenoglu, K., Azin, V., Taspinar, N., 2015. The relationship between industrial
Abid, N., Ceci, F., Ahmad, F., Aftab, J., 2022. Financial development and green
production, GDP, inflation and oil price: the case of Turkey. Procedia Econ. Finance
innovation, the ultimate solutions to an environmentally sustainable society:
25, 497–503. https://doi.org/10.1016/S2212-5671(15)00762-5.
evidence from leading economies. J. Clean. Prod. 369, 133223 https://doi.org/
Guo, L., Ou, Z., Liu, Y., et al., 2022. Technological innovations on direct carbon
10.1016/j.jclepro.2022.133223.
mitigation by ordered energy conversion and full resource utilization. Carb
Adedoyin, F.F., Bekun, F.V., Hossain, M.E., Gyamfi, B.A., Haseki, M.I., 2023. Glasgow
Neutrality 1, 4. https://doi.org/10.1007/s43979-022-00009-5.
Climate Change Conference (COP26) and its Implications in Sub-sahara Africa
Gyamfi, B.A., Agozie, D.Q., Bekun, F.V., 2022. Can technological innovation, foreign
Economies. Renewable Energy.
direct investment and natural resources ease some burden for the BRICS economies
Adom, P.K., Kwakwa, P.A., Amankwaa, A., 2018. The long-run effects of economic,
within current industrial era? Technol. Soc. 70, 102037 https://doi.org/10.1016/j.
demographic, and political indices on actual and potential CO2 emissions.
techsoc.2022.102037.
J. Environ. Manag. 218, 516–526.
Gyamfi, B.A., Kwakwa, P.A., Adebayo, T.S., 2022b. Energy Intensity Among European
Agboola, P.O., Hossain, M.E., Gyamfi, B.A., et al., 2022. Environmental consequences of
Union Countries: the Role of Renewable Energy, Income and Trade. International
foreign direct investment influx and conventional energy consumption: evidence
Journal of Energy Sector Management (ahead-of-print).
from dynamic ARDL simulation for Turkey. Environ. Sci. Pollut. Res. https://doi.
Gyimah, J., Yao, X., Tachega, M.A., Hayford, I.S., Opoku-Mensah, E., 2022. Renewable
org/10.1007/s11356-022-19656-3.
energy consumption and economic growth: new evidence from Ghana. Energy 248,
Ahmad, M., Khan, Z., Rahman, Z.U., Khattak, S.I., Khan, Z.U., 2021. Can innovation
123559.
shocks determine CO2 emissions (CO2e) in the OECD economies? A new perspective.
Habiba, U., Xinbang, C., Anwar, A., 2022. Do green technology innovations, financial
Econ. Innovat. N. Technol. 30 (1), 89–109. https://doi.org/10.1080/
development, and renewable energy use help to curb carbon emissions? Renew.
10438599.2019.1684643.
Energy 193, 1082–1093. https://doi.org/10.1016/j.renene.2022.05.084.
Agozie, D.Q., Gyamfi, B.A., Bekun, F.V., Ozturk, I., Taha, A., 2022. Environmental
Hammond, P., Opoku, M.O., Kwakwa, P.A., 2022. Relationship among corporate
Kuznets Curve hypothesis from lens of economic complexity index for BRICS:
reporting, corporate governance, going concern and investor confidence: evidence
evidence from second generation panel analysis. Sustain. Energy Technol.
from listed banks in sub Saharan Africa. Cogent Business & Management 9 (1),
Assessments 53, 102597.
2152157.
Ahmad, S., Khan, D., Magda, R., 2022. Assessing the influence of financial inclusion on
Herzer, D., Nagel, K., Dewenter, R., 2017. The effects of adult and non-adult mortality on
environmental degradation in the ASEAN region through the panel PMG-ARDL
long-run economic development: Evidence from a heterogeneous dynamic and cross-
approach. Sustainability 14, 7058. https://doi.org/10.3390/su14127058.
sectionally dependent panel of countries between 1800 and 2010, Diskussionspapier,
Akram, R., Chen, F., Khalid, F., Huang, G., Irfan, M., 2021. Heterogeneous effects of
No. 177, Helmut-Schmidt-Universität-Universität der Bundeswehr Hamburg.
energy efficiency and renewable energy on economic growth of BRICS countries: a
Fächergruppe Volkswirtschaftslehre, Hamburg.
fixed effect panel quantile regression analysis. Energy 215, 119019.
