Pursuing a carbon-neutral world: How digital exports,financial stability,and energy security contribute to green growth

Abstract

Digital exports, financial stability, and energy security are vital in determining green growth. However, no empirics in the past have shed light on the relationship between digital exports, financial stability, energy security, and green growth. The main focus of the analysis is how digital exports, financial stability, and energy security affect green growth in 33 world's leading energy-consuming economies from 2000 to 2021. To that end, the study employed the novel cross-sectionally augmented autoregressive distributed lag (CS-ARDL) model. Our findings show that digital exports and financial stability boost long-run green growth in the full sample and in Asian, European, and American models, while energy security risks hinder long-run green growth in the full sample and in Asian, European, and American models. Environmental technology promotes long-run green growth in full-sample and in European and American models, and renewable energy consumption helps boost green growth in full-sample and European models. In the short run, financial stability is crucial for green growth in the full sample and in Asian and American models, while environmental technology benefits green growth in all models. Integrating digital exports, financial stability, and sustainable energy practices for ecologically sustainable and commercially rewarding green growth in a fast-changing global context is essential.

Keywords

Digital exports financial stability energy security green growth

Introduction

The discussion of green economic development as a fresh perspective on economic activity emerged from the need to address ecological issues and the effects of climate change. Regarding green economic development, the social benefits of protecting the world's ecology and ecosystems are just as essential as the economic rewards. According to Fay,¹ green growth (GG) may promote social equality and wellbeing, enhancing circumstances for human habitation on Earth. Mastini et al.² reported green economic development goes beyond economic growth. It assures that existence on Earth has societal benefits. Countries have developed a variety of strategies and visions to define the route to greening economic sectors to meet the goal of green economic development (e.g. Vietnam's green economic growth strategy for 2021–2030; Korea's “Low Carbon, Green Growth” vision; The European Green Deal). However, several academics^3,4 believe that many nations will find it hard to implement GG for various reasons, most notably the need for a more stable financial structure.

Paris Agreement, in its Article 10, also emphasized the development of “technology framework.” This framework plays a crucial role in facilitating the execution of the Agreement and attains the long-term objective of technological development and transfer.⁵ To mitigate carbon emissions and achieve carbon neutrality objectives, the first and foremost step that needs to be taken by the nations is to have an in-depth analysis of the technological requirements related to the environment. Technology framework of the Paris Agreement helps identify the developing nations’ technological requirements within the context of carbon neutrality.⁶ Article 10 of the Paris Agreement helps developing economies foster the generation of eco-friendly technologies by providing them with suggestions and facilitating the transfer of such technologies, supporting these nations in achieving carbon neutrality objectives.⁷ Thus, the pursuit of a carbon-neutral world is crucial for fostering the GG phenomenon.

The allocation and circulation of financial resources in an economy, which is the driving force behind capital accumulation, technical advancement, and investment, is enabled by financial stability. According to Chen et al.,⁸ financial stability encourages FDI and the exchange of technology within the nations, both drivers of GG. The studies by Li et al.⁹ and Chen et al.¹⁰ address the impact of financial development and digital finance on a nation's green output. The pathways of green technological advancement and renewable energy development have been accelerated through financial markets and digital financial tools, ultimately leading to improved GG. According to Razzaq and Yang,¹¹ developing financial sector innovations promotes sustainable development, reduces energy poverty, and distributes money to businesses easily. According to Ozturk and Ullah,¹² digital finance promotes financial inclusion, a prerequisite for green development. Thus, a more stable and innovative financial structure is crucial for a nation to achieve a high pace of economic growth without damaging the ecosystem due to its ability to provide capital for green technologies and practices. The study by Li¹³ reported that financial stability has favorable role in green economic recovery.

Innovations in the financial and all other sectors largely depend on the digitalization of society, which offers a wide range of opportunities to promote GG by enhancing resource efficiency, reducing emissions, and enabling sustainable practices across various sectors of the economy. Digitalization must be ecologically friendly to maximize its beneficial effects on green development.¹⁴ Digital exports bring digitalization by generating demand for technology. Businesses that grow their presence in the global digital market often make use of digitalization.¹⁵ Digital exports have the potential to stimulate green development via a channel of digitalization. Businesses often use digitization to connect with a global audience as they broaden their reach beyond national boundaries via digital exports.¹⁶ Environmentally friendly business models and sustainable technology result from this revolutionary process.¹⁷ The study by Sun et al.¹⁸ noted that digital trade promotes GG by reducing carbon emissions through smart technologies in sustainable practices in developing countries. Due to these reasons, digital exports promote environmentally friendly behavior and facilitate the transition to a more sustainable global economy, serving as a key factor in GG.

Another crucial factor in driving GG is the nation's energy sector. The course of expansion and development for any economy heavily depends on energy. All economic operations, including travel, communication, manufacturing operations, agricultural output, healthcare, schooling, and security, rely on energy availability in economies worldwide. This underscores the vital role of energy in economic operations. Energy is the “lifeblood” of economies, and the global economy's expansion will depend on a sufficient, dependable, and inexpensive energy supply.¹⁹ Energy is essential for the efficient operation of every economic sector. Therefore, ensuring a stable energy supply is crucial, as supply fluctuations can increase energy security risks and potentially impede economic activity. Numerous economic and development objectives must be accomplished to maintain them. Any nation that aspires to join the industrialized economies must have a strong energy infrastructure to accomplish sustainable growth. The nation's energy demands must be met by this foundation of energy both now and in the future. Energy security can positively affect GG by increasing investment in renewable energy. Similar outcomes are reported by Alsagr and Ozturk²⁰ within a global framework. A summary of the literature is reported in Table 1.

Table 1.

Summary of literature review.

Author(s)	Country/region	Time span	Techniques	Variables	Outcomes
Danish et al.²¹	93 developed and developing countries	2005–2019	Bias-corrected method of moments	Digital trade and economic performance	Positive
Fu et al.²²	China	2000Q1–2020Q4	Bootstrap ARDL	Digital trade and ecological sustainability	Positive
Xiong & Luo²³	OECD	1990–2019	MMQR	Digital trade and green productivity	Positive
Murshed²⁴	South Asian countries	2000–2016	CUP-FM, CUP-BC, FGLS	ICT trade and environmental sustainability	Positive
Zhang²⁵	30 Chinese provinces	2002–2018	Panel quantile regression	Digital trade openness and green total factor productivity	Positive
Sun et al.¹⁸	15 Asian countries	1995–2021	GMM	Digital trade and green growth	Positive
Zhang et al.²⁶	G7	2000–2022	CUP-FM and CUP-BC	Digital trade and ecological footprints	Negative
Kashif et al.²⁷	OECD	2001–2019	Random effects and FGLS	ICT service exports and CO₂ emissions	Negative
Zhu et al.²⁸	30 Chinese provinces	2013–2021	Fixed effect	Digital trade and CO₂ emissions	Negative
Li et al.²⁹	283 Chinese cities	2005–2021	Quasi-natural experiment	Digital trade and carbon emissions efficiency	Positive
Chen et al.⁸	European Union countries	2000–2020	Tobit	Macroeconomic stability and green development	Positive
Cao et al.³⁰	China	2011–2018	Entropy weight method	Financial development and green growth	Positive
Zheng et al.³¹	China	1996–2020	QARDL	Financial integration and green growth	Positive
Safi et al.³²	G7	1990–2018	CS-ARDL	Financial stability and consumption-based CO₂ emissions	Negative
Baloch et al.³³	Saudi Arabia	1971–2016	ARDL	Financial instability and CO₂ emissions	Insignificant
Oanh & Dinh³⁴	Vietnam	2004–2022	Nonparametric analysis	Financial stability and sustainable development	Positive
Dinh et al.³⁵	40 developed and developing countries	2005–2020	Bayesian analysis	Financial stability and sustainable development	Positive
Khan & Yoon³⁶	88 developing countries	1980–2014	FMOLS	Financial instability and CO₂ emissions	Positive
Osabohien et al.³⁷	West Africa	2015–2022	System GMM	Financial stability and sustainable development	Positive
Ozili & Iorember ³⁸	26 Asian and European countries	2011–2018	System GMM	Financial stability and sustainable development	Positive
Su et al.³⁹	Iceland	1995–2019	Fourier ADL and NARDL	Financial stability and CO₂ emissions	Negative
Alsagr & Ozturk²⁰	Global	1996–2021	CS-ARDL	Energy security risk and green investment	Positive
Le & Nguyen ⁴⁰	Global	2002–2013	FGLS	Energy security and economic growth	Positive
Kartal⁴¹	Turkey	1980–2018	NARDL	Energy security risk and economic growth	Negative
Banna et al.⁴²	Global	1980–2021	Propensity score matching	Energy security risk and GDP growth rate	Negative
Pata & Karlilar⁴³	24 OECD countries	1995–2018	AMG	Energy security risk and fossil fuel material footprint.	Negative
Ayaz et al.⁴⁴	14 Asian countries	1996–2019	PCSE	Energy insecurity and economic growth	Negative

AMG: augmented mean group; CS-ARDL: cross-sectionally augmented autoregressive distributed lag; CUP-BC: continuously updated bias-corrected; CUP-FM: continuously updated and fully modified; FGLS: feasible generalized least squares; FMOLS: fully modified ordinary least squares; GDP: gross domestic product; GMM: generalized method of moments; ICT: information and communication technology; MMQR: method of moment quantile regression; NARDL: nonlinear autoregressive distributed lag; OECD: Organisation for Economic Co-operation and Development; PCSE: panel-corrected standard error; QARDL: quantile autoregressive distributed lag; RESET: regression specification error test.

