Recent developments in green hydrogen–environmental sustainability nexus amidst energy efficiency,green finance,eco-innovation,and digitalization in top hydrogen-consuming economies

Abstract

The present era is facing a dilemma relating to engaging the essentialities of energy resources to attain economic prosperity due to the ensuing environmental complications. Consequently, sectors such as transport, aviation, and refining industries are under the scrutiny of reducing their reliance on fossil fuels achievable with the promotion of hydrogen energy which is largely neglected in the environmental empirics. To this end, the environmental impacts of green hydrogen in the top seven hydrogen-consuming countries are assessed from 1995 to 2019. Moreover, the roles of green finance, environmental-related technologies, energy efficiency, and digitalization are considered in the model specified within the STIRPAT framework. Second generation estimators comprising cross-section autoregressive distributed lag, Common Correlated Effect Mean Group, Augmented Mean Group, and Method of Moment Quantile Regression are employed in evaluating the stated hypotheses. Feedbacks from the analysis uncovered that green hydrogen; green finance, environmental-related technologies, energy efficiencies, digitalization, and structural change promote environmental sustainability in the top seven hydrogen-consuming countries. Contrariwise, natural resource dependence and urbanization trigger CO₂ emissions, thereby exacerbating environmental complications. Based on the findings, policy measures leading toward sustainable environment are suggested.

Keywords

green hydrogen energy efficiency green finance environmental-related technologies digitalization

Introduction

The challenge of worsening climate change and the resultant environmental consequences continue to attract the center of policy debates for more than three decades despite frantic efforts from national, regional, and international governments to curb the pervasive pollution surge. There seems to be a clear indication that the ecosystem's vulnerability to degradation escalates as humanity progresses in civilization and economic development. This is predicated on the ground that global greenhouse gas (GHG) emissions in this generation era have escalated and threatened the global sustainability of the ecosystem in a manner that is incomparable to the experience of the pre-industrial age. Consequently, there are global efforts through collaboration and agreements on policies that are believed will salvage the present generation from the ravaging impacts of GHG emissions. For instance, the 2021 Climate Change Conference (COP26) of the United Nations recorded the commitments of over 193 countries on the necessity of keeping global warming below 2 degrees Celsius, believed to be detrimental to human survival and sustenance of the ecosystem.¹^{, p.26} Besides, CO₂ emissions are considered the major contributors to the aggregate volume of global GHG emissions triggered by a rise in fossil fuel consumption.² Consequently, COP26 emphasizes the need for a steady transition to 100% renewables as a precondition for reaching CO₂ neutral economy by 2050.³

The sustainability path to renewable energy remains an illusion without considering the criticality of hydrogen, among other components, for at least three reasons. First, hydrogen is a component of energy produced from various sources including nuclear power, natural gas, biogas, and renewable power sources comprising solar and wind.⁴ The possibility of deriving hydrogen from renewable sources makes it a low- CO₂ emission energy source with a high prospect of promoting sustainable development.⁵ Second, hydrogen's economic benefits are not limited to fuel or energy carriers; it is efficient in deploying power and plays a fundamental role in the transition to renewable energy. More so, it promotes environmental quality and increased economic growth, thus suggesting how critical it is in the net zero emission targets. It aids in decarbonizing the ecosystem and achieving the proposed 2050 net zero emissions target.⁶ Third, there is a rapid increase in the expansion of the global hydrogen market, with an estimated value of 129.85 billion dollars in 2021, projected to rise by 6.4% between 2022 and 2030.⁷ The preceding accentuates the criticality of hydrogen energy sources as substitutes for fossil fuels and an instrumental tool for achieving net zero emissions by 2050. Moreover, energy efficiency is another angle believed to be inevitable in minimizing the volume of energy-related CO₂ emissions surge. The efficacy of efficient energy in moderating CO₂ emission surge heightens the research interest in carbonless sources of energy in recent times.

It is instructive to clarify that the global drive for a sustainable environment transcends the transition to clean and eco-friendly energy sources to encapsulate essential policy measures advancing the roles of green finance, environmental-related technologies, and digitalization. For instance, green finance has been argued to be a critical policy option that facilitates the achievement of the green economy. Conceptually, green finance involves any form of financial service or product created to achieve favorable environmental outcomes through loans or investments that promote advancement in green projects or reduce the surge in global warming. The various instruments of green finance, comprising green bonds, green investment funds, climate change mitigation and adoption, carbon capture, green-tagged loans, air pollution abatement, and climate risk insurance, are driven through financial flows from the banking sector, insurance, micro-credit, and investment by the government, corporate business organizations, and nonprofit organization.⁸ Interestingly, the global prediction for green finance (green bonds) is estimated to reach 2.63 trillion USD by 2023.⁹ Besides, copious empirical studies confirm that green finance drives a sustainable environment by moderating GHG emissions surge.^10–12

Furthermore, environmental-related technologies have recently surfaced as viable and feasible ways of achieving global sustainability. Besides, the recent COP26 emphasizes the inevitability of technology in energy transition and carbon-neutral environment agenda.³ Among numerous arguments, the ability of technological innovations to drive increased growth rates at decreasing CO₂ emissions rates in a drive toward sustainable development has been empirically documented.¹³ The essential impacts of the digital economy in a global drive for net zero emissions is yet another stride attracting extensive research in the environment empirics. The roles of digitalization in the pathways to global sustainability are evident from the desire to derive the best from technological progress. Precisely, evidence abounds that digitalization simplifies human activities and promotes development at a low cost and huge returns without harming the ecosystem.¹⁴

Despite the foregoing narratives, researches on the environmental impacts of green hydrogen, energy efficiency, green finance, environmental-related technologies, and digitalization are scarce in the literature. This empirical scarcity is a massive disservice to the global drive for sustainable development, thus calling for an urgent study to advance this narrative empirically.

Research objective and contributions to knowledge

The dearth of empirical studies probing the green hydrogen–environmental sustainability nexus in top hydrogen-consuming economies triggers the core of empirical inquiry in the current study. Subsequently, this study primarily hopes to investigate the extent to which green hydrogen facilitates the pathways to environmental sustainability. The empirical model controls for the impacts of energy efficiency, green finance, environmental-related technologies, and digitalization. Besides, the criticality of natural resource dependence is further evaluated to extend the empirical regularity of the policy insights emanating from the study.

Drawing from the study's objectives, the contributions of this current research are sixth-fold.

First, the empirical submissions on the environmental impacts of hydrogen energy are vastly conflicting despite its potential to lead the way to global net zero emissions. For example, Jahanger et al.¹⁵ find that hydrogen energy mitigates CO₂ emissions, whereas Dash et al.¹⁶ find that it triggers CO₂ emissions. To resolve this empirical ambiguity, this study employs a recent advancement in hydrogen energy source, which is green hydrogen, otherwise called hydrogen technology which enhances the roadmap toward hydrogen economy. Besides, green hydrogen is CO₂ neutral and has a high prospect of enhancing the achievement of net zero emissions by 2050. For this reason, many advanced economies are buying into the idea of entering into a green hydrogen deal as one of the strategies for phasing out the heavy reliance on fossil fuels. For instance, countries in the European Union and related advanced economies entered into a green hydrogen deal in July 2020 to enhance the attainment of a hydrogen economy.¹⁷ Despite these potentials of green hydrogen, no studies have been conducted on how it can empirically mitigate CO₂ emissions in the top hydrogen-consuming economies. Hence, this study will constitute the first empirical investigation of the environmental effects of green hydrogen in the seven most hydrogen-endowed economies comprising China, the United States, India, Russia, the United Kingdom, Iran, and Saudi Arabia.¹⁸

Second, the engagement of green finance and environmental-related technologies in a green hydrogen-environmental sustainability model is off empirical records, thus, making this study the first to advance such novelty. Third, the study considers how digitalization and energy efficiency intervene in sustaining the environment of the top hydrogen-consuming countries. Besides, the dual engagements of the two variables are necessary to ensure smooth and hitch-free implementation of the plans and strategies toward hydrogen economy. Fourth, the current study embarks on comprehensive analyses of the green hydrogen–environment nexus both from the panel and country-specific angles.

Fifth, considering a battery of estimation techniques that address slope heterogeneity, cross-sectional dependence (CSD), endogeneity, simultaneity, and reverse causality, is a notable contribution to the extant studies. Consequently, this study utilizes a recently advanced and robust estimation technique, cross-section autoregressive distributed lag (CS-ARDL). Interestingly, CS-ARDL makes it possible for the long-run and short-run estimates to be evaluated while equally accounting for the speed of adjustment for the distortions that occurred in the short run. Additionally, the study employs three robust estimators comprising Common Correlated Effect Mean Group (CCEMG), Augmented Mean Group (AMG), and Method of Moment Quantile Regression (MMQR) in verifying the validity of the stated model. Besides, the country-specific analyses are conducted based on the novel Fully Modified Ordinary Least Square (FMOLS). Furthermore, relying on the strengths of the Stochastic Impacts by Regression on Population, Affluence, and Technology (STIRPAT) framework, this study extends the debates on the environmental impacts of green hydrogen in environment empirics. STIRPAT model accounts for non-monotonic or non-proportional impacts stemming from the various environmental forces.¹⁹ Besides, the STIRPAT model is one of the most accepted frameworks for modeling environment determinants.²⁰ Specifically, the overall estimators can provide unbiased estimates essential for deducing possible policy implications for achieving a sustainable environment in the sample countries.

Sixth, it is important to clarify that the sustainable development agenda of 2030 is impossible without resolving the lingering environmental issues. Consequently, policy implications from this study will provide practical insights into achieving the emerging global drive for hydrogen economy and carbon-neutral environment by 2050.

The empirical setting of the current research inquiry follows the succeeding structures; section one entails introduction, research, and objective/novelties. Section two focuses on stylized facts and a review of existing empirical works and section three centers on research methodologies. The presentation and economic interpretation of the findings are illustrated in section four. Section five provides details on the conclusion, relevant policies, and caveats for future research exploration.

Stylized facts: recent strides in hydrogen energy in the decarburization roadmap for the seven top hydrogen economies

This section provides detailed explanations of the key indicators in the current study, specifically as it relates to the recent developments in hydrogen energy globally and the panel of the seven top hydrogen-consuming nations. Starting with hydrogen energy contribution to the global energy mix, reports indicate positive and landmark progress. The remarkable contributions of hydrogen to the decarbornization target are apparent in sectors where CO₂ emissions abatement is complicated due to the difficulty of providing alternative energy sources to fossil fuels such as aviation, shipping, and heavy-duty transport.²¹ Currently, about 120 million tonnes of hydrogen are produced annually.²² However, there seem to be some environmental challenges attributed to the supply and demand for hydrogen. For instance, about 95% of hydrogen production comes from fossil fuels generated either through coal gasification or steam methane reforming.²² The carbonless portion entailing about 5% constitutes chlorine production generated through electrolysis, thus making the share of hydrogen from renewable sources insignificant.

The impending environmental challenges emanating from the sources of deriving hydrogen, which are mainly CO₂-inducing, accentuate the urgency and essentiality of green hydrogen. As a result, countries are committing to reach CO₂ neutrality by 2050 or earlier while ensuring their economies continue to develop.²² The recent developments in hydrogen energy are in the rising transition to hydrogen technology, as evident in the seven top hydrogen-consuming countries. For instance, the trend in Figure 1 reveals a persistent rise in green hydrogen on the aggregate from Russia, India, the United Kingdom, China, the United States, Saudi Arabia, and Iran. It is important to note that, these countries are facing a successive surge in CO₂ emissions, entrapping their environment in continuous degradation (Figure 2). The persistent rise in GHG emissions in these countries is in tandem with empirical evidence because countries such as Russia, India, the United Kingdom, China, the United States, and Saudi Arabia are major contributors to the volume of CO₂ emissions globally.