Huang, L., Lei, Z., 2021. How environmental regulation affect corporate green
Alhassan, H., Kwakwa, P.A., Adzawla, W., 2019. Farmers choice of adaptation strategies
investment: evidence from China. J. Clean. Prod. 279, 123560.
to climate change and variability in arid region of Ghana. Review of Agricultural and
IEA (International Energy Agency, 2021. Net Zero by 2050. https://www.iea.org/reports
Applied Economics (RAAE) 22 (1340–2019-779), 32–40.
/net-zero-by-2050 [Google Scholar].
Alhassan, H., Kwakwa, P.A., Donkoh, S.A., 2022. The interrelationships among financial
Kao, C., Chiang, M.H., 2000. On the estimation and inference of a cointegrated regression
development, economic growth and environmental sustainability: evidence from
in panel data. In: Nonstationary Panels, Panel Cointegration, and Dynamic Panels.
Ghana. Environ. Sci. Pollut. Control Ser. 29 (24), 37057–37070.
Advances in Econometrics, vol. 15. JAI Press, pp. 179–222. https://doi.
Ashraf, S., Jithin, P., Umar, Z., 2022. The asymmetric relationship between foreign direct
org/10.1016/S0731-9053(00)15007-8.
investment, oil prices and carbon emissions: evidence from Gulf Cooperative Council
Kihombo, S., Saud, S., Ahmed, Z., Chen, S., 2021. The effects of research and
economies. Cogent Economics & Finance 10, 1. https://doi.org/10.1080/
development and financial development on CO2 emissions: evidence from selected
23322039.2022.2080316, 2080316.
WAME economies. Environ. Sci. Pollut. Control Ser. 28 (37), 51149–51159. https://
Avagyan, A.B., 2018. Algae to Energy and Sustainable Development. Technologies,
doi.org/10.1007/s11356-021-14288-5.
Resources, Economics and System Analyses. New Design of Global Environmental
Koçak, E., Ulucak, Z.Ş., 2019. The effect of energy R&D expenditures on CO2 emission
Policy and Live Conserve Industry. Amazon. ISBN-10: 17187225, -13: 978-
reduction: estimation of the STIRPAT model for OECD countries. Environ. Sci. Pollut.
1718722552.
Res. 26, 14328–14338. https://doi.org/10.1007/s11356-019-04712-2.
Avagyan, A.B., 2021. Theory of bioenergy accumulation and transformation: application
Koseoglu, A., Yucel, A.G., Ulucak, R., 2022. Green innovation and ecological footprint
to evolution, energy, sustainable development, climate change, manufacturing,
relationship for a sustainable development: evidence from top 20 green innovator
agriculture, military activity and pandemic challenges. Athens J Sci 8, 57–80.
countries. Sustain. Dev. https://doi.org/10.1002/sd.2294.
Bai, J., Carrion-I-Silvestre, J.L., 2009. Structural changes, common stochastic trends, and
Kwakwa, P.A., 2021. What determines renewable energy consumption? Startling
unit roots in panel data. Rev. Econ. Stud. 76 (2), 471–501.
evidence from Ghana. Int. J. Energy Sect. Manag. 15 (1), 101–118.
Baskurt, B.B., Celik, S., Aktan, B., 2022. Do foreign direct investments influence
Kwakwa, P.A., Alhassan, H., 2018. The effect of energy and urbanisation on carbon
environmental degradation? Evidence from a panel autoregressive distributed lag
dioxide emissions: evidence from Ghana. OPEC Energy Review 42 (4), 301–330.
model approach to low-, lower-middle-, upper-middle-, and high-income countries.
Kwakwa, P.A., Adzawla, W., Alhassan, H., Oteng-Abayie, E.F., 2022. The effects of
Environ. Sci. Pollut. Res. 29, 31311–31329. https://doi.org/10.1007/s11356-021-
urbanization, ICT, fertilizer usage, and foreign direct investment on carbon dioxide
17822-7.
emissions in Ghana. Environ. Sci. Pollut. Control Ser. 1–15.
Bekun, F.V., Alola, A.A., Gyamfi, B.A., Kwakwa, P.A., Uzuner, G., 2022. Econometrics
Le, T.H., Le, H.C., Taghizadeh-Hesary, F., 2020. Does financial inclusion impact CO2
analysis on cement production and environmental quality in European Union
emissions? Evidence from Asia. Finance Research Letters 34, 101451.
countries. Int. J. Environ. Sci. Technol. 1–16.