Carbon neutrality refers to achieving a balance between the amount of carbon released into and absorbed by the ecosystem, also known as net-zero carbon emissions.⁴⁵ Under the framework of COP21, 124 countries have agreed to achieve carbon neutrality objective by 2050 or 2060. The main purpose of the carbon-neutrality goal is to keep the mean global temperature at 1.5°C to 2 °C above preindustrial levels. Among the advantages of carbon neutrality, the most notable are reducing global warming and improving environmental quality and human health.⁴⁶ Attainment of carbon neutrality is categorized as a revolution in the manufacturing sector and is believed to be a crucial juncture in human history. Whether carbon neutrality can be achieved within the traditional economic framework or we need to adopt a green economic model.⁴⁷ Moreover, it is important to know whether the top energy-consuming economies are working on achieving carbon neutrality objectives. These are the research questions this analysis wants to address.

Although the contemporary literature on carbon neutrality is growing,^48–50 the role of digital exports, financial stability, and energy security risk has not been explored as potential determinants of carbon neutrality within the GG framework in top energy-consuming economies. Further, past studies have suffered from methodological issues like cross-sectional dependence, resulting in inconclusive results. Prior literature only focuses on long-run effects and ignores short-run effects. Therefore, we need a comprehensive analysis to estimate the influence of digital exports, financial stability, and energy security on GG in top energy-consuming economies. Therefore, this analysis can be supported in light of the following policy framework threads: (i) shedding light on the digital exports, financial stability, energy security risk, and GG link and (ii) filling the above-stated gaps in the literature. The primary objective of the analysis is to investigate the linkage between digital exports, financial stability, energy security, and GG in top energy-consuming economies.

Against this backdrop, this study extends the literature through the following contributions. First, to our limited knowledge, this analysis is the first-ever effort to scrutinize how digital exports help achieve GG within the carbon neutrality objective. This clarifies how exports of digital items can contribute to the carbon neutrality objectives by fostering green economic growth. Second, analyzing the effect of financial stability on GG is another novel aspect of the study. This enhances our understanding regarding the development of a green economy with the help of a stable financial structure. Third, the analysis also sheds light on the connection between energy security and GG, suggesting how crucial the continuous supply of green and reliable energy sources is for GG. Fourth, the study picked the top energy-consuming economies for estimating the connection between outcome variable and regressors. Top energy-consuming economies have a crucial place in global decision making; thus, finding the factors that contribute to GG in top energy-consuming economies can prove vital for global carbon neutrality objectives. Moreover, we also perform regional analysis for Asia, Europe, and America, which provides more useful information regarding the GG dynamics in the diverse environments of these regions. Fifth, the analysis focuses on short- and long-term estimates instead of past studies, which only focus on the long-run estimates. Sixth, the analysis employs the novel cross-sectionally augmented autoregressive distributed lag (CS-ARDL) technique of Chudik and Pesaran⁵¹ that can offer robust estimates by controlling cross-sectional dependence. Seventh, this study also provides a country-specific analysis. Lastly, the study offers valuable suggestions to policymakers in top energy-consuming economies on how to achieve economic growth while mitigating environmental destruction. It gives the scientific community in-depth insights into how digital exports, financial stability, and energy security can contribute to GG.

Theoretical framework

The debate on the nexus between trade and the environment is not new and has a significant historical context. The supporters of the trade benefit hypothesis, using the idea of the environmental Kuznets curve (EKC), posit that International trade mitigates the climate impacts after a certain level of economic development.^52,53 Grossman and Krueger⁵⁴ have highlighted three effects, the “scale effect, structure effect, and technology effect” which are mainly responsible for the environmental degradation due to international trade. Two of these effects are structure and technology, which help mitigate carbon footprints, whereas the scale effect enhances the carbon footprints. Thus, this leads to the development of the trade uncertainty hypothesis.

As far as the nexus between digital trade and carbon neutrality is concerned, compared to traditional trade, it is believed to have a less detrimental influence on the environment. The idea of a digital economy has helped cross-border trade and increased productivity.⁵⁵ The increased use of information and communication technology (ICT) has greatly supported industry digitalization. Therefore, the ICT sector supports the nation's digital trade. As a result, ICT serves as a medium for digital trade.²¹ Digital trade is taking on more significance in the digital economy. It offers the main pathways for knowledge transmission, distribution, and sharing for environmentally sustainable technologies.⁵⁶ There has been a tremendous increase in global digital trade, which is currently seen as the new engine for achieving rapid economic development.⁵⁷ The relationship between digital trade and GG is debatable. In certain instances, digital trade has helped boost trading in the services industry, decreased costs, and improved resource allocation and energy efficiency. For example, digital has helped transform intelligent manufacturing processes, promoted energy efficiency, decreased transaction costs, and paved the way for the global value chain to contribute to green economic development.

The “theory of financial development” and its connection to the green economy and how the latter impacts environmental deterioration are closely intertwined. Over the past few decades, there has been an increasing focus on financial development by empirics, academics, and policymakers.⁵⁸ Several empirics have supported the favorable linkage between financial development and economic growth.⁵⁹ Due to the nexus between financial development and economic growth, these factors are believed to be intermediating factors in shaping the nation's environmental quality within the framework of EKC.⁶⁰ Two possible effects are financial stability and GG nexus. Positively, financial stability increases GG by facilitating investments in renewable energy projects. Financial stability enables green finance for low interest rates and promotes sustainable economic growth. Adversely, a higher level of financial stability is linked to more loans being provided, which helps people buy new automobiles, trucks, or equipment that emits additional carbon emissions and negatively impacts GG. Additionally, increased manufacturing activities result in increased carbon emissions, which hurt green economic growth.

Energy security, which includes reliable and accessible energy sources, is a fundamental pillar supporting the goals of green development. With less dependence on fossil fuels, governments may increase energy security while lowering carbon emissions and environmental damage by securing a diverse and sustainable energy mix. The concepts of GG are supported by this shift to cleaner energy sources, fueled by policies that support renewable energy, energy efficiency, and technical advancement.⁶¹ Energy security measures increase economic resilience by lowering sensitivity to supply interruptions and price volatility. This fosters an environment conducive to sustained economic development while minimizing environmental impacts. Alsagr and Ozturk²⁰ noted that energy security risk has a favourbale impact on green investment, promoting GG. They found that increased energy security risk enhances the energy transition, which positively affects GG.

Econometric model

The financial sector's stability enables quick transactions and maintains checks on the economy's resources, which are vital to its development. For economic development to flourish, it is vital to have a solid financial system to move money efficiently and stimulate firms to raise their output. The previous works of Carbó-Valverde and Sánchez⁶² and Alsamara et al.⁶³ confirm the positive role of the stable financial sector in fostering economic growth. On the other hand, Hasni et al.⁶⁴ suggested that financial stability helps achieve economies of scale that may reduce the environmental consequences of economic activities. According to Ma and Stern,⁶¹ the link between economic expansion and a “technical impact” may result in reduced carbon footprints due to the less wastage of resources during the industrial process.

On one side, the literature suggests that ICT may be exploited to increase growth, lower inequality, and support equitable education, it is also believed to be a crucial accelerator for carbon economic development.⁶⁵ On the other hand, the literature also shows that expanding ICT trade may also raise energy usage and, consequently, pollution. During generation and work, ICT equipment may consume a lot of energy and create harmful waste.⁶⁶ Energy resources are crucial in every sector of the economy; however, energy usage might risk ecological sustainability due to increased energy demand by several sectors of the economy.⁶⁷ Thus, the energy security literature is widely recognized for exploring the significance of energy security in economic and environmental wellbeing. The above empirical literature confirms a connection between digital exports, financial stability, energy security risk, and GG. To capture the empirical connection between our variables, following the literature, particularly the studies of Chen et al.⁸ and Zheng et al.,³¹ we have formulated the following panel model:

\begin{matrix} GG_{i t} = δ_{0} + δ_{1} {DE}_{i t} + δ_{2} {FS}_{i t} + δ_{3} {ESR}_{i t} + δ_{4} {ET}_{i t} + δ_{5} {REC}_{i t} + ε_{i t} \end{matrix}

(1)

where GG is a function of digital exports (DE), financial stability (FS), energy security risk (ESR), environmental technology (ET), renewable energy consumption (REC), and error term (

ε_{i t}

). Digital exports can play a significant role in boosting GG by promoting energy efficiency and green innovation in various sectors of the economy. Thus, we would expect DE to be positive. Financial stability can significantly impact GG, and its positive and negative effects, depending on various factors. Financial stability enhances investor confidence and access to capital, fostering market growth, which, in turn, positively impacts GG. Conversely, a lack of financial stability can reduce incentives for green innovation, thereby impeding the progress of GG. Thus, we expect a positive or negative impact on GG. Lastly, higher energy security risks can hamper GG by creating uncertainties and challenges that impede the transition to a more sustainable and environmentally friendly energy system. Therefore, DE is expected to have a negative impact.