Figure 1.

Green hydrogen.

Figure 2.

CO₂ emissions.

Asides from the prospects of abating the CO₂ emissions surge that these countries can benefit from green hydrogen, they are equally advancing in environmental-related technologies and green finance (Figures 3 and 4). For instance, China engaged in numerous green finance policies nationally to promote the transition to a green economy.²³ Besides, available data uncover that 10 countries which are rated higher in green finance investment include both the United States and the United Kingdom.²⁴ Similarly, the selected leading hydrogen-consuming nations have records of good performance in terms of development and progress in energy efficiency and digitalization (Figures 5 and 6).

Figure 3.

Environmental-related technologies.

Figure 4.

Green finance.

Figure 5.

Energy efficiency.

Figure 6.

Digitalization.

The above-stylized facts suggest that there are green lights in pursuing a sustainable environment for the selected top seven hydrogen-consuming economies. However, a mere statistical plot of the trend in these indicators cannot justify any form of substantial causal nexus between these indicators and environmental sustainability. The initial submission necessitates the need to subject the relationship between the highlighted variables to empirical verification.

Literature review

The pervasiveness of climate change on environmental sustainability continues to increase despite frantic efforts from policymakers and research pundits. Consequently, copious empirical studies have been advanced to identify the moderators and promoters of global warming from different angles. To ensure a cautious review of the extant studies, this section is delineated into three subsections as follows: (i) energy structure–environment nexus; (ii) eco-innovation–environment nexus; and (iii) green finance–digitalization–environment nexus.

Energy structure–environment nexus

This section focuses on studies relating to the interlock between energy structure (comprising hydrogen, renewable energy, and energy efficiency) and environment in recent times. Feedbacks from the research studies on the impacts of energy consumption on environmental pollution are instigating renewed attention in eco-friendly energy sources. For instance, Özbay et al.² examine the channel of interaction between hydroelectricity and CO₂ emissions in China from 1985 to 2018. The empirical model controls for the impacts of globalization, urbanization, and economic growth using the famous quantile regression and quantile-on-quantile estimators. Findings reveal that hydroelectricity substantially mitigates CO₂ emissions across the quantiles. Conversely, economic growth, globalization, and urbanization exacerbate CO₂ emission surges. Dash et al.¹⁶ probe the environmental hydropower consumption in BRICS. The specification of the model includes other key indicators such as population, economic growth, and industrialization. The study reveals that hydropower consumption triggers environmental contamination. Equally, the impactful roles of economic growth, population, and industrialization are empirically confirmed. Jahanger et al.²⁵ examine the association between hydropower and CO₂ emissions based on quarterly data from 1965 to 2018 in Malaysia. The essential roles of urbanization are estimated in the model based on Quantile Autoregressive Lagged (QARDL) estimator. Findings uncover that hydropower mitigates CO₂ emissions.

Some studies focus on assessing two-way effects of hydrogen on the environment such as Chang et al.¹⁵ who evaluate the extent to which the direct and indirect effects of hydropower energy consumption dictate the variation in CO₂ emissions for a panel of top-10 hydropower energy-consuming countries from 1991 to 2018. The model specified for the nexus is evaluated based on Quantile-on-Quantile (QQ) which is noted to be suitable for time series. Findings from the analysis reveal that CO₂ emissions inversely respond to changes in hydropower energy consumption. Similarly, Alsaleh and Abdul-Rahim²⁶ report how hydropower effectively and efficiently enhances water quality in a panel study conducted for 27 EU countries from 1990 to 2019. Alnour et al.²⁷ evaluate the association of hydropower consumption with environmental quality in Sudan from 1990Q1 to 2018Q4. Similarly, the effects of urban population and economic growth are investigated with the use of SVAR model to effectively check the likely observable structural shocks. The results uncover economic growth and hydropower consumption as substantial mitigating factors for environmental pollution. Conversely, urban population growth positively drives pollution emissions in the country. Zheng et al.²⁸ examine disaggregated level impacts of hydroelectricity on fossil fuel consumption-related CO₂ emissions (FFCO₂) in Bangladesh. The components of FFCO₂ considered comprise emissions emanating from coal, gas, and oil estimated using the novel Fourier-based econometric methods which are robust for addressing issues relating to structural break. Findings from the empirical analysis expose that hydroelectricity consumption significantly moderates emissions from the highlighted components. On the flip side, the emissions are substantially escalated based on the engagements of economic globalization in the model.

The emerging efforts in resolving the challenges of global warming from the angle of energy consumption have seen continuous demand for the promotion of renewable energy. For instance, Khezri et al.²⁹ examine the association of renewable energy and CO₂ emissions in 29 Asia-Pacific countries for a period of 18 years straddling 2000−2018. The analyses include covariates such as economic complexity index, energy consumption, urbanization, and trade openness. Results uncover that renewable energy through solar and wind energy reduces CO₂ emissions. Similarly, economic complexity mitigates CO₂ emissions, whereas trade openness and urbanization trigger CO₂ emissions. Cui et al.³⁰ assess the functional impacts of renewable energy on ecological footprint from 1980 to 2017. The essential roles of urbanization, economic complexity, human capital, and economic development are examined based on FMOLS, DOLS, and canonical cointegrating regression. Feedback from the empirical model exposes that renewable energy and human capital minimize ecological footprint. Conversely, economic development, urbanization, and economic complexity induce ecological footprint. Y. Sun et al.³¹ estimate the impacts of renewable energy shocks on CO₂ emissions from 1991 to 2018 in the ten most polluted economies. The effects of globalization, economic growth, and green energy using the STIRPAT framework are based on MMQR. The results reveal renewable energy moderates renewable in lower and upper quantiles, whereas green innovation is only found efficient in mitigating CO₂ emissions in the fourth and ninth quantiles. Contrarily, globalization, population, and economic growth induce CO₂ emissions across all quantiles.

The contributions of energy consumption to the surging emissions motivate global advocacy on the need to ensure it is efficiently utilized. Within these empirical strides, Lei et al.³² investigate the extent to which energy efficiency moderates CO₂ emissions surge in the Chinese economy from 1991 to 2019. The model also considers the impacts of renewable energy while relying on the nonlinear ARDL estimator to draw the direct and indirect effects of the main regressors. Feedbacks from the empirical analyses show that negative shock in energy efficiency triggers carbon emissions, whereas positive shock impedes CO₂ emissions. Additionally, the feedback from positive and negative shocks of renewable energy moderate and promote CO₂ emissions, respectively. Akram et al.³³ estimate how energy efficiency mitigates CO₂ emissions in MINT economies comprising Mexico, Indonesia, Nigeria, and Turkey for the period stretching 1990–2014. The empirical model caters for the essential impacts of nuclear energy, renewable energy, and economic growth estimated based on the nonlinear panel autoregressive distributed lag model. The empirical results expose that energy efficiency and renewable energy mitigate and induce CO₂ emissions based on varying shocks. Murshed et al.³⁴ examine the nexuses of renewable and nuclear energy on environmental quality in G7 economies by controlling for the role of economic growth and economic complexity from 1995 to 2016. The study captures environmental quality with two variables comprising CO₂ emissions and carbon footprints. Findings reveal that nuclear energy substantially reduces both pollutants to raise the bar of environmental quality in the G7 economies. On the contrary, renewable energy escalates the surges in CO₂ emissions and carbon footprints. Jamil et al.³⁵ estimate the extent to which renewable energy and remittance cause significant changes in CO₂ emissions of selected G-20 countries from 1990 to 2019. The set of control variables employed in the study includes economic growth, financial development, and trade openness. The empirical evidence relies on FMOLS and DOLS estimators. Results support the existence of inverse nexus between renewable energy and CO₂ emissions. On the opposite end, financial development, economic growth, and remittance exacerbate the emission surge.

Eco-innovation–environment nexus

There is a growing interest in the engagement of technological innovation in environmental empirics due to the recent progress recorded globally. In this empirical perspective, Hussain et al.³⁶ examine the association between environmental-related technologies (ERTs) and consumption-based emissions from 1990 to 2016 in seven emerging economies. The study's scope extends to evaluating the role of renewable energy using CCEMG and another set of second-generation estimators. Results show that ERTs are efficient in mitigating consumption-based CO₂ emissions. Also, renewable energy turns out to negatively predict the emissions. Conversely, economic growth triggers the emissions in the sample economies. Alataş³⁷ probe the nexus between environmental technologies and transport CO₂ emissions in fifteen selected European Union economies from 1977 to 2015. The empirical evidence relies on robust estimators comprising AMG and CCEMG. Findings reveal a significant reduction in transport sector CO₂ emissions as a reaction to the impacts of environmental technologies. More so, energy consumption, economic growth, and urbanization escalate emissions contribute to the emissions from the transport sector.

Similarly, Abid et al.³⁸ examine the tripartite impacts of technological innovation and financial development from 1990 to 2019 in G8 economies. The empirical verification of the model specified is anchored on FMOLS estimator. Feedbacks show that technological innovation, financial development, and foreign direct investment reduce CO₂ emissions. Furthermore, Amin et al.³⁹ estimate how eco-innovation impacts consumption-based CO₂ emissions in the Next-Eleven economies from 1995 to 2019. In addition, the role of energy productivity is considered in a model estimated based on CS-ARDL and AMG techniques. Findings show that the emissions respond inversely to eco-innovation, exports, and energy productivity. On the contrary, imports and economic growth significantly induce a rise in the emissions. Chien et al.⁴⁰ evaluate how eco-innovation and solar energy impact CO₂ emissions in China from 1990 to 2018. Feedbacks from the analyses expose that eco-innovation and solar energy curb CO₂ emission surge for the Chinese economy. Conversely, population size and economic growth jointly escalate the emissions. Yunzhao⁴¹ focuses on how eco-innovation impacts CO₂ emissions in E7 economies from 1995 to 2018. The empirical model endogenizes renewable energy, and environmental taxes. Findings show that the three explanatory variables comprising eco-innovation, environmental taxes, and renewable energy mitigate CO₂ emissions.

Green finance–digitalization–environment nexus

The divergence in the role of financial deepening in the pervasive surge in CO₂ emissions motivates the recent advocacy and emerging attention on green finance. Consequently, numerous empirical studies are emerging toward advancing the enhancing roles of green finance in a pathway to sustainable environment. For instance, Sharif et al.⁴² assess the impacts of green finance on CO₂ emissions in G7 economies from 1995 to 2019 by controlling for the combined effects of green technological innovation. The study adopts advanced estimators that are robust for CSD and slope heterogeneity. Findings show that green finance promotes a sustainable environment by moderating CO₂ emissions. Similarly, green technological innovation supports sustainability agenda, whereas globalization and economic growth impede it. Zhao et al.¹² probe the impacts of green finance and green growth on CO₂ emissions in China from 2014 to 2018. Results uncover the enhancing effects of both exogenous indicators on a sustainable environment through a substantial reduction in CO₂ emissions. C. Sun³¹ investigates how green finance associates with CO₂ emissions by relying on the famously emerging big data and machine learning simulation tests. The simulation is conducted through the system, and the outputs correlate with the actual situation. Feedbacks show that the green finance model performs relatively well with CO₂ emissions following the constructed machine learning and big data.