Li, Z.-Z., Li, R.Y.M., Malik, M.Y., Murshed, M., Khan, Z., Umar, M., 2020. Determinants of
Bindi, G., 2018. The Resource Curse Hypothesis: an Empirical Investigation. Department
carbon emission in China: how good is green investment? Sustainable Production
of Economics, Lund University. http://lup.lub.lu.se/student-papers/record/
and Consumption. https://doi.org/10.1016/j.spc.2020.11.008.
8958321.

11
M. Musah et al. Journal of Environmental Management 340 (2023) 117911

Li, S., Yu, Y., Jahanger, A., Usman, M., Ning, Y., 2022. The impact of green investment, Shen, Y., Su, Z.-W., Yousaf, M., et al., 2020. Does green investment, financial
technological innovation, and globalization on CO2 emissions: evidence from MINT development and natural resources rent limit carbon emissions? A provincial panel
countries. Front. Environ. Sci. 10, 868704 https://doi.org/10.3389/ analysis of China. Sci. Total Environ. https://doi.org/10.1016/j.
fenvs.2022.868704. scitotenv.2020.142538.
Li, K., Ying, H., Ning, Y., et al., 2022a. China’s 2060 carbon-neutrality agenda: the nexus Sterl, S., Brecha, R., 2020. How West Africa can expand power supply and meet
between energy consumption and environmental quality. Environ. Sci. Pollut. renewable climate goals. Quartz Publication online. https://qz.com/africa/186574
Control Ser. https://doi.org/10.1007/s11356-022-19456-9. 3/west-africa-can-expand-renewable-power-meet-climate-goals.
Liu, H., Wang, K.-T., Khudoykulov, K., Tai, T.D., Ngo, T.Q., Phan, T.T.H., 2022. Does Sun, Y., Li, H., Zhang, K., Kamran, H.W., 2021. Dynamic and Casual Association between
economic development impact CO2 emissions and energy efficiency performance? Green Investment, Clean Energy and Environmental Sustainability Using Advance
Fresh evidences from Europe. Front. Energy Res. 10, 860427 https://doi.org/ Quantile A.R.D.L. Framework. Economic Research-Ekonomska Istraživanja. https://
10.3389/fenrg.2022.860427. doi.org/10.1080/1331677X.2021.1997627.
Murshed, M., Samiha, K., Rahman, A.K.M.A., 2022a. Roadmap for achieving energy Tackie, E.A., Chen, H., Ahakwa, I., Atingabili, S., 2022. Exploring the dynamic nexus
sustainability in Sub-Saharan Africa: the mediating role of energy use efficiency. among economic growth, industrialization, medical technology, and healthcare
Energy Rep. 8, 4535–4552. https://doi.org/10.1016/j.egyr.2022.03.138. expenditure: a PMG-ARDL panel data analysis on income-level classification along
Matuszewska-Janica, A., Żebrowska-Suchodolska, D., Ala-Karvia, U., Hozer-Koćmiel, M., West African economies. Front. Public Health 10, 903399. https://doi.org/10.3389/
2021. Changes in Electricity Production from Renewable Energy Sources in the fpubh.2022.903399.
European Union Countries in 2005–2019. Energies 14 (19), 6276. Ullah, S., Ali, K., Shah, S.A., et al., 2022. Environmental concerns of financial inclusion
Murshed, M., Nurmakhanova, M., Al-Tal, R., Mahmood, H., Elheddad, M., Ahmed, R., and economic policy uncertainty in the era of globalization: evidence from low &
2022b. Can intra-regional trade, renewable energy use, foreign direct investments, high globalized OECD economies. Environ. Sci. Pollut. Res. https://doi.org/
and economic growth mitigate ecological footprints in South Asia? Energy Sources B 10.1007/s11356-022-18758-2.
Energy Econ. Plann. https://doi.org/10.1080/15567249.2022.2038730. UNCTAD, 2022. Digitalization as a Driver of Structural Transformation in African LDCs.
Musah, M., 2022a. Financial inclusion and environmental sustainability in Ghana: https://unctad.org/news/blog-digitalization-driver-structural-transformation-afr
application of the dynamic ARDL estimator. Environ. Sci. Pollut. Control Ser. ican-ldcs.
https://doi.org/10.1007/s11356-022-19994-2. Usman, M., Kousar, R., Yaseen, M.R., Makhdum, M.S.A., 2020. An empirical nexus
Musah, M., 2022b. Stock Market Development and Environmental Quality in EU Member between economic growth, energy utilization, trade policy, and ecological footprint:
Countries: a Dynamic Heterogeneous Approach. Environment, Development and a continent-wise comparison in UpperMiddle-Income countries. Environ. Sci. Pollut.