Data and descriptive analysis

The study's primary objective is to assess the impact of digital exports, financial stability, and energy security on GG in the top energy-consuming economies from 2000 to 2021. We collected data from 33 of the world's leading energy-consuming economies. This sample is selected based on the availability of data. Our dependent variable is GG. Following Gao et al.⁶⁸ our study measures GG in terms of environmentally adjusted multifactor productivity growth. The data series for GG have been sourced from the Organisation for Economic Co-operation and Development (OECD). We measured the DE variable by comparing the percentage of ICT goods exported to total goods exported. World development indicator (WDI) is the data source for the digital exports variable. The FS variable is assessed using the bank Z-score formulated by the global financial development database. The third independent variable, ESR is determined through an index developed by the Global Energy Institute. In addition to these three independent variables, our study also encompasses ET and REC as control variables. The control variables are selected based on Porter hypothesis⁶⁹ and the hypothesis of renewable energy-led GG.⁷⁰ ET and REC are also essential components of GG. They promote sustainable economic development, reduce environmental harm, enhance energy security, and contribute to global efforts to combat climate change. The means and standard deviations (SDs) are reported as follows: GG (mean: 2.280; SD: 3.224), DE (mean: 1.313; SD: 1.292), FS (mean: 2.515; SD: 0.603), ESR (mean: 6.899; SD: 0.189), ET (mean: 6.024; SD: 2.020), and REC (mean: 1.804; SD: 3.609). Variables description, descriptive statistics, and the sources are documented in Tables 2 and 3. Analysis conducted using Stata 17.

Table 2.

Variables and descriptive statistics.

Variable	Symbol	Definition	Sources
Green growth	GG	Environmentally adjusted multifactor productivity growth (%)	OECD
Digital exports	DE	ICT goods exports (% of total goods exports)	WDI
Financial stability	FS	Bank Z-score	GFDD
Energy security risk	ESR	Energy security risk index	Global Energy Institute
Environmental technology	ET	Environmental related technologies (Total Patent)	OECD
Renewable energy consumption	REC	Total energy consumption from nuclear, renewables, and other (quad Btu)	EIA

Source: Author's calculations.

EIA: energy information administration; GFDD: global financial development database; OECD: Organisation for Economic Co-operation and Development; WDI: world development indicators.

Table 3.

Descriptive statistics.

	Mean	Median	Maximum	Minimum	SD	Skewness	Kurtosis
GG	2.280	2.153	13.44	−10.71	3.224	0.219	3.842
DE	1.313	1.385	3.606	−3.548	1.292	−0.788	3.501
FS	2.515	2.534	3.675	−4.055	0.603	−2.092	22.17
ESR	6.899	6.934	7.519	6.252	0.189	−0.694	3.389
ET	6.024	5.759	10.68	0.500	2.020	0.093	3.293
REC	1.804	0.495	24.71	−0.168	3.609	3.682	17.38

Source: Author's calculations.

DE: digital exports; ESR: energy security risk; ET: environmental technology; FS: financial stability; GG: green growth; REC: renewable energy consumption.

Table 4 offers the outcomes of the correlation matrix. This matrix informs whether the variables are correlated to each other. The degree of correlation between variables varies between −1 and +1, where −1 represents a perfect negative correlation, while +1 signifies the perfect correlation. However, the 0 indicates no correlation between the variables. From the outcome of the correlation matrix, we infer that the highest positive value, 0.558, lies between ET and REC, followed by 0.407 between ET and FS. None of the values is either +1 or −1, implying that all the correlation values are within the legal range. Hence, we can add all the values into the correlation matrix and proceed with our regression.

Table 4.

Correlation matrix.

	GG	DE	FS	ESR	ET	REC
GG	1
DE	0.143	1
FS	0.014	0.296	1
ESR	−0.022	−0.004	−0.007	1
ET	0.060	0.208	0.407	−0.177	1
REC	0.130	0.211	0.381	−0.225	0.558	1

Source: Author's calculations.

DE: digital exports; ESR: energy security risk; ET: environmental technology; FS: financial stability; GG: green growth; REC: renewable energy consumption.

Table 5 proposes the outcomes of the variance inflation factor (VIF), which is a multicollinearity test. The rule of thumb is to check if the VIF is lower than 5, as it suggests a low level of multicollinearity. The outcomes confirm that VIF for all variables (ET, REC, FS, DE, and ESR) is less than 5, as well as the mean VIF. These results imply that multicollinearity is not a concern in our selected variables.

Table 5.

VIF results.

Variable	VIF	1/VIF
ET	1.57	0.636
REC	1.57	0.637
FS	1.33	0.753
DE	1.11	0.897
ESR	1.07	0.934
Mean VIF	1.33

Source: Author's calculations.

DE: digital exports; ESR: energy security risk; ET: environmental technology; FS: financial stability; REC: renewable energy consumption; VIF: variance inflation factor.

Econometrics methods

CSD and homogeneity tests

Cross-sectional dependence (CSD) is the first among several preliminary tests that need to be applied before we examine the short- and long-run estimates. This CSD may appear due to the connection between the economies due to trade integration, cultural relations, common borders, economic and financial dependence, and mutual social traits.¹⁸ Due to the above-stated connections between the economies, the shock in one country could impact all other countries in the region or even in other regions. This may lead to erroneous results. Thus, we believe it is appropriate to check the CSD using Pesaran⁷¹ test to obtain accurate estimates. This test is suitable where interactions between regions are significant. Following is the equation that represents the CSD equation:

C D_{test} = \sqrt{\frac{2 T}{N (N - 1)}} (\sum_{i = 1}^{N - 1} \sum_{k = i + 1}^{N} {\hat{τ}}_{i k})

(2)

The study also used a slope heterogeneity test to confirm whether the relationships between variables are homogeneous across different cross-sectional units within the panel data. Therefore, slope heterogeneity is another important test essential for obtaining accurate estimates. To that end, the slope heterogeneity test of Pesaran and Yamagata⁷² is applied, and the equations representing this test are given below:

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

(3)

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

(4)

Panel unit root tests

The next step is unit root testing to assess whether a dataset is stationary or nonstationary. The CSD and slope heterogeneity tests also help us to identify whether we need to apply the first-generation or second-generation unit root tests. If there are signs of CSD in our data, the first-generation unit root tests become handicapped and fail to produce accurate outcomes.³⁸ The problems due to CSD in the outcomes can be overcome by implementing the second-generation unit root test “CIPS” (cross-sectionally augmented Im–Pesaran–Shin) extended by Pesaran.⁷³ Recently, these unit root tests have become famous for countering the issue of CSD. The basic specification of the test is shown below:

Δ V_{i, t} = α_{i} + α_{i} X_{i, t - 1} + α_{i} Y_{t - 1} + \sum_{l - 0}^{p} α_{i 1} Δ \bar{V_{t - 1}} \sum_{i - 1}^{p} α_{i} Δ Y_{i, t - 1} + μ_{i t}

(5)

Equation (5) displays the cross-section averages as $\bar{V_{t - 1}}$ . The value of cross-sectionally augmented Dickey–Fuller (CADF) may attain the value of CIPS. Specification (6) represents the CIPS:

\hat{CIPS} = \frac{1}{N} \sum_{i = 1}^{n} C_{ADF}^{i}

(6)

Panel cointegration tests

Cointegration tests are crucial for examining the long-run nexus between DF, FS, ESR, ET, REC, and GG. If there is no cointegration between the variables, the long-run relationship between these variables is believed to be spurious. For policy formulation, a valid long-run relationship should exist; thus, proving cointegration is necessary. There are several cointegration tests in econometric literature. Among those tests, Westerlund⁷⁴ is a second-generation test, while Pedroni⁷⁵ and Kao⁷⁶ belong to the first-generation. Westerlund⁷⁴ is our choice for this analysis because it can help us recognize the true nature of the long-run nexus between the variables in the case of CSD in the data.