Furthermore, Huang and Chen⁴³ evaluate the association of green finance with environmental quality in selected 30 provinces in China for the period straddling 2009–2017. Findings reveal that green finance positively and significantly promotes environmental quality. The study reveals the presence of identifiable threshold effects at which green finance effectively drives environmental quality. Similarly, Kirikkaleli and Adebayo⁴⁴ estimate the impacts of green finance on environmental quality in Brazil based on quarterly data running from 2001Q1 to 2018Q4. The study verifies the empirical model by utilizing dynamic ARDL to draw the feedbacks in both the long and short runs. Results indicate that green finance significantly promotes environmental quality. Similarly, political risk and green innovation support the pathways to environmental quality. Conversely, an inverse relationship is evident in the nexus between economic growth and environmental quality. Some studies have given particular focus to investigate the role of digitalization in the environment empirics. For example, Hou et al.⁴⁵ investigate how digital finance drives sustainable environment in the presence of green finance and green technological innovation in selected five provinces in China. Findings indicate that the execution of green finance initiatives is important for achieving sustainable environment. Similarly, both green technological innovation and digital finance substantially drive environmental quality. Wei and Ullah⁴⁶ evaluate the extent to which digital infrastructure supports the strides toward achieving environmental quality. In addition, the impacts of international tourism assessed in the stated model based on three estimators comprising FMOLS, DOLS, and quantile regression. Results which show that both indicators positively contribute to environmental quality are supported by feedbacks from quantile regression.

Gap in the literature

The extant literature has been flooded with empirical studies dedicated to unraveling the ambiguities enthralling the survival of the present and future ecosystems from the adverse effects of global warming. The preceding paragraphs consist of a review of the extant studies. Despite the extensive empirical studies, some lacunas are apparent. For instance, the preponderance of the empirical studies focuses more on renewable energy in combating global warming without little effort on the components of the energy sources such as hydrogen and others. Besides, the empirical submissions on hydrogen are grossly conflicting. For instance, Jahanger et al.²⁵ find that hydrogen energy mitigates CO₂ emissions, whereas Dash et al.¹⁶ find that it triggers CO₂ emissions. This suggests the need for exploring eco-friendly hydrogen which is empirically proven with the emergence of hydro technology. Moreover, the verification of hydrogen on environmental pollution in the top hydrogen economies is not empirically confirmed.

Method

This section provides details on the methodological approach adopted in carrying out the empirical examination of the research model.

Research scope, data, and source

The scope of the current research covers panel analyses for selected top seven hydrogen-consuming economies comprising China, the United States, India, Russia, the United Kingdom, Iran, and Saudi Arabia. The endogenous variable is CO₂ emissions employed as a proxy for environmental sustainability such that a reduction in CO₂ emissions is perceived as a movement toward CO₂ neutrality translated as a sustainable environment. Contrariwise, a rise in CO₂ emissions is denoted as a deterrent to decarbonizing the environment suggesting unsustainable environment. The exogenous variables are green hydrogen, energy efficiency, environmental-related technologies, green finance, digitalization, urbanization, natural resource dependence, and structural change. The years under consideration are from 1995 to 2019, motivated by numerous reasons, of which two are prominent. First, the commencement year of investigation, being 1995, is chosen because most of the key variables, such as green hydrogen, energy efficiency, and environmental-related technologies, are not available on equal grounds. Second, the end year of the study, 2019, is selected because most of the indicators are not available beyond 2019. Concise information about the variables and the sources from which they are collected are provided in Table 1.

Table 1.

Data analyses.

Indicators	Name	Measurements	Sources
CO₂	Carbon emissions	MMtonnes CO₂	EIA
GHDRO	Green hydrogen	Number of patents, with country fractional value	OECD
GRFIN	Climate change adaptation technologies	Number of patents, with country fractional value	OECD
EVTECH	Environmental related technologies	Number of patents, with country fractional value	OECD
DIG	Digitalization	Individuals using the Internet (% of population)	WDI
EGEF	Energy efficiency	Number of patents, with country fractional value	OECD
NRD	Natural resource dependence	Total natural resources rents (% of GDP)	WDI
STCH	Structural change	Services, value added (% of GDP)	WDI
URB	Urbanization	Urban population (% of total population)	WDI

Note: Energy Information Administration (EIA), Organization for Economic Cooperation and Development (OECD), World Development Indicator (WDI),

Theoretical framework

To model the environmental consequence of hydrogen in the top seven hydrogen-consuming economies, amidst the other selected covariates, the study relies on the famous STIRPAT model propounded by Dietz and Rosa.⁴⁷ STIRPAT framework constitutes one of the commonly adopted models used to explain environmental quality determinants.⁴⁸ The model hypothesizes a strong case for the role of population and affluence in enhancing rapid and persistent increases in global greenhouse gas (GHG) emissions in the present and future. The three key indicators that constitute the build-up model for the STIRPAT framework are population (P), affluence (A), and technology (T). The following equation explains the relationship.

I = β P_{i}^{ϑ 1} \times A_{i}^{ϑ 2} \times T_{i}^{ϑ 3} \times π_{i}

(1)

From Equation (1),

β

represents the constant,

σ_{1}, σ_{2}, σ_{3}

denotes the exponential factors, P represented by urban population (URB), A is denoted by service sector-led economic growth conceived as structural change (STCH), and T denoted by environmental-related technology (EVTECH) while

π

denotes the error term. An estimable version of this model in linear form entails the below equation.

\ln I_{i t} = β_{0} + σ_{1} (\ln U R B_{i t}) + σ_{2} (\ln STC H_{i t}) + σ_{2} (\ln EVTEC H_{i t}) + π_{i}

(2)

Research hypotheses

The economic intuitions justifying the directions of why and how the exogenous variables influence significant variation in the endogenous variable are explained in this section. This becomes essential to enhance a detailed understanding of the reactions and interactions among the indicators in the empirical model. Regarding green hydrogen–CO₂ emissions nexus, there are emerging facts advancing how green hydrogen facilitates the achievement of green economy, leading to sustainable environment. This submission is supported by Dong et al.,⁵ thus leading us to maintain indirect association between green hydrogen and CO₂ emissions as follows $\frac{δ C O_{2}}{δ GHDRO} < 0$ . The pervasiveness of energy consumption in inducing GHG emissions surge motivates the global call for efficient utilization of energy resources. Besides, copious researches confirm the moderating effects of energy efficiency on CO₂ emissions. Hence, we hypothesize an inverse relationship between the two indicators as thus $\frac{δ C O_{2}}{δ EGEF} < 0$ .

The environmental impacts of green finance have been documented from the CO₂ mitigating angle.^11,31 Based on this empirical position, we anticipate an indirect nexus between green finance and CO₂ emissions thus $\frac{δ C O_{2}}{δ GRFIN} < 0$ . There are burgeoning arguments supporting the efficacy of environmental-related technologies in sustaining global environment. Expressly, appreciable strands of studies confirmed that environmental-related technologies mitigate CO₂ emissions.⁴⁹ In that case, an indirect nexus between technological innovations and CO₂ emissions is hypothesized thus $\frac{δ C O_{2}}{δ EVTECH} < 0$ . The gradual adoption of digitalization in virtually all spheres of human endeavor is motivated by the empirical confirmation of the substantial roles of digitalization in reducing CO₂ emissions.^50–52 Following the preceding submission, we predict that CO₂ emissions will respond negatively to changes in digitalization thus $\frac{δ C O_{2}}{δ DIG} < 0$ . The consequential impacts of production activities stemming from the agriculture and most importantly the manufacturing sector have raised concerns on the need to be conscious of carbon emissions contributed by economic activities. Consequently, there are emerging interests on the need to concentrate on the service sector. The concerns are borne out of the fact that the service sector forms the most important part of structural change that has mitigated CO₂ emission surge. According to Adebayo et al.,⁵³ structural change is an instrumental tool that is capable of militating against persistent surge in CO₂ emission. Consequently, the relationship is stated thus $\frac{δ C O_{2}}{δ STCH} < 0$ . The impacts of natural resource dependence have empirically proven to be CO₂-enhancing,⁵⁴ leading to the conclusion of direct relationship $\frac{δ C O_{2}}{δ NRD} > 0$ . The rate of increase in population growth has significant impacts on the sustainability of the environment. Besides, population growth has significantly influenced the rising level of urbanization and quite notable strands of empirical submission expose that urbanization triggers CO₂ emissions. Consequently, a positive nexus can be hypothesized between urbanization and CO₂ emissions as follows $\frac{δ C O_{2}}{δ URB} > 0$ .

Model specification

The model illustrating the relationship between the exogenous and endogenous variables is specified based on the existing studies^13,49 with some modifications as thus stated:

C O_{2 i t} = σ_{0} + σ_{1} GHDR O_{i t} + σ_{2} EGE F_{i t} + σ_{3} GRFI N_{i t} + σ_{4} EVTEC H_{i t} + σ_{5} DI G_{i t} + σ_{5} NR D_{i t} + σ_{5} STC H_{i t} + σ_{5} UR B_{i t} + α_{i t}

(3)

Given that CO₂ denotes CO₂ emissions, GHDRO signifies green hydrogen, EGEF represents energy efficiency, GRFIN signifies green finance, EVTECH denotes environmental-related technologies, DIG implies digitalization, NRD represents natural resource dependence, STCH stands for structural change, and URB represents urbanization. The stochastic term in the above equation is stated as

α

whereas the cross-section and time are represented by t and i.

Estimation procedures

Following the rule of thumb and the standard procedures evident in the extant studies, these steps are summarized in Figure 7 and further elaborated in the subsequent subsections.

Figure 7

Flow chart of empirical procedures.

Slope homogeneity and Cross-Sectional Dependency tests

The relationship among economic indicators in a panel model is often characterized by the issue of common factors that can distort the model of providing accurate predictions. Consequently, testing for the presence of CSD has become a standard empirical procedure for deciding on the best estimator between first and second-generation groups. Hence, the presence of CSD is often considered a priority in recent empirical studies.^1,13,55 Besides, slope heterogeneity is another critical econometric problem that can lead to erroneous results in panel regression. The presence of slope heterogeneity suggests that all slope coefficients are the same across all cross-sectional units. The CSD model states thus;

CSD = \sqrt{\frac{2 T}{N (N - 1) N}} . \sum_{i = 1}^{N - 1} \sum_{j = i + 1}^{N} Cor r_{i, l}

(4)

Similarly, the slope homogeneity test equation is given below.

{\tilde{Δ}}_{SH} = {(N)}^{\frac{1}{2}} (2 k)^{- \frac{1}{2}} (\frac{1}{N} \tilde{S} - k)

(5)

{\tilde{Δ}}_{ASH} = {(N)}^{\frac{1}{2}} {(\frac{2 k (t - K - 1)}{T + 1})}^{- \frac{1}{2}} (\frac{1}{N} \tilde{S} - 2 k)

(6)

Based on the above-stated equations,

SH

signifies delta tilde and

ASH

represents adjusted delta tilde.