Sustainability. https://doi.org/10.1007/s10668-022-02521-1. Res. 27 (31), 38995–39018. https://doi.org/10.1007/s11356-020-09772-3.
Musah, M., Owusu-Akomeah, M., Kumah, E.A., et al., 2022a. Green investments, Usman, M., Makhdum, M.S.A., Kousar, R., 2021. Does financial inclusion, renewable and
financial development, and environmental quality in Ghana: evidence from the non-renewable energy utilization accelerate ecological footprints and economic
novel dynamic ARDL simulations approach. Environ. Sci. Pollut. Control Ser. growth? Fresh evidence from 15 highest emitting countries. Sustain. Cities Soc. 65,
https://doi.org/10.1007/s11356-021-17685-y. 102590 https://doi.org/10.1016/j.scs.2020.102590.
Musah, M., Boateng, F., Kumah, E.A., 2022b. Financial Flows and Environmental Quality Weili, L., Khan, H., khan, I., et al., 2022. The impact of information and communication
in ECOWAS Member States: Accounting for Residual Cross-Sectional Dependence technology, financial development, and energy consumption on carbon dioxide
and Slope Heterogeneity. Environment, Development and Sustainability. https:// emission: evidence from the Belt and Road countries. Environ. Sci. Pollut. Res. 29,
doi.org/10.1007/s10668-022-02755-z. 27703–27718. https://doi.org/10.1007/s11356-021-18448-5.
Nsiah, A.Y., Yusif, H., Tweneboah, G., Agyei, K., Baidoo, S.T., 2021. The effect of Westerlund, J., 2007. Testing for error correction in panel data. Oxf. Bull. Econ. Stat. 69
financial inclusion on poverty reduction in SubSahara Africa: does threshold matter? (6), 709–748. https://doi.org/10.1111/j.1468-0084.2007.00477.x.
Cogent Social Sciences 7 (1), 1903138. https://doi.org/10.1080/ Westerlund, J., 2008. Panel cointegration tests of the Fisher effect. J. Appl. Econom. 23
23311886.2021.1903138. (2), 193–233. https://doi.org/10.1002/jae.967.
Nwani, C., Effiong, E.L., Okpoto, S.I., Okere, I.K., 2021. Breaking the carbon curse: the World Bank, 2022. World Development Indicators. https://databank.worldbank.
role of financial development in facilitating lowcarbon and sustainable development org/source/world-development-indicators#.
in Algeria. Afr. Dev. Rev. 33, 300–315. https://doi.org/10.1111/1467-8268.12576. World Economic Forum. For millions of West Africans, climate change is already here.
Obobisa, E.S., Chen, H., Mensah, I.A., 2022. The impact of green technological https://www.weforum.org/agenda/2021/10/west-and-central-africa-climate-
innovation and institutional quality on CO2 emissions in African countries. Technol. migrants/.
Forecast. Soc. Change 180, 121670. https://doi.org/10.1016/j. Xue, L., Haseeb, M., Mahmood, H., Alkhateeb, T.T.Y., Murshed, M., 2021. Renewable
techfore.2022.121670. energy use and ecological footprints mitigation: evidence from selected South Asian
OECD, 2021. Digital Transformation for Youth Employment and Agenda 2063 in West economies. Sustainability 13, 1613. https://doi.org/10.3390/su13041613.
Africa. https://www.oecd-ilibrary.org/sites/0a5c9314-en/1/3/8/index.html?ite Yannan, D., Ahmed, A.A.A., Kuo, T.H., Malik, H.A., Nassani, A.A., Haffar, M., Iramofu, D.
mId=/content/publication/0a5c9314-en&_csp_=278187cb8c57cd774dd54777c P.F., 2022. Impact of CSR, innovation, and green investment on sales growth: new
d6982b0&itemIGO=oecd&itemContentType=book. evidence from manufacturing industries of China and Saudi Arabia. Economic
Ofori-Abebrese, G., Baidoo, S.T., Essiam, E., 2020. Estimating the effects of financial Research-Ekonomska Istraživanja 35 (1), 4537–4556.