CS-ARDL

Once the CSD, slope heterogeneity, and unit root properties are confirmed, we need to identify an efficient and robust estimator, such as the CS-ARDL estimator. In the presence of the above-stated issues in the panel data, the CS-ARDL can help us concurrently analyze the short- and long-run nexus between DF, FS, ESR, and GG. Chudik and Pesaran⁵¹ developed this technique, which surpasses numerous other procedures. The significance of the CS-ARDL is growing in the environmental literature⁷⁶ due to its superiority over other methods. All other panel cointegration approaches are applicable only if the variables have a unit root or integrated order one. At the same time, the CS-ARDL successfully captures the relationship between the variables without worrying about the stationary features of the variables. Moreover, its superiority is also established due to its ability to offer the short- and long-run impact of regressors on the outcome variable. Further, controlling the CSD, slope heterogeneity, endogeneity, and serial correlation during panel analysis are some other unique characteristics of this approach that give an edge to CS-ARDL over other methods. To empirically test the variables in the CS-ARDL framework, we develop the following equation (7):

\begin{matrix} Δ G G_{i t} = C_{i} + Ψ_{i} (G G_{i t - 1} - Υ_{i} Z_{i t - 1} - ϕ_{1 i} {\bar{GG}}_{t - 1} - ϕ_{2} {\bar{Z}}_{t - 1}) + \sum_{j = 1}^{p - 1} ω_{i j} Δ G G_{i t - j} \\ + \sum_{j = 0}^{q - 1} λ_{i j} Δ Z_{i t - j} + π_{1 i} Δ {\bar{GG}}_{t} + π_{2 i} Δ {\bar{Z}}_{t} + ε_{i t} \end{matrix}

(7)

where $Δ G G_{i t}$ is the dependent variable, $Z_{i t - 1}$ the independent variables, ${\bar{GG}}_{t - 1}$ the mean of the dependent variable, and ${\bar{Z}}_{t - 1}$ the mean of independent variables in the long run. Regarding short-run, $Δ G G_{i t - j}$ is the dependent variable, $Δ Z_{i t - j}$ the independent variables, $Δ {\bar{GG}}_{t}$ the mean of a dependent variable, $Δ {\bar{Z}}_{t}$ the mean of the independent variables, and $ε_{i t}$ the error term. Meanwhile, j represents the cross-sectional dimension, and time is denoted by t. To enhance robustness, we used the pooled mean group-autoregressive distributed lag (PMG-ARDL) method. This approach mitigates autocorrelation problem and simultaneously provides short- and long-term effects. Additionally, the study also employs the panel quantile regression method. This method accounts for the entire distribution while addressing time-varying heterogeneity and outliers.⁷⁷ The country-specific analysis enhances the credibility of our outcomes. For this purpose, fully modified ordinary least squares method is applied. This was designed to offer appropriate estimates in a cointegration analysis. This approach attains asymptotic efficiency by transforming the OLS technique to deal with the issues of serial correlation and endogeneity, which may erupt during cointegration analysis.⁷⁸ This approach provides long-run country-specific results for analysis. Lastly, we have performed some diagnostic tests to assess the validity of the PMG-ARDL estimates. These tests include the Langrage multiplier (LM) test of first-order serial correlation, the Breusch–Pagan (BP) test for heteroskedasticity, and Remsay's RESET test for model misspecification. In addition, the long-run relationship between the variables is tested via the error correction model (ECM) test.

Empirical results and discussion

The outcomes of the CSD test are shown in Table 6. When using panel data, it is vital to evaluate CSD. With globalization, countries are economically and financially interconnected. Therefore, unexpected shocks in a single nation can affect the entire world. Ignoring CSD in this context can lead to biased estimates. The Pesaran⁷⁹ CSD test was therefore used to assess the presence of CSD. Our results validate the existence of CSD since all the test statistics related to all variables (GG, DE, FS, ESR, ET, and REC) in Table 6 are significant. This result implies that the values of each variable are dependent across the cross-sections, which are countries in this case.

Table 6.

Cross-section dependence tests.

	GG	DE	FS	ESR	ET	REC
Pesaran's test	52.41***	9.621***	10.61***	25.92***	21.44***	13.74***
Prob.	0.000	0.000	0.000	0.000	0.000	0.000
Off-diagonal elements	0.516	0.438	0.329	0.460	0.410	0.397

Note: ***p < 0.01, **p < 0.05, *p < 0.1.

DE: digital exports; ESR: energy security risk; ET: environmental technology; FS: financial stability; GG: green growth; REC: renewable energy consumption.

The findings of a slope homogeneity test are displayed in Table 7. We use both unadjusted and delta-adjusted statistics. Since both statistics exhibit significant p-values, the slope homogeneity null hypothesis is definitively rejected; thus, we may conclude that our model exhibits slope heterogeneity. In other words, it rejects the “slope homogeneity across cross-sectional units.” Thus, we may conclude that our model exhibits slope heterogeneity.

Table 7.

Slope heterogeneity test.

	Delta	p-value
$\hat{Δ}$	7.100	0.000
${\hat{Δ}}_{adjusted}$	8.601	0.000

Note: ***p < 0.01, **p < 0.05, *p < 0.1.

The stationary characteristics of the variables are determined once CSD is established. Table 8 displays the findings of CIPS. The CIPS test results indicate that the variables GG, ESR, ET, and REC are stationary at I(1). In contrast, DE and FS are integrated at I(0) and stationary at level. These results show that the study's variables are stationary at different integration orders. Therefore, we must use an econometric method to handle the heterogeneous order of integrations. Westerlund⁷⁴ cointegration test is employed to determine if model variables have an accurate long-term relationship. In other words, it is used to decide whether or not they are cointegrated. Table 9 displays the cointegration results. Since one group and one-panel statistic out of all four of our statistics are significant, there is a confirmation of cointegration between our variables. This suggests that these variables have a valid long-run relationship.

Table 8.

Unit root test results.

	CIPS
	I(0)	I(1)	Decision
GG	0.123	−3.104***	I(1)
DE	−2.088***		I(0)
FS	−2.219***		I(0)
ESR	−1.148	−3.626***	I(1)
ET	0.011	−2.674***	I(1)
REC	−1.102	−5.170***	I(1)

Note: CIPS table critical values: 1%: −1.820, 5%: −1.730, 10%: −1.690. ***p < 0.01, **p < 0.05, *p < 0.1.

CIPS: cross-sectionally augmented Im–Pesaran–Shin; DE: digital exports; ESR: energy security risk; ET: environmental technology; FS: financial stability; GG: green growth; REC: renewable energy consumption.

Table 9.

Results of Westerlund cointegration test.

Statistic	Value	z-value	p-value
Gt	−1.977*	1.500	0.067
Ga	−4.772	2.795	0.997
Pt	−9.259*	1.293	0.098
Pa	−4.798	0.480	0.316

Note: ***p < 0.01, **p < 0.05, *p < 0.1.

Table 10 shows the CS-ARDL results. Long-run estimated coefficients are more significant and are discussed in detail. In every model, the association between DE and GG is favorable. In particular, GG rises in the full sample, Asian, European, and American models by 1.578%, 1.521%, 2.161%, and 2.478%, respectively, with a 1% increase in DE. This result is supported by Xu,⁸⁰ who argued that digital exports increase information related to environmentally friendly technologies and practices. The study of Xiong and Luo²³ also supported and noted that digital exports reduce resource consumption, which can have a positive environmental impact. Murshed²⁴ documented that digital trade provides technological support for GG. These studies argued that digital trade improves GG and energy efficiency by stimulating technological innovation and diffusion, decreasing production costs and enlarging demand. Similarly, Zhang²⁵ stated that digital trade improves the industrial process and transportation supply chain, which enhances overall environmental performance, ultimately leading to GG. The positive nexus between digital trade and GG also corroborates with the study done by Sun et al.¹⁸ The study argued that digital trade affects consumers’ technology adoption behavior and motivates them to utilize imported ICT-based technologies. Wang et al.⁸¹ stated that with continuous innovation in ICT sector, digital trade has rapidly increased throughout the globe and has become new determinant of green economic growth. Staiger⁸² noted that digital trade occupies a significant share of worldwide trade. Digital trade can enhance efficiency, optimize resource allocation, reduce costs, and improve service trade, contributing to significant green economic growth.

Table 10.

Group and regional-specific results (CS-ARDL).

	Full sample		Asia		America		Europe
	Coefficient	z-stat	Coefficient	z-stat	Coefficient	z-stat	Coefficient	z-stat
Long-run
DE	1.578**	2.511	1.521*	1.835	2.478***	2.926	2.161**	2.357
FS	1.346**	1.991	1.058**	2.452	0.082***	3.059	0.411*	1.859
ESR	−1.542**	−2.485	−2.054***	−3.461	−1.124**	−2.242	−2.154***	−3.115
ET	2.336*	1.781	0.013	0.144	0.425*	1.921	0.308**	2.112
REC	0.174*	1.944	1.021	0.841	1.885	1.324	1.552*	1.689
Short-run
DE	1.578	1.111	1.041	0.992	0.758	0.781	0.165	1.111
FS	0.855**	2.312	0.797*	1.731	0.690**	2.114	1.185	0.295
ESR	−0.515	−0.713	−1.106	−0.285	−0.255	−0.971	−0.252	−1.550
ET	1.654*	1.811	1.865*	1.752	2.678*	1.872	1.398*	2.526
REC	0.105	1.492	0.056	0.918	0.512	0.631	0.056	1.219

Note: ***p < 0.01, **p < 0.05, *p < 0.1.