Panel unit root and cointegration tests

The inevitability of avoiding erroneous results in the parameter estimates provided by the panel regression model necessitates the validity of stationarity tests.⁵⁶ In the presence of CSD, second generation unit root estimators are often adopted. Moreover, the cross-sectionally augmented IPS (CIPS) estimator credited to Pesaran⁵⁷ is used in favor of the second-generation specified as follows:

Δ Q_{i t} = σ_{i} + σ_{i} R_{i, t - 1} + σ_{i} {\bar{Q}}_{t - 1} + \sum_{m = 0}^{ρ} σ_{i l} Δ \bar{Q_{t - 1}} + \sum_{m = 0}^{ρ} σ_{i l} Δ {\bar{Q}}_{t - 1}

(7)

The equation depicting t-statistic for the cross-sectionally augmented IPS can be specified as follows:

\bar{CIPS} = \frac{1}{N} \sum_{i = 1}^{n} CADF

(8)

The next step after ascertaining the nature of unit root is the long-run test The existence of slope heterogeneity and CSD suggests second generation cointegration test with Westerlund⁵⁸ is the most appropriate. The model can be specified as follow.

σ_{i} (U) Δ y_{i t} = η 1_{i t} + η 2_{i t} + δ (η_{i t} - 1 - {σ^{'}}_{i} W_{i t} - 1) + φ_{i} (U) θ_{i t} + ϖ_{i}

(9)

μ_{1 i} = δ_{i} (1) β_{2 i} - δ_{i} θ_{1 i} + δ_{i} β_{2 i} and η 2_{i} = - δ θ 2_{i}

(10)

The subsequent equation depicts the various t-statistics for Westerlund cointegration.

G_{t} = \frac{1}{N} \sum_{i - 1}^{N} \frac{{σ^{'}}_{i}}{S E (σ^{'})}

(11)

G_{σ} = \frac{1}{N} \sum_{i - 1}^{N} \frac{T {σ^{'}}_{i}}{σ^{'} (1)}

(12)

P_{T} = \frac{{σ^{'}}_{i}}{S E (σ^{'})}

(13)

P_{σ} = T σ^{'}

(14)

The criteria for evaluating the validity of the model are based on the group statistics (

G a

and

G t

) and panel statistics (

P a

and

P t

Cross-Sectional Autoregressive Distributed Lag

Confirming the long-run relationship in a panel model will validate the essentiality of evaluating how the variables interact in the short-run and long-run. The conventional practice in the empirical studies suggests giving precedence to CS-ARDL proposed by the estimator.⁵⁹ The rationale behind adopting the CS-ADRL estimator bothers on the estimator's ability to address econometric issues relating to endogeneity, slope heterogeneity, and CSD in panel models.⁶⁰ The model states thus

Δ Y_{i, t} = ψ_{i} + \sum_{l = 1}^{ρ} ψ_{i l} Δ Y_{i t} - 1 + \sum_{l = 0}^{ρ} ψ_{i l}^{'} IND V_{s, i, t - 1} + \sum_{l = 0}^{1} ψ_{i l}^{'} {\bar{CSM}}_{i, t - 1} + ω_{i, t}

(15)

The validity and reliability of the CS-ARDL estimator are evaluated through the adoption of additional estimators comprising CCEMG and AMG.

Common Correlated Effects Mean Group and Augmented Mean Group

The choice of employing CCEMG bothers the associated issues relating to heterogeneity and CSD of slope coefficient. The Common Correlated Effects (CCE) estimation technique is credited to the original work of ⁶¹ and augmented by.⁶² Among different strengths, CCE is robust for econometric issues relating to nonstationary, structural breaks, and unobserved common factors. The equation depicting CCE method states thus:

y_{i t} = α_{1 i} + φ_{i} χ_{i t} + σ_{i} f_{t} + ϖ_{i t}

(16)

It should be noted that

χ_{i t}

and

y_{i t}

are observables in Equation (16) whereas

ϕ_{i}

denotes the time series slope on the exogenous indicators. More so,

α_{1 i}

represents the panel fixed effects indicating heterogeneous time-invariant slope across the panel, the unobserved common factor is denoted by

f_{t}

while the factor loadings are represented by

σ_{i}

, and

ϖ_{i t}

is error terms. Equation (16) can be expanded by estimating the mean of the cross-sectional units for the endogenous and exogenous indicators as follow;

y_{i t} = α_{1 i} + φ_{i} χ_{i t} + δ_{i} y_{i t}^{'} + θ_{i} χ_{i t} + ϖ_{i t}

(17)

Moreover, some issues concerning autocorrelation and heteroscedasticity limit the utilization of the Ordinary Least Square (OLS) estimator in evaluating Equation (17). However, the mean group (MG) method can be employed to resolve the econometric issue emanating from heterogeneous slope which is obtainable by taking the mean of each coefficient for the individual regression as thus:

C C E M G = N^{- 1} \sum_{i = 1}^{N} \hat{β}

(18)

Besides, the AMG method initiated by ⁶³ can equally be derived as follow;

A M G = N^{- 1} \sum_{i = 1}^{N} \tilde{β}

(19)

Fully Modified Ordinary Least Squares

The current research expands the frontier of knowledge in hydrogen-environment debates by substantiating the panel analyses with time series analyses for each of the top hydrogen-consuming countries. The time series analysis is done based on the novel FMOLS estimator ⁶⁴ which is robust for the issues relating to slope heterogeneity.⁶⁵ The FMOL estimator is based on ⁶⁶ which accounts for both heterogeneous panel and heterogeneous group mean evaluated using two statistics. First, three statistics comprising $Z_{\hat{v}, N, T}$ , $Z_{\hat{ρ} N, T - 1}$ , and $Z_{t N, T}$ are evaluated following the pooled residuals based on the within-the dimension. It can be stated thus:

Z_{\hat{v}, N, T} = T^{2} N^{3 / 2} \sum_{t = 1}^{N} \sum_{t = 1}^{T} L_{11 i}^{2} {\hat{ℓ}}_{i, t 1}^{2}

(20)

Z_{\hat{ρ} N, T 1} = T \sqrt{N} \sum_{i = 1}^{N} {\sum_{t = 1}^{T} {\hat{L}}_{11 i}^{2} {\hat{ℓ}}_{i, t 1}^{2}}^{1} \sum_{t = 1}^{N} \sum_{t = 1}^{T} {\hat{L}}_{11 i}^{2} ({\hat{ℓ}}_{i, t 1} Δ {\hat{ℓ}}_{i, 1} {\hat{λ}}_{i})

(21)

Z_{t N, T} = {\tilde{σ}}_{N, T} \sum_{I = 1}^{N} {\sum_{T = 1}^{T} {\hat{L}}_{11 i}^{2} {\hat{ℓ}}_{i, t 1}^{2}}^{1 / 2} \sum_{t = 1}^{N} \sum_{t = 1}^{T} {\hat{L}}_{11 i}^{2} ({\hat{ℓ}}_{i, t 1} Δ {\hat{ℓ}}_{i, 1} {\hat{λ}}_{i})

(22)

Given that

{\hat{ℓ}}_{i, t 1}

vectors the residual in the OLS estimator.¹

The second statistic bothers on the pooled residuals stated along the between-the dimension of the panel that is found robust for a non-homogeneous autocorrelation parameter in the cross-section. It can be specified as follows:

{\tilde{Z}}_{\hat{ρ} N, T}^{1} = \sum_{t = 1}^{N} \sum_{t = 1}^{T} {\hat{ℓ}}_{i, t}^{2}^{1} \sum_{t = 1}^{T} ({\hat{ℓ}}_{i, t 1} Δ {\hat{ℓ}}_{i, t} {\hat{λ}}_{i})

(23)

{\tilde{Z}}_{t N, T}^{1} = \sum_{t = 1}^{N} \sum_{t = 1}^{T} {\hat{ℓ}}_{i, t}^{2} 1^{1 / 2} \sum_{t = 1}^{T} ({\hat{ℓ}}_{i, t 1} Δ {\hat{ℓ}}_{i, t} {\hat{λ}}_{i})

(24)

Equations (23) and (24) are employed to derive the group means statistics of each time series. The five models can be asymptotically distributed in the following equation:

\frac{X_{N, T} μ \sqrt{N}}{\sqrt{v}} \Rightarrow N (0, 1),

(25)

Following from Equation (25)

X_{N, T}

replicates the above-stated test statistics. Moreover,

μ

and

ν

signify the mean and the variance of each test The critical values for illustrated test statistics are available in table 2 of Pedroni.⁶⁷

Method of Moments Quantile Regression

The current study employs the Method of Moments Quantile Regression (MMQR) credited to Machado and Silva.⁶⁸ The MMQR estimator is a robust estimation technique that accounts for the presence of outliers and is efficient in the case of unobserved slope heterogeneity inherent in a panel model. Besides, MMQR estimator provides estimates based on conditional heterogeneous covariance impacts of the exogenous indicators on CO₂ emissions. This is achieved by facilitating the individual impacts of the regressors to influence the variation in the entire distribution rather than just shifting means. Following Machado and Silva,⁶⁸ the model explicating the MMQR estimator can be stated thus:

Q_{Y} (τ / X_{i t}) = (α_{i} + δ_{i} q (τ)) + X_{i t}^{'} β + Z_{i t}^{'} y q (τ)

(26)

Pre-estimation analyses

This research relies on four approaches to explicate the nature of the dataset employed in examining the environmental impacts of green hydrogen amidst other covariates in hydrogen-consuming countries. These channels include summary statistics, normality tests, correlation matrix, and trend analysis. The motivation for the analyses is anchored on the opportunity it will provide for having a good idea about the nature of the indicators as they relate to the sample countries. To start with, Table 2 showcases the various values relating to the summary statistics. As evident in the table, the average value for the seven countries is 2429 suggesting a high value of CO₂ emissions emanating from each selected country. This value is corroborated by the movement displayed in Figure 8, where CO₂ emissions have unstable movements in the last two decades with most parts being on the increase. The average value for green hydrogen stands at 53.2, environmental-related technologies average 1226.4, green finance has a mean value of 216.5, digitalization averages 41.2, and energy efficiency averages 91.8. These averages are suggestive of the fact that the respective indicators perform well in the selected countries, an assertion that is supplemented by the trend analysis presented in Figure 8. The consciousness of the top hydrogen-consuming nations to flatten the curve of CO₂ emissions is evident in the average value of natural resource dependence which is 9.4 with an apparently declining trend in the recent time as displayed in Figure 8. The effort to decarbonize the economy is equally apparent from the high value of service sector contribution to GDP in the selected countries which is supported by a rising trend as presented in Figure 8. Urbanization seems to be on the rise giving an average value of 67.9 which is supported by increasing movements in Figure 8.

Figure 8.

Trend analyses of the selected indicators.

Table 2.

Summary statistics and normality analysis.

Indicators	CO₂	GHDRO	EVTECH	GRFIN	DIG	EGEF	NRD	STCH	URB
Mean	2429.89	53.19	1226.39	216.47	41.23	91.82	9.35	55.32	67.97
Median	1313.51	11.00	214.25	50.48	42.30	13.33	3.41	53.77	79.10
Maximum	10,773.25	313.85	10,555.73	2276.22	96.50	767.25	55.48	77.44	86.12
Minimum	212.67	0.33	0.33	1.00	0.00	0.25	0.24	31.13	26.61
Std. Dev.	2786.92	85.28	2376.81	448.79	32.64	179.17	12.87	12.44	20.23
Skewness	1.52	1.65	2.44	2.67	0.09	2.34	1.99	0.10	−1.06
Kurtosis	4.45	4.35	8.07	9.43	1.47	7.41	6.13	1.84	2.43
Jarque-Bera	83.07	92.87	361.18	509.27	17.24	302.37	186.70	10.02	35.23
Probability	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.01	0.00
Observations	175	175	175	175	175	175	175	175	175
Correlation matrix
CO₂	1
GHDRO	−0.06	1
EVTECH	0.51	0.28	1
GRFIN	0.50	0.28	0.31	1
DIG	0.24	0.48	0.37	0.35	1
EGEF	0.52	0.23	0.45	0.41	0.35	1
NRD	−0.33	−0.27	−0.31	−0.29	−0.13	−0.30	1
STCH	0.11	0.47	0.37	0.30	0.34	0.45	−0.52	1
URB	−0.23	0.42	0.18	0.22	0.41	0.11	0.28	0.40	1

Besides, the dataset exhibits some significant level of abnormality as apparent from the values of the skewness, kurtosis, and Jarque-Bera in Table 2. Moreover, the correlation matrix in Table 2 suggests the model is free from multicollinearity that could distort reliable prediction on the extent to which the exogenous variables predict the endogenous variables.