inclusion on welfare in sub-Saharan Africa. Cogent Bus Manage 7 (1), 1839164. Yao, S., Yu, S., Jia, W., 2022. Does distorted allocation of capital factors inhibit green
https://doi.org/10.1080/23311975.2020.1839164. technology innovation in Chinese cities? An empirical analysis based on spatial
Ozturk, I., Ullah, S., 2022. Does digital financial inclusion matter for economic growth effect. Environmental Science and Pollution Research 1–16.
and environmental sustainability in OBRI economies? An empirical analysis. Yoshino, N., Morgan, P., 2016. Overview of financial inclusion, regulation, and
Resources, Conservation and Recycling 185, 106489. education. In: ADBI Working Paper 591. Asian Development Bank Institute, Tokyo.
Pata, U.K., Yilanci, V., Hussain, B., Naqvi, S.A.A., 2022. Analyzing the role of income Available: http://www.adb.org/publications/overview-financial-inclusion-regulat
inequality and political stability in environmental degradation: evidence from South ion-and-education/.
Asia. Gondwana Res. 107, 13–29. https://doi.org/10.1016/j.gr.2022.02.009. Yousefi, S.R., Alshamsi, H.A., Amiri, O., Salavati-Niasari, M., 2021. Synthesis,
Pesaran, M.H., 2004. General diagnostic tests for cross section dependence in panels. characterization and application of Co/Co3O4 nanocomposites as an effective
Cambridge Work Pap in Econ 0435 3, 1–39. https://ideas.repec.org/p/cam/c photocatalyst for discoloration of organic dye contaminants in wastewater and
amdae/0435.html. antibacterial properties. J. Mol. Liq. 337, 116405.
Pesaran, M.H., Yamaga, T., 2008. Testing slope homogeneity in large panels. J. Econ. Yuan, R., Liao, H., Wang, J., 2022. A nexus study of carbon emissions and financial
142, 50–93. development in China using the decoupling analysis. Environmental Science and
Raghutla, C., Chittedi, K.R., 2020. Financial development, energy consumption, Pollution Research 1–16.
technology, urbanization, economic output and carbon emissions nexus in BRICS Zafar, M.W., Shahbaz, M., Hou, F., Sinha, A., 2019. From nonrenewable to renewable
countries: an empirical analysis. Manag. Environ. Qual. Int. J. 290–307. energy and its impact on economic growth: the role of research & development
Ren, S., Hao, Y., Wu, H., 2022. How does green investment affect environmental expenditures in Asia-Pacific Economic.
pollution? Evidence from China. Environmental and Resource Economics 81 (1), Zahan, I., Chuanmin, S., 2021. Towards a green economic policy framework in China:
25–51. role of green investment in fostering clean energy consumption and environmental
Renzhi, N., Baek, Y.J., 2020. Can financial inclusion be an effective mitigation measure? sustainability. Environmental Science and Pollution Research 28 (32),
Evidence from panel data analysis of the environmental Kuznets curve. Finance Res. 43618–43628.
Lett. 37, 101725 https://doi.org/10.1016/j.frl.2020.101725. Zaidi, S.A.H., Hussain, M., Zaman, Q.U., 2021. Dynamic linkages between financial
Sattar, A., Hassan, A., Hussain, M.N., Sakhi, U., Elahi, A.R., 2022. Impact of foreign inclusion and carbon emissions: evidence from selected OECD countries. Resources
direct investment on socio-economic development in belt and road countries. Cogent Environ Sustain 4, 100022. https://doi.org/10.1016/j.resenv.2021.100022.
Economics & Finance 10, 1. https://doi.org/10.1080/23322039.2022.2143772. Zhao, L., Zhang, L., Sun, J., He, P., 2022. Can public participation constraints promote
Shabir, M., Ali, M., Hashmi, S.H., Bakhsh, S., 2022. Heterogeneous effects of economic green technological innovation of Chinese enterprises? The moderating role of
policy uncertainty and foreign direct investment on environmental quality: Cross- government environmental regulatory enforcement. Technological Forecasting and
country evidence. Environmental Science and Pollution Research 29 (2), 2737–2752. Social Change 174, 121198.
Shahbaz, M., Lean, H.H., Shabbir, M.S., 2012. Environmental Kuznets curve hypothesis
in Pakistan: cointegration and granger causality. Renew. Sustain. Energy Rev. 16,
2947–2953.

12