CS-ARDL: cross-sectionally augmented autoregressive distributed lag; DE: digital exports; ESR: energy security risk; ET: environmental technology; FS: financial stability; GG: green growth; REC: renewable energy consumption.

Likewise, FS and GG are positively correlated in every model. For instance, for every 1% increase in FS, GG will increase in the full sample, Asian, European, and American models by 1.346%, 1.058%, 0.411%, and 0.082%, respectively. This finding aligns with Li,¹³ who noted that financial stability increases investment in green infrastructure. This green investment contributes to GG by reducing environmental impact. Another possible justification is that financial stability provides access to capital for businesses and projects with environmentally sustainable objectives. Safi et al.³² highlighted that financially stable economies are more likely to improve economic progress and environmental quality. Baloch et al.³³ revealed that financial stability is a key determinant of green economic growth as financially stable institutions provide loans to enterprises at low interest rates, support mobilization of savings and risk management, and improve the efficiency of capital markets. Moreover, Baloch et al.³³ described that financially stable economies support enterprises with financial resources to converge toward clean and renewable energy utilization. The study claimed that financial stability is a gateway toward eco-friendly and renewable energy technology that curbs CO₂ emissions and enhances GG.

In all models, the long-run ESR has a negative impact on GG. This shows that in the full sample, American, European, Asian models, GG falls by 1.542%, 1.124%, 2.154%, and 2.054%, with a 1% increase in ESR. In support of this finding, Akinyemi et al.⁸³ argued that energy security risks discourage investments in renewable energy sources. This infers that high energy security risks push countries to rely more heavily on fossil fuels energy sources. This leads to an increase in carbon emissions and hinders progress toward GG goals. Another possible reason is that high energy security risks within the economy ignore sustainable and environmentally friendly practices. The energy security risk leads to social and economic disparities and hinders efforts to promote GG equitably. Blum and Legey⁸⁴ described that energy security is essential for any economy's sustainable development and GG process. According to Kartal,⁴¹ energy security is the lifeblood of global economic growth. The study stated that a reliable and adequate energy supply is essential for sustainable growth as energy supply fuels all economic activities, such as health, education, food production, industrial processes, communication, and transportation. Therefore, the risk to energy security reduces GG.

While the ET estimates are significantly favorable in the full sample, European, and American models. In particular, a 1% increase in ET results in increases in GG of 2.336%, 0.308%, and 0.425% in the full sample, European, and American models, respectively. This finding is corroborated by Chen et al.,¹⁰ who noted that progress in ETs stimulates clean energy consumption and thus boosts GG. Meanwhile, ETs increase green production efficiency and promote sustainable development. The study by Hwang and Díez⁸⁵ stated that fossil fuel energy combustion is replaced by clean energy sources, thus leading to GG. The results reveal that development in ETs is crucial for energy transition that leads to GG. Ahmad et al.⁸⁶ argued that ETs increase sustainability in the agriculture and industrial sectors by reducing environmental pollution. Therefore, ET is an effective method for stimulating GG. REC estimates are only positively correlated with GG in the full sample and Europe. A 1% increase in REC causes increases in GG of 0.174% and 1.552% in the full sample and European. This result is congruent with the findings of Mohsin et al.,⁸⁷ who stated that renewable energy is a determinant of GG. This result finding is backed by Chang and Fang,⁷⁰ who noted that REC reduces carbon emissions by ensuring GG. The renewable energy-led growth hypothesis also supports our result.

The short-run effects are mostly unimportant except for the FS and ET variables. In the short run, GG increases by 0.855%, 0.797%, and 0.108% in the full sample, Asian, and American models, respectively, for every 1% rise in FS. Additionally, GG is enhanced by 1.654%, 1.865%, 2.678%, and 1.398% in the full sample, Asian, and American models, respectively, for every 1% rise in ET.

Table 11 reports the long- and short-run estimates from the robust PMG-ARDL model. The long-run estimates of PMG-ARDL model show that the association between DE and GG is significantly positive only in the full sample, Asia, and Europe. The coefficient estimates infer that with a 1% increase in DE, GG increases by 0.794% in the full sample, 1.375% in Asia, and 1.602% in Europe in the long-run. Similarly, FS is significantly and positively associated with GG in all models. For every 1% upsurge in FS, GG enhances by 0.841% in the full sample, 1.850% in Asia, 0.557% in America, and 0.853% in Europe. ESR reports a negative impact on GG in all four long-run models, but the effect is statistically significant only in the full sample model. The findings show that with a 1% increase in ESR, GG reduces by 2.106% in the full sample model. ET reports a significantly positive impact on GG in three models, the full sample, Asian sample, and European sample. The results show that with a 1% increase in ET, GG enhances by 0.189% in the full sample, 0.741% in Asia, and 0.273% in Europe. Although REC reports a positive impact on GG in all four models in the long run, the effect is statistically insignificant. The short-run effects are mostly insignificant except ET variable. ET has a statistically significant and positive effect on GG only in the case of Asia. The results show that with a 1% increase in ET, GG enhances by 0.596% in Asia in the short-run. In the end, ECM statistics is given for all samples. The ECM term confirms the cointegration association among variables and describes the convergence speed towards long-run equilibrium.

Table 11.

Group and regional-specific results (PMG-ARDL)-robustness.

	Full sample		Asia		America		Europe
Variable	Coefficient	t-stat	Coefficient	t-stat	Coefficient	t-stat	Coefficient	t-stat
Long-run
DE	0.794***	3.074	1.375*	1.670	0.046	0.140	1.602***	4.297
FS	0.841**	2.379	1.850**	2.212	0.557**	2.445	0.853**	2.041
ESR	−2.106**	−2.087	−2.212	−0.961	−1.178	−0.548	−0.358	−0.258
ET	0.189*	1.702	0.741**	2.510	0.186	0.980	0.273*	1.689
REC	0.014	0.238	0.082	1.157	0.058	0.232	0.267	0.217
Short-run
D(DE)	1.852	1.609	1.396	0.925	2.879	1.634	1.574	1.446
D(FS)	0.986	1.007	0.634	0.786	1.768	1.227	1.101	1.242
D(ESR)	−1.020	−1.439	−2.989	−0.656	−1.993	−0.756	−2.045	−1.114
D(ET)	0.501	0.901	0.596*	1.890	−0.931	−0.976	1.114	1.290
D(REC)	−0.508	−0.115	−1.491	−0.418	0.863	0.261	−0.900	−0.394
C	8.521***	11.88	19.72***	4.917	6.588***	4.992	2.054***	4.788
Diagnostic tests
ECM (−1)	−0.701***	−12.49	−0.820***	−4.667	−0.851***	−8.701	−0.649***	−9.092
LM test (p-value)	0.658		0.521		0.498		0.621
BP test (p-value)	0.452		0.324		0.374		0.398
RESET test (p-value)	0.878		0.804		0.947		0.754

Note: ***p < 0.01, **p < 0.05, *p < 0.1.

BP: Breusch–Pagan; DE: digital exports; ECM: error correction model; ESR: energy security risk; ET: Environmental technology; FS: financial stability; GG: green growth; PMG-ARDL: pooled mean group-autoregressive distributed lag; REC: renewable energy consumption.

The ECM statistics is negative in all four models, which states that convergence will occur at the speed of 70% in the full sample, 82% in Asia, 85% in America, and 65% in Europe in 1 year. Some other diagnostics are also presented in Table 11. For instance, LM test confirms that there is no issue of autocorrelation in our models, the BP test confirms that models don’t suffer from heteroskedasticity, and the RESET test indicates no signs of model misspecification.

Table 12 provides the outcomes of the panel quantile regression. From the table, we observe that the estimates of DE are positively linked to GG from 0.10 to 0.90 quantiles, suggesting DE fosters GG at all levels. Likewise, the estimates attached to FS are positive and significant at all quantiles, that is, 10th to 90th. This result reveals that FS boosts GG at most of its levels. Moreover, the ET is positively and significantly linked to GG at all its levels, that is, from 10th to 90th quantiles, while the REC fosters GG only at the highest levels, that is, 75th and 90th quantiles. In contrast, the ESR helps reduce the GG from the 25th to the 90th quantiles, implying that ESR hurts all levels of GG.

Table 12.

Panel quantile regressions (full sample).