Empirical results and discussion

Cross-sectional dependence, correlation, and slope homogeneity results

Table 3 denotes the outcomes of the tests for CSD and slope homogeneity which are necessary to determine the appropriateness of estimation techniques between the first and second generations. As evident from the outcomes, the null hypothesis of independence in the panel model cannot be accepted because of the significant level of the probability values which are less than 5%. Interestingly, these values are significant at a 1% level emphasizing the coherence of the results in explicating considerable interloping among the variables across the cross-section units. Likewise, a high correlation is noticeable in the model with a specific range between 61% and 95%. This outcome can lead us to posit that macroeconomic instability in one of the seven selected top hydrogen-consuming countries could result in momentous variation in others.

Table 3.

Panel unit roots and cointegration tests outcomes.

Indicators	CIPS (2007)		IPS (2003)
Indicators	Level	First difference	Level	First difference
CO2	−2.158	−4.277***	−1.950	−4.382***
GHDRO	−2.086	−5.385***	−1.542	−5.323***
GRFIN	−2.383	−4.939***	−1.971	−5.948***
EVTECH	−1.501	−4.441***	−1.386	−4.885***
DIG	−1.115	−4.698***	−1.615	−7.995***
EGEF	−2.081	−5.015***	−1.903	−5.308***
NRD	−2.106	−3.229***	−1.866	−4.901***
STCH	−2.330	−3.229***	−1.470	−4.529***
URB	−0.407	−2.994**	−1.153	−3.914***
Westerlund cointegration results
Statistic		Value	Z-value	p value
Gt		−2.635**	−1.723	.043
Ga		−5.256***	−3.113	.003
Pt		−7.447***	−2.823	.002
Pa		−9.635**	−1.784	.037

The significant levels at 1%, 5%, and 10% represented by ***, **, and *, respectively.

The outcomes of the slope homogenous test are provided in Table 4 with the major findings suggesting that the null hypothesis of the homogeneous slope coefficient cannot be accepted. This submission is posited based on the significant delta and adjusted delta tildes statistics. Consequently, we conclude that the first-generation unit root tests are not appropriate to conduct the empirical verification rather the second-generation unit root tests remain the most efficient and appropriate.^3,13

Table 4.

Dependency and slope homogeneous outcomes.

Indicators	Pesaran (2015)	Pesaran (2004)	Correlation
CO2	10.121***	10.120***	.714
GHDRO	14.149***	14.150***	.649
GRFIN	10.121***	10.120***	.714
EVTECH	17.321***	17.320***	756
DIG	21.790***	21.792***	.951
EGEF	9.293***	9.290***	.607
NRD	14.548***	14.550***	.835
STCH	13.827***	13.830***	.703
URB	20.952***	20.950***	.914
Slope homogeneity	Delta	8.241***
	Adj. Delta	9.712***

The significant levels at 1%, 5%, and 10% represented by ***, **, and *, respectively.

Panel unit root and cointegration results

The confirmation of slope interdependence and heterogeneity in the cross-section units leading to the appropriateness of the second-generation unit root method influences the adoption of the cross-sectionally augmented panel unit root IPS (CIPS) test Similarly, to ensure strong empirical regularity of the emanating results, we employ the first-generation unit root tests as robustness. The results in Table 3 uncover that all the variables have unit roots at levels. Nonetheless, the series become stationary after subjecting them to the first difference. Similarly, the outcomes from the first-generation stationarity tests support the outcomes of CIPS. The cointegration test results in Table 5 support the existence of long-run association among the indicators going by the group (Gt and Ga) and panel (Pt and Pa) statistical values.

Table 5.

Results of the panel quantile regression.

Indicators	Endogenous variable: CO₂ emissions
	Lower quantiles			Middle quantiles			Upper quantiles
	Q10	Q20	Q30	Q40	Q50	Q60	Q70	Q80	Q90
LGHDR	–0.075 (0.046)	–0.079 (0.042)	–0.075 (0.040)	–0.147 (0.080)	–0.242*** (0.057)	–0.292*** (0.046)	–0.344*** (0.049)	–0.225*** (0.045)	–0.131*** (0.03)
LGRFIN	–0.336*** (0.043)	–0.325*** (0.037)	–0.204*** (0.067)	–0.345*** (0.083)	–0.181* (0.095)	–0.171** (0.076)	–0.229*** (0.082)	–0.181** (0.075)	–0.101** (0.049)
LEVTECH	–0.264*** (0.055)	–0.302*** (0.047)	–0.332*** (0.087)	–0.306*** (0.107)	–0.514*** (0.123)	–0.567*** (0.098)	–0.612*** (0.106)	–0.571*** (0.096)	–0.507*** (0.064)
LDIG	–0.124*** (0.025)	–0.123*** (0.022)	–0.093** (0.04)	–0.212*** (0.049)	–0.157*** (0.057)	–0.107** (0.045)	–0.097** (0.049)	–0.054** (0.024)	–0.045** (0.022)
LEGEF	–0.249*** (0.05)	–0.214*** (0.043)	–0.210*** (0.078)	–0.264*** (0.097)	–0.162 (0.111)	–0.081** (0.039)	–0.108** (0.046)	–0.029** (0.012)	–0.017*** (0.005)
LSTCH	−5.208*** (0.442)	−5.167*** (0.379)	−4.755*** (0.694)	−3.965*** (0.856)	−3.984*** (0.983)	−3.927*** (0.788)	−4.187*** (0.849)	−5.073*** (0.771)	−4.733*** (0.509)
LNRD	0.110* (0.059)	0.186*** (0.051)	0.197*** (0.033)	0.182*** (0.025)	0.168*** (0.013)	0.131*** (0.015)	0.115** (0.054)	0.095** (0.063)	0.086** (0.041)
LURB	0.774*** (0.223)	0.720*** (0.191)	0.508*** (0.135)	0.713* (0.431)	1.013** (0.495)	1.164*** (0.397)	1.394*** (0.428)	1.698*** (0.388)	1.421*** (0.256)
_cons	21.551*** (1.068)	21.591*** (0.915)	21.085*** (1.678)	17.052*** (2.069)	16.084*** (2.375)	15.392*** (1.904)	15.681*** (2.053)	18.358*** (1.863)	18.588*** (1.231)
Pseudo R2	0.534	0.540	0.539	0.555	0.575	0.577	0.599	0.569	0.588
Observations	175	175	175	175	175	175	175	175	175

The significant levels at 1%, 5%, and 10% represented by ***, **, and *, respectively. Brackets denote standard errors.

Long-run results

Panel empirical results

To arrive at far-reaching policy insights, this research relies on CS-ARDL estimator. Furthermore, CCEMG and AMG estimators are employed to enhance the robustness of the CS-ARDL results. The results in Table 6 show that green hydrogen mitigates CO₂ emissions in the short and long run. This is noticeable from the negatively signed and significant green hydrogen coefficients suggesting that a one percent increase in green hydrogen leads to a corresponding decrease in CO₂ emissions. The effects of green finance are observed to be significant and negative, implying an inverse nexus with CO₂ emissions. Consequently, a substantial improvement in green finance leads to a proportionate decline in CO₂ emissions. Besides, environmental-related technologies in Table 7 negatively and significantly impact CO₂ emissions in the short and long run. Digitalization exerts negative effects on CO₂ emissions in the long-run implying that advancement in the digital economy leads to a substantial decline in CO₂ emissions. More so, energy efficiency and structural change significantly and negatively impact CO₂ emissions in the short and long run. Conversely, triggering effects are observable from natural resource dependence and urbanization on CO₂ emissions, suggesting that both indicators induce substantial increases in CO₂ emissions. It is instructive to clarify that the empirical results emanating from long-run estimators employed as robustness checks comprising CCEMG and AMG provide sturdy support to justify the validity and reliability of the CS-ARDL findings. Going by the findings of the error correction term, the correction of the distortion in the short run can be corrected at 80% speed of adjustment. The reported corrections are evident following the fulfillment of the ECT to the rule of thumb relating to negative and significant coefficients. The various outcomes of the empirical analyses are presented in Figure 9.

Figure 9.

Graphical presentation of environmental sustainability's drivers in top hydrogen-consuming economies.

Table 6.

Country results of the green hydrogen–environmental sustainability nexus.

Variables/coefficients	Australia	Canada	China	India	Russia	Saudi Arabia	The United States
LGHDRO	−0.245** (0.089)	−0.352*** (0.102)	−0.422* (0.212)	−0.119* (0.069)	−0.213*** (0.069)	−0.332*** (0.099)	−0.145** (0.055)
LGRFIN	−0.168*** (0.059)	−0.136** (0.066)	−1.507*** (0.155)	−1.089 (0.711)	−0.047 (0.077)	−4.595*** (0.109)	−0.174*** (0.052)
LEVTECH	−0.168* (0.096)	−0.129** (0.059)	−0.803*** (0.189)	−0.172** (0.056)	−0.137** (0.065)	−2.106*** (0.889)	−0.624*** (0.117)
LDIG	−1.175*** (0.321)	−2.170*** (0.560)	−2.953** (1.363)	−6.191*** (1.676)	1.598** (0.447)	−0.539*** (0.125)	−2.673*** (0.743)
LEGEF	−0.676*** (0.003)	−0.633*** (0.105)	−4.063*** (0.954)	−3.347 (3.308)	−0.645 (0.434)	−3.859** (1.445)	−0.762** (0.298)
LSTCH	−0.474*** (0.056)	−0.447*** (0.198)	−3.169*** (1.221)	−0.658** (0.248)	−8.993 (6.123)	−0.138*** (0.087)	−1.612*** (0.225)
LNRD	0.223 (0.861)	1.589 (2.627)	3.549*** (0.310)	2.190*** (0.446)	1.462*** (0.388)	7.312*** (1.876)	3.397*** (0.245)
LURB	10.624*** (3.589)	22.603*** (5.223)	14.866*** (4.301)	22.568*** (4.223)	13.043*** (2.223)	5.514* (2.761)	3.193** (1.229)
C	−32.118 (24.927)	−22.436 (16.005)	15.204* (8.267)	−45.037*** (4.671)	−75.463 (70.266)	44.879** (15.887)	41.4623* (22.015)
R-squared	0.961	0.898	0.997	0.994	0.791	0.974	0.932
Adjusted R-squared	0.944	0.854	0.996	0.991	0.699	0.962	0.902

The significant levels at 1%, 5%, and 10% represented by ***, **, and *, respectively. Brackets denote standard errors.

Table 7.

Empirical results from short-run and long-run estimates.