Variables	(0.10)	(0.25)	(0.50)	(0.75)	(0.90)
DE	0.512***	0.475***	0.384***	0.227*	0.637***
	(0.107)	(0.106)	(0.129)	(0.130)	(0.175)
FS	0.560**	0.281	0.771***	0.394*	0.507**
	(0.250)	(0.218)	(0.182)	(0.215)	(0.249)
ESR	−1.164	−1.479**	−1.678**	−1.981**	−1.425*
	(0.717)	(0.741)	(0.808)	(0.845)	(0.774)
ET	0.381***	0.195**	0.238**	0.305***	0.390***
	(0.081)	(0.080)	(0.098)	(0.099)	(0.133)
REC	−0.020	−0.006	0.083	0.287***	0.381***
	(0.045)	(0.045)	(0.055)	(0.055)	(0.074)
Constant	3.776	9.264*	4.091	5.330	6.623
	(4.981)	(4.939)	(6.030)	(6.077)	(8.153)

Note: Standard errors in parentheses. ***p < 0.01, **p < 0.05, *p < 0.1.

DE: digital exports; ESR: energy security risk; ET: environmental technology; FS: financial stability; REC: renewable energy consumption.

Table 13 underscores the country-specific effects. The estimated coefficients of DE a positive significant impact on GG in 23 countries (China, the USA, Japan, Germany, Canada, Brazil, Turkiye, the UK, Poland, France, Spain, the Netherlands, Belgium, Chile, Greece, Austria, Israel, Portugal, Finland, Ireland, Switzerland, Sweden, Denmark), but it has negative impact in five countries (Mexico, Argentina, Romania, Bulgaria, and Norway). Furthermore, FS exhibits a positive and significant effect on GG in 26 countries (China, the USA, India, Japan, Germany, Canada, Mexico, Turkiye, the UK, Poland, Spain, the Netherlands, Czechia, Belgium, Chile, Colombia, Romania, Greece, Israel, Hungary, Bulgaria, Ireland, Norway, Switzerland, Sweden, and Denmark) and negative effect in three countries (France, Austria, and Croatia). The results also infer that ESR negatively affects GG in 22 countries (the USA, India, Canada, Mexico, Brazil, Turkiye, the UK, Poland, France, Spain, the Netherlands, Chile, Romania, Greece, Austria, Israel, Finland, Bulgaria, Ireland, Norway, Sweden, and Denmark). The estimates of control variables ET and REC are positively significant in almost half of the countries, and their interpretation is the same as in the case of primary variables.

Table 13.

Country-specific results (FMOLS).

	DE		FS		ESR		ET		REC
Country	Coefficient	z-stat	Coefficient	z-stat	Coefficient	z-stat	Coefficient	z-stat	Coefficient	z-stat
China	2.38*	1.84	2.08***	3.02	−2.01	−0.57	1.26**	2.37	0.13**	2.66
The USA	2.31***	4.12	2.34***	5.59	−4.13***	−4.31	1.71***	3.85	0.01	0.18
India	0.44	1.24	3.74***	3.41	−4.45***	−6.37	0.55**	2.42	0.65**	1.97
Japan	3.05***	3.52	4.17***	6.72	−3.71	−0.79	0.15	0.56	0.14	1.03
Germany	4.61***	4.17	1.45***	3.44	−1.91	−1.03	2.13***	4.64	1.68**	2.43
Canada	2.96***	8.04	2.68***	3.54	−2.87***	−6.16	1.97***	9.43	0.71***	3.16
Mexico	−1.85**	−2.53	3.16***	8.48	−2.64***	−2.75	−0.05	−0.97	3.18**	2.08
Brazil	1.65***	4.91	0.31	0.22	−3.17***	−2.69	−0.24	−1.07	−0.92*	−1.95
Turkiye	3.41***	5.63	2.44***	4.96	−3.54***	−5.41	0.37***	3.85	3.50***	3.42
The UK	3.35***	7.35	3.97**	2.23	−7.45***	−3.52	0.16	0.59	2.61***	3.61
Poland	0.55**	2.17	0.85***	6.15	−9.87***	−6.25	−1.24	−0.12	3.12***	2.68
France	2.82***	5.29	−1.41***	−5.01	−6.03***	−4.06	1.15	1.55	1.87***	4.37
Spain	2.39***	2.66	2.73***	8.95	−7.18***	−7.82	0.68***	8.57	0.73	1.05
Argentina	−5.25***	−6.39	−1.63	−1.09	−5.12	−1.47	−1.11	−1.14	3.01	0.22
The Netherlands	3.51***	7.34	1.06***	7.04	−3.24***	−2.99	−0.37	−0.39	2.45***	4.12
Czechia	0.74	0.74	3.12**	2.54	3.73	1.45	−0.76	−0.62	−1.99*	−1.68
Belgium	1.22***	3.26	0.62***	2.78	−2.61	−1.21	−0.18	−1.03	−2.46	−1.57
Chile	0.52**	2.53	3.24***	8.16	−1.22***	−5.16	2.11***	9.37	3.11**	2.44
Colombia	0.25	1.32	1.26***	2.79	2.08	1.41	0.25**	2.52	4.87***	5.99
Romania	−2.14***	−3.68	2.88***	7.97	−1.45***	−6.05	2.76***	4.48	4.87***	3.28
Greece	3.45***	6.25	2.48***	8.74	−3.24***	−2.81	0.54	1.61	−1.07	−0.32
Austria	1.96***	5.25	−1.04***	−5.08	−1.19***	−7.38	0.77***	7.41	1.95***	5.84
Israel	1.95***	3.58	3.15***	6.37	−1.96***	−5.17	2.12***	10.6	2.61***	6.74
Portugal	3.28***	4.17	−0.54	−1.54	−1.09	−0.65	0.37***	6.51	2.76***	6.97
Hungary	−1.26	−1.21	1.14	0.82	−1.71	−0.35	−0.05	−0.35	2.14***	5.45
Finland	3.61***	4.68	−0.12	−0.15	−2.52***	−4.03	2.92**	2.26	3.87	1.01
Bulgaria	−5.21***	−3.54	2.02***	5.45	−7.98***	−5.54	0.86***	8.89	2.76***	4.31
Ireland	2.44***	2.87	0.43***	2.99	−2.88***	−3.73	2.08***	3.78	3.64***	6.85
Norway	−1.81***	−4.52	2.22***	4.23	−4.71***	−4.74	1.89***	8.14	0.23	0.19
Switzerland	1.85***	6.46	0.94***	3.38	1.75	0.98	3.15***	7.83	−1.61	−0.46
Sweden	2.91***	5.67	1.77***	6.42	−5.13***	−7.92	0.24	0.97	3.67***	3.24
Denmark	2.51***	5.52	3.02***	3.23	−1.56***	−7.27	0.03	0.29	1.79***	3.16
Croatia	0.59	0.63	−4.82***	−9.01	−7.72	−1.43	0.42*	1.76	2.54***	6.29

Note: ***p < 0.01, **p < 0.05, *p < 0.1.

FMOLS: fully modified ordinary least squares.

Conclusion, policy implications, and limitations

Conclusion

High-energy-consuming economies are at the forefront of global environmental sustainability and GG concerns. These economies are characterized by substantial energy consumption patterns that lead to environmental degradation and pose challenges to achieving sustainable development goals. Addressing these challenges is critical not only for the wellbeing of these nations but also for global efforts to combat climate change and promote a greener future. While the emerging literature investigates novel aspects of GG, it has largely neglected the importance of digital exports, financial stability, and energy security risks. Against this backdrop, our study investigates the influence of digital exports, financial stability, and energy security risk on GG. To achieve this, we have collected data from 33 high-energy-consuming economies from 2000 to 2021. Employing the CS-ARDL estimation technique, the following outcomes are obtained. Some of the main highlights of our findings are as follows: (1) digital exports and financial stability boost GG in the full sample, Asian, European, and American models only in the long run; (2) energy security risks hinder long-run GG in the full sample, Asian, European, and American models; (3) ET promote long-run GG in the full sample, European, and American models, and REC helps boost GG in full and European models; (4) in the short-run, financial stability promote GG in the full sample, Asian, and American models, while ET benefits GG in all models; (5) the country-specific estimates of digital exports and financial stability are positively linked to GG, while the estimates of energy security risk are negatively significant in most of the economies.

Policy implications

Based on the results, our study proposed the following policy recommendations. First and foremost, digital exports enhance the GG in all regions. However, among all regions, the positive impact of digital exports on GG is the largest in the American region, followed by the European and Asian regions. Second, financial stability also favorably influences GG in all three regions, with the largest impact being observed in the Asian region, followed by the European and American regions. In contrast, energy security risks reduce GG in all three regions. This suggests that policymakers in all three regions should prioritize digital exports and financial stability in their GG policies, while also working to reduce energy security risks to foster GG. In this context, the following are the suggestions for the policymakers in respective regions. In Asia, policymakers must prioritize the development of digital infrastructure to support digital exports. In addition, they should keep an eye on cybersecurity issues to promote digital trade in the regions. Financial stability is one of the concerns of the Asian region that regulatory reforms and investment in financial technologies can overcome. Lastly, to enhance the level of energy security in the region, policymakers in the Asian region should enhance their investment in renewable energy and reliable energy sources.