Indicators	Endogenous variable: CO₂ emissions
	CS-ARDL		CCEMG	AMG
	Short-run	Long-run	CCEMG	AMG
LGHDR	−0.028*** (0.009)	−0.012** (0.005)	−0.024*** (0.004)	−0.003 (0.009)
LGRFIN	−0.038** (0.016)	−0.030** (0.012)	−0.145*** (0.027)	−0.006** (0.003)
LEVTECH	−0.103* (0.058)	−0.059* (0.031)	−0.090*** (0.028)	−0.104*** (0.022)
LDIG	−0.020 (0.018)	−0.107***	−0.023***	−0.015***
LDIG	−0.020 (0.018)	(0.026)	(0.005)	(0.003)
LEGEF	−0.029*** (0.005)	−0.018** (0.008)	−0.030*** (0.014)	−0.004 (0.017)
LSTCH	−0.845*** (0.147)	−0.459* (0.267)	−0.944* (0.556)	−0.481* (0.288)
LNRD	0.056*** (0.017)	0.028** (0.014)	0.047** (0.023)	0.025*** (0.007)
LURB	0.848*** (0.088)	0.428*** (0.039)	0.984*** (0.045)	0.649 (0.527)
ECT (−1)	−0.798*** (0.126)
RMSE			0.012	0.023
WALD			36.300***	17.053***
Observations	168	168	175	175

The significant levels at 1%, 5%, and 10% represented by ***, **, and *, respectively. Brackets denote standard errors.

Country-specific empirical results

The essentiality of extending the novelties of this paper to knowledge frontier on the hydrogen–environmental sustainability nexus motivates the consideration of country-specific analyses of the nexus using the novel fully modified OLS (FMOLS). The results in Table 6 reveal that green hydrogen substantially moderates the surge in CO₂ emissions across the individual country of the top seven hydrogen-consuming economies. Hence, a percentage rise in the rate of green hydrogen consumption in these countries leads to a corresponding decline in CO₂ emissions. The environmental impacts of green finance are substantially noticed in Australia, Canada, China, Saudi Arabia, and the United States suggesting an inverse relationship between green hydrogen and CO₂ emissions in the economies. Environmental-related technologies significantly moderate CO₂ emissions in the seven selected countries as evidenced by the negative and statistical significance of environmental-related technologies on CO₂ emissions. Similarly, digitalization exerts mitigating effects on CO₂ emissions across the selected countries. Structural change hinders carbon emissions surge in all the countries except Russia.

The inducing roles of natural resource dependence are statistically supported in China, India, Russia, Saudi Arabia, and the United States. Intuitively, a percentage increase in natural resource depletion leads to a substantial rise in the stock of CO₂ emissions thereby complicating the pollution tragedy embattling the concerned economies. Besides, urbanization proves to be a key driver of CO₂ emissions across the seven countries. This implies that an increase in urban settlement occasioned by the influx of people from the rural to the urban area will escalate the share of urban contributions to the total GHG in the various top hydrogen-consuming countries. It should equally be noted that the predictive power of the model in explaining the extent to which changes in the exogenous variables (comprising green hydrogen, green finance, environmental related technologies, digitalization, energy efficiency, structural change, natural resource dependence, and urbanization) lead to variation in the endogenous indicator (CO₂ emissions) is accentuated by the high values of both R-squared and adjusted R-squared ranging between 79% and 99%.

Discussion

Panel empirical fallouts

The present study explores the impacts of green hydrogen on environmental sustainability amidst the intervening roles of energy efficiency, environmental-related technologies, digitalization, structural change, natural resource dependence, and urbanization on CO₂ emissions in the top seven hydrogen-consuming economies. The research objective is pursued through the novel STIRPAT framework estimated based on second-generation estimators comprising CS-ARDL, CCEMG, and AMG. Feedbacks from the empirical analyses show that green hydrogen promotes environmental sustainability by mitigating the surge in CO₂ emissions. Consequently, promoting the consumption of green hydrogen energy as an alternative to fossil fuels will not only reduce environmental vulnerability to pollution but also reduce the huge reliance on fossil fuels in sectors such as aviation, transport, and other industries utilizing fossil fuels as energy sources.

These empirical outcomes corroborate previous studies such as Dong et al.⁵ and Wang⁶⁹ that find empirical support for the mitigating impacts of green hydrogen and hydroelectricity on CO₂ emissions.

The empirical outcomes uncover the moderating role of green finance on CO₂ emissions in the selected sample countries as obvious from the negative signs of the green finance indicators in the short and long run. The economic intuition from this outcome is that persistent advancement in green finance will pave the way for flattening the curve of CO₂ emissions. Besides, since eco-friendly projects and investments such as green bonds, air pollution abatement, CO₂ capture, and CO₂ mitigation and adaptation, among others, are enhanced through the notion of green finance, an increase in such projects and investment is expected to drive environmental sustainability through a significant reduction in CO₂ emissions. These results are well supported by the submissions from Zhao et al.¹² and C. Sun.³¹ The estimated model reveals that environmental-related technologies hinder a significant increase in CO₂ emissions, suggesting that they are positive predictors of the sustainable environment. This outcome is intuitional and aligns well with the reality of what is operating in recent times. Specifically, environmental-related technologies such as wastewater treatment, industrial emissions elimination, recycling and water management, waste to energy, vertical garden, and farms are effective and efficient in decarbonizing the ecosystem. Besides, notable strands of empirical studies allude that environmental-related technologies facilitate the achievement of net zero emissions..^13,70

The impacts of digitalization on CO₂ emissions mitigations are only significant in the long-run, which suggests that it takes some considerable periods before the adoption of digitalization could effectively mitigate CO₂ emissions. Besides, it takes some level of economic status for the individual and household to adopt digitalization in economic activities. The empirical findings of Adha et al.,⁵⁰ Ren et al.,⁵² and Balogun et al.⁷¹ support the notion that digitalization mitigates CO₂ emissions. The moderating role of energy efficiency is significant only in the long-run whereas insignificant impacts are reported in the short run. This is inconsonant with Lei et al.³² and R. Li et al.⁷² who provide empirical support for energy efficiency in inhibiting incessant surges in CO₂ emissions. The impacts of structural change are negatively significant on CO₂ emissions in the short and long run across the top seven hydrogen-consuming countries. Consequently, a substantial increase in service sector's contribution to GDP will significantly reduce CO₂ emissions surge. The plausibility of this result lies in the fact that the service sector emits less or insignificant CO₂ to the ecosystem which is suggestive of why the advanced economies are transiting from manufacturing-based economies to service-based. The moderating roles of structural change driven by the service sector have been empirically documented.¹³ The impacts of natural resource dependence on CO₂ emissions are positive and significant in the short and long run suggesting that as the countries deplete the available stock of natural resources, the environment is polluted. This submission is well supported by the empirical findings in the study conducted by Shen et al.⁵⁴ and Ibrahim and Ajide.⁷³ Urbanization is noted to drive CO₂ emissions surge in the estimated short and long-run models. Intuitively, a persistent increase in the rate of population concentrated in urban areas escalates the level of CO₂ emissions. This result is equally arguable on the ground that industrial activities are concentrated more in the urban areas with the majority emitting considerable volume of CO₂ in the atmosphere. The empirical submissions in Cui et al.³⁰ and Y. Chen et al.⁷⁴ support the results in this study.

Country-level empirical fallouts

The empirical findings in Table 6 corroborate the panel findings in Table 7 considering that green hydrogen substantially reduces CO₂ emissions. The results are plausible considering the contributions of the seven countries to global volume of hydrogen consumption. Besides, it is a confirmation of the submission of advancing green hydrogen as the essential energy source to complement other clean energy strides focusing on substantial reduction in CO₂ emissions in a stride toward net zero emissions.²² The emerging roles of green finance in the promotion of sustainable environment through the significant reduction in CO₂ emissions are apparent in five countries consisting of Australia, Canada, China, Saudi Arabia, and the United States. However, India and Russia are yet to have substantial level of investment in green finance that could trigger a significant decline in CO₂ emissions. The significant impacts of environmental-related technologies in moderating CO₂ emissions across the seven countries further corroborate empirical confirmation submitted by recent studies.^70,75 The mitigating effects of digitalization on CO₂ emissions corroborate the panel outcomes. Similarly, energy efficiency, structural change, natural resource dependence, and urbanization supplement the findings reported in the panel analyses.

Disintegration of the impacts of green hydrogen and other regressors using the method of moment quantile regression

The empirical verification of the impacts of green hydrogen on sustainable environment in the top seven hydrogen-consuming countries are elaborated through the MMQR estimator. Based on the outcomes in Table 5, the impacts of green hydrogen are substantial in mitigating CO₂ emissions from the 50th to 90th quantiles implying that the early stage of hydrogen energy consumption is not substantial enough to offset the existing CO₂ emissions driven by fossil fuels. In other words, the result could be perceived as accentuating the fact that shifting from fossil fuels to hydrogen energy would require some time lags for fulfilling replacement and CO₂ mitigation to take place. Green finance, environmental-related technologies, digitalization, energy efficiency, and structural change significantly moderate CO₂ emissions across the quantiles. Conversely, natural resource dependence and urbanization promote the surge in CO₂ emissions across the quantiles. The graphical presentation of the quantile regression is presented in Figure 10. Similarly, a graphical comparison of the results that emanate from the four estimators is presented in Figure 11. This is necessary to have a glance view of how the robust estimators faired in the current research.

Figure 10.

Quantile regression plot.

Figure 11.

Plots of coefficient.

Outcomes of panel causality nexus

The feedback on the heterogeneous panel causality is provided in Table 8 for the direction of the relationship between each of the exogenous variables and CO₂ emissions in the selected seven top hydrogen-consuming nations. Based on the findings, it is evident that bidirectional causality exists between green hydrogen and CO₂ emissions such that policy measures implemented to promote the consumption of green hydrogen in the net zero emissions target will significantly influence the substantial reduction in CO₂ emissions. Similarly, policy measures implemented to curtail the surge in CO₂ emissions can stimulate an increase in the consumption of clean energy such as green hydrogen through efforts to phase out fossil fuels. Bidirectional causality is confirmed between green finance, environmental-related technologies, energy efficiency, and natural resource dependence on CO₂ emissions. By implication, policy measures implemented to promote investment in green finance, increase the level of advancements in technologies, and enhance efficient use of energy will lead to the substantial reduction in CO₂ emissions. Additionally, CO₂ emission mitigation and adaptation policies implemented toward the CO₂ neutrality target could require raising the level of investment in green finance, promoting technological innovation, and efficiently utilizing energy resources in the seven top hydrogen-consuming nations.

Table 8.

Findings of the heterogeneous panel causality.

Null Hypothesis:	W-Stat.	Zbar-Stat.	Decision
LGHDR → LCO₂	5.211**	2.256	Bidirectional
LCO₂ → LGHDR	4.223*	1.931	Bidirectional
LGRFIN → LCO₂	5.662***	3.538	Bidirectional
LCO₂ → LGRFIN	13.739***	11.915	Bidirectional
LEVTECH → LCO₂	4.815***	2.660	Bidirectional
LCO₂ → LEVTECH	17.765***	16.089	Bidirectional
LDIG→LCO₂	4.381**	2.209	Unidirectional
LCO₂ → LDIG	2.721	0.488	Unidirectional
LEGEF → LCO₂	4.533**	2.368	Bidirectional
LCO₂ → LEGEF	5.719***	3.598	Bidirectional
LSTCH → LCO₂	5.123***	4.236	Bidirectional
LCO₂ → LSTCH	2.116	1.087
LNRD → LCO₂	5.209***	3.069	Unidirectional
LCO₂ → LNRD	2.195	0.365	Unidirectional
LURB → LCO₂	3.907*	1.719	Unidirectional
LCO₂ → LURB	3.645	1.446

The significant levels at 1%, 5%, and 10% represented by ***, **, and *, respectively. Brackets denote standard errors.