The American region, with countries like America and Canada, is the most advanced economy in the world and has the highest level of digital penetration into its economies. In this region, policymakers should aim to incentivize digital innovation and exports, enhance financial systems to better withstand economic shocks, and ensure energy security through diversified renewable energy sources. Europe is also one of the most developed regions. In Europe, policymakers should focus on continuous innovation in digital technologies to maintain their leadership position in digital exports. European nations must navigate the delicate balance between their people's concerns over privacy and security and the corporate sector's need for unrestricted data flow to engage in digital trade and global value chains. This can be achieved by strengthening the laws regarding digital secrecy and building dedicated secure servers overseas to keep digital records. The rising financial stability risks due to geopolitical and economic tensions, global financial stability is at high risk, particularly in Europe and America. Thus, policymakers must ensure financial stability with stringent regulatory oversight. Due to rising energy demand in Europe, energy security policies should focus on enhancing renewable energy integration and cross-border energy co-operation. Some of the other combined challenges in all three regions that may hinder GG include varying levels of technological advancement, regulatory differences, and energy resource disparities. These challenges should be addressed by strengthening regional collaborations and standardizing regulations, leading to high-quality economic growth throughout the region.

Limitations

The present study, while informative, is subject to certain limitations that need consideration. Firstly, the macroeconomic variables are subject to external shocks that may introduce the nonlinearity into the series. However, this study estimates the symmetric impact of our selected variables on GG, making our results less interesting and reliable. In the future, the empirics should depend on the asymmetric assumption while estimating the impact of digital exports, financial stability, and energy security risks on GG. Second, the research focuses on regional analysis. A comparative analysis between advanced, emerging, and developing economies could add more credibility to the study. Third, our evaluation of GG is based on the OECD measure. It is recommended that future research consider incorporating alternative measures from the World Bank and UNEP to enrich the empirical analysis. Lastly, future research endeavors should incorporate a broader array of determinants into their analyses to develop a more comprehensive understanding of the factors influencing GG. This may encompass variables like digital financial inclusion, energy prices, and environmental educational factors, among others, to provide a general perspective.

Highlights

This study examines the link between digital exports, financial stability, energy security, and GG.

Employing the CS-ARDL model for analysis.

Findings show that digital exports and financial stability promote long-run GG in the full sample and in Asia, America, and Europe.

Energy security risks hinder long-run GG in the full sample and in Asia, America, and Europe.

Footnotes

List of abbreviations

Declaration of conflicting interests

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

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Sidra Sohail

Yang Yang is currently affiliated with the Hunan Institute of Engineering, Xiangtan, Hunan, China, and the University of Queensland, Brisbane, QLD, Australia. His primary research interests lie in energy and environmental economics.

Xu Yunbao is currently working at the School of Management, Hunan Institute of Engineering, Xiangtan, Hunan, China. He has published extensively in reputed national and international journals.

Xu Qifeng received a bachelor's degree in accounting from Hunan University of Finance and Economics. She continues her master's degree education in the field of Management Science and Engineering at Guanxi University of Science and Technology. Her main research directions include technology innovation management, enterprise innovation, and industrial economy. She has published a few papers on these research areas.

Guan Yongjun obtained a doctorate in Management from Central South University. He serves as the dean of the School of Economics and management at Guanxi University of Science and Technology. He mainly engaged in research related to financial management and technological innovation management, as well as high-quality development of the manufacturing industry.

Sidra Sohail completed her PhD in Economics at the Pakistan Institute of Development Economics, Islamabad, Pakistan. She is currently working as an Assistant Professor at ILMA University, Karachi, Pakistan. Her area of interest is in the field of public finance and fiscal policy. She has published extensively in reputed national and international journals.

References

Fay

. Inclusive green growth: The pathway to sustainable development. Washington, DC: World Bank Publications, 2012.

Mastini

Kallis

Hickel

. A green new deal without growth? Ecol Econ 2021; 179: 106832.

Lin

Zhu

. Fiscal spending and green economic growth: evidence from China. Energy Econ 2019; 83: 264–271.

Hickel

Kallis

. Is green growth possible? New Polit Econ 2020; 25: 469–486.

Zhou

. Can intellectual property rights within climate technology transfer work for the UNFCCC and the Paris Agreement? Int Environ Agreem Polit Law Econ 2019; 19: 107–122.

Minas

. The Paris Agreement's technology framework and the need for 'transformational change’. CCLR 2020; 14: 241.

Rimmer

. Article 10 of the Paris agreement 2015: technology development and transfer. In: A commentary on the Paris Agreement on climate change. Cheltenham and Northampton (MA): Edward Elgar, 2021, pp.237–259.

Chen

Ramzan

Hafeez

, et al.

Green innovation-green growth nexus in BRICS: does financial globalization matter?

J Innov Knowl 2023; 8: 100286.

Shao

Chang

, et al.

Does digital finance promote the green innovation of China's listed companies?

Energy Econ 2022; 114: 106254.

10.

Chen

Lyulyov

Pimonenko

, et al. Green development of the country: role of macroeconomic stability. Energy Environ 2024; 35: 2273–2295.

11.

Razzaq

Yang

. Digital finance and green growth in China: appraising inclusive digital finance using web crawler technology and big data. Technol Forecast Soc Change 2023; 188: 122262.

12.

Ozturk

Ullah

. Does digital financial inclusion matter for economic growth and environmental sustainability in OBRI economies? An empirical analysis. Resour Conserv Recycl 2022; 185: 106489.

13.

. Measuring the combining effects of financial stability and climate risk for green economic recovery. Econ Change Restruct 2023; 56: 1225–1241.

14.

Hassan

Yang

Usman

, et al.

Green growth as a determinant of ecological footprint: do ICT diffusion, environmental innovation, and natural resources matter?

PLoS ONE 2023; 18: e0287715.

15.

Dong

Ullah

. Towards a green economy in China? Examining the impact of the internet of things and environmental regulation on green growth. Sustainability 2023; 15: 12528.

16.

Tang

Jiang

. Effects of digital global value chain participation on CO2 emissions embodied in digital exports: new evidence from PSTR approach. Energy Econ 2023; 126: 106913.

17.

Jänicke

. “Green growth”: from a growing eco-industry to economic sustainability. Energy Policy 2012; 48: 13–21.

18.

Sun

Wang

. Digital trade growth and mineral resources in developing countries: implications for green recovery. Resour Policy 2024; 88: 104338.

19.

Bologna

. Energy and sustainable economic development. In: Renewable energy for unleashing sustainable development: Blending technology, finance and policy in low and middle income economies. Cham: Springer International Publishing, 2013, pp.181–193.

20.

Alsagr

Ozturk

. How do energy security risk and ICT affect green investment? Econ Anal Policy 2024. DOI: https://doi.org/10.1016/j.eap.2024.04.027.

21.

Danish Khan

Haneklaus

. Sustainable economic development across globe: the dynamics between technology, digital trade and economic performance. Technol Soc 2023; 72: 102207.

22.

Huang

, et al. Digital trade and environmental sustainability: the role of financial development and ecological innovation for a greener revolution in China. Econ Res-Ekon Istraž 2023; 36. DOI: https://doi.org/10.1080/1331677X.2022.2125889.

23.

Xiong

Luo

. Investigating the relationship between digital trade, natural resources, energy transition, and green productivity: moderating role of R&D investment. Resour Policy 2023; 86: 104069.

24.

Murshed

. An empirical analysis of the non-linear impacts of ICT-trade openness on renewable energy transition, energy efficiency, clean cooking fuel access and environmental sustainability in South Asia. Environ Sci Pollut Res 2020; 27: 36254–36281.

25.

Zhang

Song

, et al. How financial development and digital trade affect ecological sustainability: the role of renewable energy using an advanced panel in G-7 countries. Renewable Energy 2022; 199: 1005–1015.

26.

Zhang

. Trade openness and green total factor productivity in China: the role of ICT-based digital trade. Front Environ Sci 2021; 9: 809339.

27.

Kashif

Shi

Naseem

, et al. ICT service exports and CO2 emissions in OECD countries: the moderating effect of regulatory quality. Econ Change Restruct 2024; 57: 94.

28.

Zhu

Bao

Qin

. Impact analysis of digital trade on carbon emissions from the perspectives of supply and demand. Sci Rep 2024; 14: 14540.

29.

Zheng

Zhao

, et al. How does digital trade impact urban carbon emissions efficiency? Evidence from China's cross-border e-commerce pilot zones. J Cleaner Prod 2024; 456: 142363.

30.

Cao

Law

Samad

ARBA

, et al. Effect of financial development and technological innovation on green growth—analysis based on spatial Durbin model. J Cleaner Prod 2022; 365: 132865.

31.

Zheng

Ahmed

Xie

, et al.

Green growth and carbon neutrality targets in China: do financial integration and ICT matter?

J Cleaner Prod 2023; 405: 136923.

32.