Unidirectional causality is evident between natural resource dependence and CO₂ emissions suggesting that policy measures that promote the depletion of natural resources will instigate substantial increases in CO₂ emissions. Furthermore, unidirectional causality is apparent in the nexuses of digitalization and urbanization with CO₂ emissions. Intuitively, policy measures executed to promote digitalization would enhance carbon emission mitigation while economic plans that drive urbanization drive CO₂ emissions. A graphical representation of the panel causal nexus is presented in Figure 12(a). Moreover, the country-specific causality for each of the seven countries is presented diagrammatically in Figure 12(b)–(h). It is evident from the figures the causalities reveal heterogeneous trends across the countries ranging from bidirectional to unidirectional and no causality. For instance, while unidirectional causality is evident between structural change and CO₂ emissions in Australia, a bidirectional nexus is evident in China. More so, no causality was reported between green finance, energy efficiency, and CO₂ emissions in India. Besides, energy efficiency and structural change did not cause CO₂ emissions in Russia.

Figure 12.

(a) Panel causal relationship. (b) Causal relationship in Australia. (c) Causal relationship in Canada. (d) Causal relationship in China. (e) Causal relationship in India. (f) Causal relationship in Russia. (g) Causal relationship in Saudi Arabia. (h) Causal relationship in the United States.

Conclusion, policy insights, and caveats

The various viewpoints that emanate from the analyses of the current study which lead to the derivation of essential policy implications and the possible limitations are discussed in the following subsections.

Conclusion

The pervasiveness of climate change constitutes the core hindrance to attaining sustainable ecosystem globally. To lay this long-embattled issue to rest, policymakers, governments, international organizations, and research pundits are working unanimously to arrive at practical solutions in halting the unceasing rise in CO₂ emissions. Consequent to the foregoing, this study examines the impacts of green hydrogen on CO₂ emissions in selected top seven hydrogen-consuming countries from 1995 to 2019. The empirical model controls for the effects of green finance, environmental-related technologies, energy efficiencies, digitalization, structural change, natural resource dependence, and urbanization. The study aligns with the usual empirical procedures evident from extant studies by conducting various tests and the feedbacks are instrumental in choosing the second-generation estimators comprising CS-ARDL, AMG, CCEMG, and MMQR.

The empirical findings from the estimated model expose that green hydrogen; green finance, environmental-related technologies, energy efficiencies, digitalization, and structural change promote environmental sustainability in the top seven hydrogen-consuming countries. This is apparent from the significant reduction in CO₂ emissions in reaction to the impacts of the aforementioned indicators. In contrast, natural resource dependence and urbanization induce the surge in CO₂ emissions thereby exacerbating environmental complications. The preceding submissions are important for supporting sustainable environment in the top seven hydrogen-consuming nations. Results from CS-ARDL, AMG, and CCEMG are corroborated by MMQR through examination of the nexus in three categories comprising lower, middle, and upper quantiles.

Policy insights

Some relevant and viable policies observed to be highly fundamental in driving the pathways to environmental sustainability in the top seven hydrogen-consuming countries. Consequently, the following policy insights are thus provided.

The roles of green hydrogen in the estimated model promote environmental sustainability by mitigating CO₂ emissions, suggesting that green hydrogen energy is supportive of the net zero emissions targets. Consequently, governments and policymakers in the top seven hydrogen-consuming nations should implement policy measures that will enhance the complete transition to hydrogen technology. Besides, since fossil fuel-based hydrogen production and consumption has been noted to drive CO₂ emission, the government should mandate key industrial activities such as producing fertilizer, treating metals, refining petroleum, and processing foods to adopt green hydrogen which is renewable energy-based and promotes decarbonization. Moreover, industries such as transport, aviation, and other fossil fuel-dependent industries should be supported to embrace green hydrogen.

Green finance is observed to drive CO₂ neutrality across the panel of top hydrogen-consuming countries. Consequently, policy measures that enhance green projects and investment should be promoted. The strides ongoing in China, the United Kingdom, and the United States in terms of massive investment in green finance should be extrapolated to the other seven top hydrogen-consuming economies. The government should persuade private bodies and nonprofit organizations to partake in promoting investment in green bonds, green projects, green-tagged loans, and climate risk insurance.

The functionality of technology in promoting sustainable environment of the top seven hydrogen-consuming nations is well proven from the empirical front. This indicates that government should strive more to promote technology in all facets of the economy.

The environmental impacts of energy efficiency are noted to support and drive sustainable environment. This implies that promoting efficient energy consumption at homes leading to less energy utilization for cooking, heating, cooling, and running appliances is a great opportunity to pave the way for a sustainable environment. Companies and businesses should be encouraged to observe and pursue the efficient utilization of energy resources. The implementation of energy-efficient policies becomes more important for countries like China, the United States, Russia, and India, which are major drivers of global GHG emissions. It is believed promoting energy efficiency in these countries will not only flatten the curve of CO₂ emissions in their environment but also reduce the global volume of GHG emissions.

The emerging efforts in promoting the digital economy should be embraced by the top hydrogen-consuming economies owing to the empirically documented mitigating impacts of digitalization on CO₂ emissions leading to the sustainability of the environment. Besides, the structural change driven by the service sector should be supported and promoted from the policy front. The triggering impacts of natural resource dependence can be curtailed through CO₂ taxation and diversification of the economy to the real sectors. The inducing roles of urbanization on CO₂ emissions can be controlled through the sponsorship of infrastructural development in the rural areas and the encouragement of industrial concentration in the rural areas to create jobs for the rural habitats.

Caveats and future research opportunities

The present study has been able to provide reliable empirical assessment of the impacts of green hydrogen on environmental sustainability. That notwithstanding, few limitations are noticeable as thus. First, the disaggregated of different components of hydropower and hydroelectricity can be of immeasurable assistance in reaching a robust policy implication. Second, the environmental impacts of green hydrogen on other variants of pollutants such as nitrous oxide, methane, P.M._2.5, and sulfur dioxide could also provide extended empirical views of how hydrogen contributes to the overall level of environmental sustainability. Third, the current inquiry can be extended to some selected continents like the European Union, Sub-Saharan African countries, and the Asian region. The preceding limitations are worthwhile research opportunities that future studies could explore to ensure global sustainability in the post-COP26 era.

Authors’ contribution statement

Jiahao Shen: conceptualization, modeling, and data collection. Lanre Ibrahim Ridwan: conceptualization, literature review, and proofreading. Lukman Raimi: modeling, analysis, and policy insights. Mamdouh Abdulaziz Saleh Al-Faryan: proofreading and policy insights.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Jiahao Shen

Lanre Ibrahim Ridwan

Mamdouh Abdulaziz Saleh Al-Faryan

Notes

Jiahao Shen is currently a doctoral candidate in School of Foreign Studies, Nanjing University and a research assistant in Institute of African Studies, Nanjing University. His research interest focuses on African economic studies, interdisciplinary research of terminology and translation, language testing, and so forth.

Lanre Ibrahim Ridwan is a potential research and a PhD candidate at University of Lagos, Department of Economics. His research interest focuses on energy economics, environment, international trade, and climate change conferences (COP26).

Lukman Raimi is an Assistant Professor of Entrepreneurship, the Coordinator of the Entrepreneurship major and the Chairman of the learning and development committee (LDC) at the Universiti Brunei Darussalam. He is a member of the Brunei Mentors for Entrepreneurs Network (BMEN), the ASEAN Mentors for Entrepreneurs Network (AMEN), the Founder's Institute USA, and a mentor for the Tony Elumelu Foundation. He is also on the editorial board of several international journals.

Mamdouh Abdulaziz Saleh Al-Faryan is an international researcher with special interest in corporate governance and finance, development, environment, and finance.

References

Ibrahim

. Beyond COP26: can income level moderate fossil fuels, carbon emissions, and human capital for healthy life expectancy in Africa? Environ Sci Pollut Res.2022; 29(58): 87568–87582.

Özbay

Athari

Saliba

, et al. Towards environmental sustainability in China: role of globalization and hydroelectricity consumption. Sustainability 2022; 14: 4182

Ibrahim

. Post-COP26: can energy consumption, resource dependence, and trade openness promote carbon neutrality? Homogeneous and heterogeneous analyses for G20 countries. Environ Sci Pollut Res 2022; 29(57): 86759–86770.

Energy Information Administration, E. I. A. Use of hydrogen—U. S. Energy Information Administration (EIA) 2022, June. https://www.eia.gov/energyexplained/hydrogen/use-of-hydrogen.php

Dong

Yang

, et al. A green hydrogen credit framework for international green hydrogen trading towards a carbon neutral future. Int J Hydrogen Energy 2022; 47: 728–734

Yusaf

Laimon

Alrefae

, et al. Hydrogen energy demand growth prediction and assessment (2021–2050) using a system thinking and system dynamics approach. Appl Sci 2022; 12: 2 Article 2.

Hossain Lipu

Miah

MdS

Ansari

, et al. (2021). Data-driven hybrid approaches for renewable power prediction toward grid decarbonization: applications, issues and suggestions. J Cleaner Prod, 328, 129476

United Nations Environment Programme, U.N. Green Financing . UNEP – UN Environment Programme. 2018, January 23. http://www.unep.org/regions/asia-and-pacific/regional-initiatives/supporting-resource-efficiency/green-financing

George

Haas

Joshi

(eds) 2022). Introduction to the business of sustainability: an organizing framework for theory, practice and impact. In Handbook on the Business of Sustainability. Cheltenham Glos: Edward Elgar Publishing. https://doi.org/10.4337/9781839105340.00007.

10.

Al Mamun

Boubaker

Nguyen

. Green finance and decarbonization: evidence from around the world. Finance Res Lett 2022; 46: 102807

11.

Ibrahim

Al-mulali

Ozturk

, et al.

On the criticality of renewable energy to sustainable development: Do green financial development, technological innovation, and economic complexity matter for China?

Renewable Energy. 2022; 199: 262–277.

12.

Zhao

Taghizadeh-Hesary

Dong

, et al. How green growth affects carbon emissions in China: the role of green finance. Econom Res Ekonom Istraživanja 2022; 0: 1–22

13.

Ibrahim

Ozturk

Al-Faryan

MAS

, et al. Exploring the nexuses of disintegrated energy consumption, structural change, and financial development on environmental sustainability in BRICS: modulating roles of green innovations and regulatory quality. Sustain Energ Tech Assess 2022b; 53: 102529

14.

Alshubiri

Jamil

Elheddad

. The impact of ICT on financial development: Empirical evidence from the Gulf Cooperation Council countries. Int J Eng Bus Manag 2019; 11: 184797901987067.

15.

Chang

Ali

, et al. How does hydropower energy asymmetrically affect environmental quality? Evidence from quantile-based econometric estimation. Sustain Energ Technol Assess 2022; 53: 102564

16.

Dash

Sethi

. Designing hydro-energy led economic growth for pollution abatement: evidence from BRICS. Environ Sci Pollut Res 2022; 29: 31252–31269

17.

World Hydrogen Energy Conference, W. H. E. C. 23. World Hydrogen Energy Conference WHEC-2022. Bidgecongress. 2022, June. https://www.bidgecongress.org/event/23-world-hydrogen-energy-conference-whec-2022/

18.

Sönnichsen

. Global biofuel production by country 2021 . Statista 2022, July. https://www.statista.com/statistics/274168/biofuel-production-in-leading-countries-in-oil-equivalent/

19.

Sheng

Tang

. Spatiotemporal variation patterns of water pollution drivers: the case of China’s south-north water transfer project. Sci Total Environ 2021; 761: 143190

20.

Guo

Wang

. Analysis of the spatial relevance and influencing factors of carbon emissions in the logistics industry from China. Environ Sci Pollut Res 2022; 29: 2672–2684

21.

international Energy Agency, international E. A. Hydrogen – Analysis. IEA. 2022, September. https://www.iea.org/reports/hydrogen

22.

Africa-EU Energy Partnership. Green hydrogen: bridging the energy transition in Africa & Europe—Africa–EU energy partnership . 2022, September. https://africa-eu-energy-partnership.org/publications/green-hydrogen-bridging-the-energy-transition-in-africa-europe/

23.

Zhang

. Relationship and integrated development of low-carbon economy, food safety, and agricultural mechanization. Environ Sci Pollut Res 2021; 28: 68679–68689

24.

Saeed Meo

Karim

MZA

. The role of green finance in reducing CO2 emissions: an empirical analysis. Borsa Istanb Rev 2022; 22: 169–178

25.

Jahanger

Awan

, et al. The impact of hydropower energy in Malaysia under the EKC hypothesis: evidence from quantile ARDL approach. SAGE Open 2022; 12(3): 21582440221109580.

26.

Alsaleh

Abdul-Rahim

. Moving toward sustainable environment: the effects of hydropower industry on water quality in EU economies. Energ Environ 2022; 33: 1304–1325

27.

Alnour

Ali

Abdalla

, et al.

How do urban population growth, hydropower consumption and natural resources rent shape environmental quality in Sudan?

World Dev Sustain 2022; 1: 100029

28.

Zheng

Wei

Murshed

, et al. Repercussions of hydroelectricity use on carbon emissions in Bangladesh: evidence using novel Fourier-bootstrapped ARDL and Fourier-gradual shift causality analyses. Eval Rev 2022. https://doi.org/10.1177/0193841X221135674

29.

Khezri

Heshmati

Khodaei

. Environmental implications of economic complexity and its role in determining how renewable energies affect CO2 emissions. Appl Energ 2022; 306: 117948

30.

Cui

Weng

Nadeem

, et al. Exploring the role of renewable energy, urbanization and structural change for environmental sustainability: comparative analysis for practical implications. Renew Energ 2022; 184: 215–224

31.

Sun

. The correlation between green finance and carbon emissions based on improved neural network. Neural Comput Appl 2022; 34: 12399–12413

32.

Lei

Xie

Hafeez

, et al. Assessing the dynamic linkage between energy efficiency, renewable energy consumption, and CO2 emissions in China. Environ Sci Pollut Res 2022; 29: 19540–19552

33.

Akram

Umar

Xiaoli

, et al. Dynamic linkages between energy efficiency, renewable energy along with economic growth and carbon emission. A case of MINT countries an asymmetric analysis. Energ Rep 2022; 8: 2119–2130

34.

Murshed

Saboori

Madaleno

, et al. Exploring the nexuses between nuclear energy, renewable energy, and carbon dioxide emissions: the role of economic complexity in the G7 countries. Renew Energ 2022; 190: 664–674

35.

Jamil

Liu

Gul

, et al. Do remittance and renewable energy affect CO2 emissions? An empirical evidence from selected G 20 countries. Energ Environ 2022; 33: 916–932

36.

Hussain

Mir

Usman

, et al. Analysing the role of environment-related technologies and carbon emissions in emerging economies: a step towards sustainable development. Environ Technol 2022; 43: 367–375

37.

Alataş

. Do environmental technologies help to reduce transport sector CO2 emissions? Evidence from the EU 15 countries. Res Transp Econ 2022; 91: 101047

38.

Abid

Mehmood

Tariq

, et al. The effect of technological innovation. FDI, and financial development on CO2 emission: evidence from the G8 countries. Environ Sci Pollut Res 2022; 29: 11654–11662

39.

Amin

Zhou

Safi

. The nexus between consumption-based carbon emissions, trade, eco-innovation, and energy productivity: empirical evidence from N-11 economies. Environ Sci Pollut Res 2022; 29: 39239–39248

40.

Chien

Hsu

C-C

Andlib

, et al. The role of solar energy and eco-innovation in reducing environmental degradation in China: evidence from QARDL approach. Integr Environ Assess Manag 2022; 18: 555–571

41.

Yunzhao

. Modelling the role of eco innovation, renewable energy, and environmental taxes in carbon emissions reduction in E−7 economies: evidence from advance panel estimations. Renew Energ 2022; 190: 309–318

42.

Sharif

Saqib

Dong

, et al. Nexus between green technology innovation, green financing, and CO2 emissions in the G7 countries: the moderating role of social globalisation. Sustain Dev 2022. https://doi.org/10.1002/sd.2360

43.

Huang

Chen

. The spatial spillover and threshold effect of green finance on environmental quality: evidence from China. Environ Sci Pollut Res 2022; 29: 17487–17498

44.

Kirikkaleli

Adebayo

. Political risk and environmental quality in Brazil: role of green finance and green innovation. Int J Finance Econom 2022. https://doi.org/10.1002/ijfe.2732

45.

Hou

Zhu

Wang

, et al. Will green financial policy help improve China’s environmental quality? The role of digital finance and green technology innovation. Environ Sci Pollut Res 2022: 1–13. https://doi.org/10.1007/s11356-022-22887-z

46.

Wei

Ullah

. International tourism, digital infrastructure, and CO2 emissions: fresh evidence from panel quantile regression approach. Environ Sci Pollut Res 2022; 29: 36273–36280

47.

Dietz

Rosa

. Effects of population and affluence on CO2 emissions. Proceedings of the National Academy of Sciences 1997; 94(1): 175–179.

48.

Khan

SAR

Ibrahim

Al-Amin

, et al. An ideology of sustainability under technological revolution: striving towards sustainable development. Sustainability 2022a; 14: 8 Article 8

49.

Abbasi

Hussain

Haddad

, et al. The role of financial development and technological innovation towards sustainable development in Pakistan: fresh insights from consumption and territory-based emissions. Technol Forecast Soc Change 2022; 176: 121444

50.

Adha

Hong

C-Y

Agrawal

, et al. ICT, carbon emissions, climate change, and energy demand nexus: the potential benefit of digitalization in Taiwan. Energ Environ 2022. https://doi.org/10.1177/0958305X221093458

51.

Chen

. How CO2 emissions respond to changes in government size and level of digitalization? Evidence from the BRICS countries. Environ Sci Pollut Res 2022; 29: 457–467

52.

Ren

Hao

, et al.

Digitalization and energy: how does internet development affect China’s energy consumption?

Energ Econom 2021; 98: 105220

53.

Adebayo

Oladipupo

Rjoub

, et al. Asymmetric effect of structural change and renewable energy consumption on carbon emissions: designing an SDG framework for Turkey. Environ Dev Sustain 2022: 1–29. https://doi.org/10.1007/s10668-021-02065-w

54.

Shen

Ibrahim

Ajide

, et al.

Tracking environmental sustainability pathways in Africa: do natural resource dependence, renewable energy, and technological innovations amplify or reduce the pollution noises?

Energ Environ 2022. https://doi.org/10.1177/0958305X221124221

55.

Usman

Balsalobre-Lorente

Jahanger

, et al.

Pollution concern during globalization mode in financially resource-rich countries: do financial development, natural resources, and renewable energy consumption matter?

Renew Energ 2022b; 183: 90–102

56.

Khan

SAR

Ibrahim

Al-Amin

, et al. An ideology of sustainability under technological revolution: striving towards sustainable development. Sustainability 2022b; 14: 4415

57.

Pesaran

. A simple panel unit root test in the presence of cross-section dependence. J Appl Econometrics 2007; 22: 265–312

58.

Westerlund

. Testing for error correction in panel data. Oxf Bull Econ Stat 2007; 69: 709–748

59.

Chudik

Pesaran

. Large panel data models with cross-sectional dependence: a survey. SSRN Electronic J 2013. https://doi.org/10.2139/ssrn.2319840

60.

Usman

Ozturk

Naqvi

SMMA

, et al. Revealing the nexus between nuclear energy and ecological footprint in STIRPAT model of advanced economies: fresh evidence from novel CS-ARDL model. Prog Nucl Energy 2022a; 148: 104220

61.

Pesaran

. Estimation and inference in large heterogeneous panels with a multifactor error structure. Econometrica 2006; 74: 967–1012

62.

Kapetanios

Pesaran

Yamagata

. Panels with non-stationary multifactor error structures. J Econom 2011; 160: 326–348

63.

Eberhardt

Bond

. Cross-section dependence in nonstationary panel models: a novel estimator [MPRA Paper]. 2009, October 7. Howard House Wagon Lane Bingley https://mpra.ub.uni-muenchen.de/17692/

64.

Pedroni

. Fully modified OLS for heterogeneous cointegrated panels. In Baltagi

Fomby

Carter Hill

(Eds.), Nonstationary panels, panel cointegration, and dynamic panels. Emerald Group Publishing Limited. (Vol. 15, 2001; 93–130). https://doi.org/10.1016/S0731-9053(00)15004-2

65.

Omokanmi

Ibrahim

Ajide

, et al.

Exploring the dynamic impacts of natural resources and environmental pollution on longevity in resource-dependent African countries: does income level matter?

Res Pol 2022; 79: 102959

66.

Pedroni

. Panel cointegration: asymptotic and finite sample properties of pooled time series tests with an application to the PPP hypothesis. Econ Theory 2004; 20: 597–625

67.

Pedroni

. Critical values for cointegration tests in heterogeneous panels with multiple regressors. Oxf Bull Econ Stat 1999; 61: 653–670

68.

Machado

Silva

. Quantiles via moments. Journal of Econometrics 2019; 213(1): 145–173.

69.

Wang

Meng

Zhou

, et al. Novel coal-to-methanol process with near-zero carbon emission: pulverized coal gasification-integrated green hydrogen process. J Cleaner Prod 2022; 339: 130500

70.

Chu

. Determinants of ecological footprint in OCED countries: do environmental-related technologies reduce environmental degradation? Environ Sci Pollut Res 2022; 29: 23779–23793

71.

Balogun

A-L

Marks

Sharma

, et al. Assessing the potentials of digitalization as a tool for climate change adaptation and sustainable development in urban centres. Sustain Cities Soc 2020; 53: 101888

72.

Wang

. The impact of energy efficiency on carbon emissions: evidence from the transportation sector in Chinese 30 provinces. Sustain Cities Soc 2022; 82: 103880

73.

Ibrahim

Ajide

. The dynamic heterogeneous impacts of nonrenewable energy, trade openness, total natural resource rents, financial development and regulatory quality on environmental quality: evidence from BRICS economies. Res Pol 2021; 74: 102251

74.

Chen

Lee

C-C

Chen

. Ecological footprint, human capital, and urbanization. Energ Environ 2022; 33: 487–510

75.

Doğan

Chu

Ghosh

, et al.

How environmental taxes and carbon emissions are related in the G7 economies?

Renew Energ 2022; 187: 645–656

76.

Sun

Anwar

Razzaq

, et al. Asymmetric role of renewable energy, green innovation, and globalization in deriving environmental sustainability: evidence from top-10 polluted countries. Renew Energ 2022; 185: 280–290