Safi

Wahab

Zeb

, et al. Does financial stability and renewable energy promote sustainable environment in G-7 countries? The role of income and International trade. Environ Sci Pollut Res 2021; 28: 47628–47640.

33.

Baloch

Danish Meng

, et al. Financial instability and CO2 emissions: the case of Saudi Arabia. Environ Sci Pollut Res 2018; 25: 26030–26045.

34.

Oanh

TTK

Dinh

. Digital financial inclusion, financial stability, and sustainable development: evidence from a quantile-on-quantile regression and wavelet coherence. Sustain Dev 2024. DOI: https://doi.org/10.1002/sd.3021.

35.

Dinh

Oanh

TTK

NTH

. Financial stability and sustainable development: perspectives from fiscal and monetary policy. Int J Finance Econ 2024. DOI: https://doi.org/10.1002/ijfe.2981.

36.

Khan

Yoon

. Financial instability and environmental degradation: a panel data investigation. Appl Econ 2021; 53: 6319–6331.

37.

Osabohien

Adeleke

Osabohien

, et al. Operational risk management, financial sector stability and sustainable development in West Africa. Discov Sustain 2024; 5: 214.

38.

Ozili

Iorember

. Financial stability and sustainable development. Int J Finance Econ 2024; 29: 2620–2646.

39.

Umar

Kirikkaleli

, et al. Testing the asymmetric effect of financial stability towards carbon neutrality target: the case of Iceland and global comparison. Gondwana Res 2023; 116: 125–135.

40.

Nguyen

. Is energy security a driver for economic growth? Evidence from a global sample. Energy Policy 2019; 129: 436–451.

41.

Kartal

. Are the effects of energy security on economic growth symmetric or asymmetric in Turkey? An application of non-linear ARDL. Ege Acad Rev 2022; 22: 487–502.

42.

Banna

Alam

Chen

, et al. Energy security and economic stability: the role of inflation and war. Energy Econ 2023; 126: 106949.

43.

Pata

Karlilar

. The integrated influence of energy security risk and green innovation on the material footprint: an EKC analysis based on fossil material flows. J Cleaner Prod 2024; 435: 140469.

44.

Ayaz

Prodromou

, et al. Energy security dimensions and economic growth in non-OECD Asia: an analysis on the role of institutional quality with energy policy implications. Energy Policy 2024; 188: 114090.

45.

Chen

Msigwa

Yang

, et al. Strategies to achieve a carbon neutral society: a review. Environ Chem Lett 2022; 20: 2277–2310.

46.

Yuan

Umar

, et al.

The race to zero emissions: can renewable energy be the path to carbon neutrality?

J Environ Manag 2022; 308: 114648.

47.

Zhao

Shang

Magazzino

, et al. Multi-step impacts of environmental regulations on green economic growth: evidence in the lens of natural resource dependence. Resour Policy 2023; 85: 103919.

48.

Wang

Guo

Chen

, et al. Carbon peak and carbon neutrality in China: goals, implementation path and prospects. China Geol 2021; 4: 720–746.

49.

Magazzino

. Carbon neutrality and economic development. In: Recent developments in green finance, green growth and carbon neutrality. Amsterdam: Elsevier, 2023, pp.109–127.

50.

Zhou

. Worldwide carbon neutrality transition? Energy efficiency, renewable, carbon trading and advanced energy policies. Energy Rev 2023; 2: 100026.

51.

Chudik

Pesaran

. Common correlated effects estimation of heterogeneous dynamic panel data models with weakly exogenous regressors. J Econom 2015; 188: 393–420.

52.

Cole

. Development, trade, and the environment: how robust is the environmental Kuznets curve? Environ Dev Econ 2003; 8: 557–580.

53.

Jiang

Zhong

, et al. Revisiting environmental Kuznets curve for carbon dioxide emissions: the role of trade. Struct Change Econ Dyn 2019; 50: 245–257.

54.

Grossman

Krueger

. Environmental impacts of a North American Free Trade Agreement (No. 158). 1991.

55.

Zhang

Lyu

, et al. Digital economy: an innovation driving factor for low-carbon development. Environ Impact Assess Rev 2022; 96: 106821.

56.

Wang

Liu

Zhao

, et al. Research on carbon emission reduction effect of China's regional digital trade under the “double carbon” target––combination of the regulatory role of industrial agglomeration and carbon emissions trading mechanism. J Cleaner Prod 2023; 405: 137049.

57.

Magazzino

Santeramo

. Financial development, growth and productivity. J Econ Stud 2023; 51: 1–20.

58.

Zaidi

SAH

Zafar

Shahbaz

, et al. Dynamic linkages between globalization, financial development and carbon emissions: evidence from Asia Pacific Economic Cooperation countries. J Cleaner Prod 2019; 228: 533–543.

59.

Safarzyńska

van den Bergh

. Financial stability at risk due to investing rapidly in renewable energy. Energy Policy 2017; 108: 12–20.

60.

Marti

Puertas

. Sustainable energy development analysis: energy Trilemma. Sustain Technol Entrep 2022; 1: 100007.

61.

Stern

. Biomass and China's carbon emissions: a missing piece of carbon decomposition. Energy Policy 2008; 36: 2517–2526.

62.

Carbó-Valverde

Sánchez

. Financial stability and economic growth. In: Crisis, risk and stability in financial markets. London: Palgrave Macmillan UK, 2013, pp.8–23.

63.

Alsamara

Mrabet

Jarallah

, et al. The switching impact of financial stability and economic growth in Qatar: evidence from an oil-rich country. Q Rev Econ Finance 2019; 73: 205–216.

64.

Hasni

Dridi

Ben Jebli

. Do financial development, financial stability and renewable energy disturb carbon emissions? Evidence from Asia–Pacific economic cooperation economics. Environ Sci Pollut Res 2023; 30: 83198–83213.

65.

Nejati

Shah

. How does ICT trade shape environmental impacts across the north-south regions? Intra-regional and inter-regional perspective from dynamic CGE model. Technol Forecast Soc Change 2023; 186: 122168.

66.

Ahmed

Nathaniel

Shahbaz

. The criticality of information and communication technology and human capital in environmental sustainability: evidence from Latin American and Caribbean countries. J Cleaner Prod 2021; 286: 125529.

67.

Dincer

Rosen

. Energy, environment and sustainable development. Appl Energy 1999; 64: 427–440.

68.

Gao

Ullah

Zafar

, et al.

How does nuclear energy consumption contribute to or hinder green growth in major nuclear energy-consuming countries?

Prog Nucl Energy 2024; 170: 105111.

69.

Porter

. American's green strategy. Sci Am 1991; 264: 168.

70.

Chang

Fang

. Renewable energy-led growth hypothesis: new insights from BRICS and N-11 economies. Renewable Energy 2022; 188: 788–800.

71.

Pesaran

. General diagnostic tests for cross section dependence in panels (No. 1229). CESifo Working Paper, 2004.

72.

Pesaran

Yamagata

. Testing slope homogeneity in large panels. J Econom 2008; 142: 50–93.

73.

Pesaran

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

74.

Westerlund

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

75.

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.

76.

Kao

. Spurious regression and residual-based tests for cointegration in panel data. J Econom 1999; 90: 1–44.

77.

Hao

Umar

Khan

, et al.

Green growth and low carbon emission in G7 countries: how critical the network of environmental taxes, renewable energy and human capital is?

Sci Total Environ 2021; 752: 141853.

78.

Ali

Yuanying

Zhang

, et al. What contributes more to BRI economic growth, renewable or non-renewable energy consumption: a third generation panel data analysis. Environ Sci Pollut Res 2024; 31: 22102–22118.

79.

Pesaran

Shin

Smith

. Pooled mean group estimation of dynamic heterogeneous panels. J Am Stat Assoc 1999; 94: 621–634.

80.

. The impact of natural resources on green growth: the role of green trade. Resour Policy 2022; 78: 102720.

81.

Wang

Ren

Xie

. Going “green trade”: assessing the impact of digital technology application on green product export. Technol Soc 2024; 77: 102487.

82.

Staiger

. Does digital trade change the purpose of a trade agreement? (No. w29578). National Bureau of Economic Research, 2021.

83.

Akinyemi

Osabuohien

Alege

, et al. Energy security, trade and transition to green economy in Africa. Int J Energy Econ Policy 2017; 7: 127–136.

84.

Blum

Legey

. The challenging economics of energy security: ensuring energy benefits in support to sustainable development. Energy Econ 2012; 34: 1982–1989.

85.

Hwang

Díez

ÁS

. Renewable energy transition and green growth nexus in Latin America. Renewable Sustainable Energy Rev 2024; 198: 114431.

86.

Ahmad

Youjin

Žiković

, et al. The effects of technological innovation on sustainable development and environmental degradation: evidence from China. Technol Soc 2023; 72: 102184.

87.

Mohsin

Taghizadeh-Hesary

Iqbal

, et al. The role of technological progress and renewable energy deployment in green economic growth. Renewable Energy 2022; 190: 777–787.