Embracing Eco-Digitalization and Green Finance Policies for Sustainable Environment: Do the Engagements of Multinational Corporations Make or Mar the Target for Selected MENA Countries?

Abstract

It is an undeniable fact that the digital economy and green financing are persistently gaining global attention as effective tools for achieving rapid economic growth and development. However, the environmental effects of these indicators are just evolving, leaving the research community with insufficient policies for harnessing the much-anticipated sustainability agenda. Hence, this research provides the first empirical evidence of the impacts of eco-digitalization and green financing on the sustainable environment in selected Middle East and North African countries from 1995 to 2019. The empirical model considers the roles of multinational corporations, renewable and nonrenewable energy, economic growth, and population growth as covariates models based on the STIRPAT framework. The stated hypotheses are verified based on Cross-Sectionally Augmented Mean Group, Mean Group, Common Correlated Mean Group, and Panel Quantile Regression. Findings show that from eco-digitalization, green financing, and renewable energy drive sustainable environment agenda. On the flip side, nonrenewable energy, economic growth, and population growth largely deter delivering on the blueprint. The estimated results are corroborated by findings from panel quantiles regression. Furthermore, the panel causality uncovers the existence of bidirectional and unidirectional causality in the estimated model. Policy insights that support the pathways toward sustainability in MENA economies are suggested based on the findings.

1. Introduction

The issue of environmental degradation amidst global warming and untiring climate change has become a burning global problem raising concerns about the fragility of sustainable future generations if frantic and conscientious efforts are not taken in this era. Specifically, the issue of environmental sustainability constitutes the core of the three dimensions of sustainable development (SD), suggesting that the vast majority of the stated objective and indicators in the 2030 sustainable development goals (SDGs) would remain unattainable if the issues relating to environmental challenges are not resolved. The major dimensions of SDGs comprise economic, social, and environmental revolving around the seventeen goals [1,2]. However, the environmental dimension has received the most attention from the literature and global agreements in recent times. Hence, various environmental agreements have been concluded and adopted for the past four decades or more with a view to unanimously reaching a global practice that can drive sustainability of the present and future environments. Specifically, the United Nations through its agency on the environment has continually organized the Conference of the Parties (COP) from 1992 (COPI) to 2022 (COP27) with the majority setting a global standard of maintaining global warming less than 2 °C and not above 1.5 °C to ensure the world is save from destruction. Furthermore, it is estimated that global warming is associated with a number of environmental challenges and undesirable health outcomes comprising aggravation of erosion, landslides, salinization, flooding, desertification, and soil biodiversity loss among others.

It is worth noting that the consequences of global warming are more devastating on developing regions of the world such as countries in the MENA region believed to be vulnerable to environmental problems despite contributing less to the stock of global greenhouse gas (GHG) emissions. For instance, evidence abounds that temperature increase in the MENA region is estimated to be two times more than the globally projected average of those categorizing the region as the most exposed to the devastating outcomes of climate change [3]. Corroborating this view, Simone and Elisa [4] opine that the vulnerability of MENA countries to climate change is evident in the extremely unfavorable temperature, insufficient groundwater, and inconsistent rainfall. Available statistics show that carbon emissions in the MENA region have maintained a persistent rise for more than three decades (see Figure 1). This is suggestive of the fact that policy implications are needed to drive some factors and reduce the carbon emission surge.

Although the empirical literature is awash with a series of factors moderating the surge in carbon emissions, the problem remains insignificantly resolved, thus necessitating further efforts in researching the best practices that can be adopted along with the identified factors. Among the various drivers of environmental sustainability, eco-digitalization, and green finance constitute the most recent believed to be effective and predictable in achieving the goals of a sustainable environment. The choice of eco-digitalization is based on the grounds that aside from the fact that the digital economy is noted to mitigate environmental degradation, eco-digitalization provides mitigation measures and is perceived to be more efficient [5]. Furthermore, green finance has been advanced as a useful tool for achieving a sustainable environment through the various green capital projects that are financially promoted [6,7]. Similarly, the roles of Multinational Corporations (MTCs) in environmental debates have been argued from two angles comprising the pollution haven hypothesis and pollution halo hypothesis. In the former hypothesis, it is assumed that MTCs transfer environmental pollution to the host countries through dumping and exploitation of weak environmental regulations. In the latter hypothesis, MTCs are observed to drive environmental quality by transferring green technology to the host countries. Consequently, it becomes highly pertinent to examine which of the two hypotheses empirically apply to the environmental situation of the MENA region.

An assessment of the trend in the drivers of environmental sustainability in MENA countries provides some insightful details worthy of noting for understanding the peculiarity of the economies. For instance, eco-digitalization is maintaining some appreciable development over the past two decades as evident in Figure 2. In addition, the region has witnessed unprecedented progress in green finance with a persistent rise for more than 20 years as exposited in Figure 3. Moreover, the contributions of MTCs to the overall growth of the MENA region recorded some percentage increases which were however followed by decreases in the last and recorded persistent decline in the last 12 years (see Figure 4). That notwithstanding, the reported contributions to GDP significantly impact the growth of the economy and variation in the quality of the environment.

The evolving nature of empirical research gauging the tripartite impacts of eco-digitalization, green finance, and Multinational Corporations (MTCs) on environmental sustainability motivates the research interest of the current paper. Hence, the objective of this study is to examine the tripartite effects of eco-digitalization, green finance, and Multinational Corporations in MENA countries. The study relies on the theoretical underpinning of Stochastic Impacts by Regression on Population, Affluence, and Technology (STIRPAT) to estimate the extent to which eco-digitalization, green finance, Multinational Corporations (MTCs), renewable and nonrenewable energy, economic growth, and population growth on carbon emissions in a panel of nine selected MENA economies. The findings from the analyses are expected to answer the following questions; (1) what are the effects of eco-digitalization, green finance, and MTCs on environmental sustainability in the MENA region? (2) to what extent can the engagement of renewable and nonrenewable energy influence the variation in the sustainable environment of the MENA economies? (3) how significant are the distributional effects of the exogenous variables on environmental sustainability in MENA?

Leveraging on the preceding research objectives and questions, the study provides three novelties to the extant literature. First, the environmental effects of eco-digitalization are an evolving research area and when it comes to the case of MENA countries, we are not aware of any existing empirical evidence. Hence, this study constitutes the first empirically backed evidence on the nexus between eco-digitalization and carbon emissions in a panel of selected MENA countries. Second, the assessment of green finance in a single model with eco-digitalization and Multinational Corporations (MTCs) in tripartite effects is the first to be advanced in this study. Third, exploring the STIRPAT model in an extended version incorporating renewable and nonrenewable energy, eco-digitalization, green finance, and MTCs is believed to be novel in deducing practicable policy implications for achieving a sustainable environment in the MENA region.

The analyses In the current study unveil some policy implications that advance the pathway to achieving a sustainable environment in MENA. To achieve the stated objectives, we subject the dataset to preliminary analyses such as checking the averages of the series through summary statistics and ascertaining the nature of the dataset to know whether it is normally or abnormally distributed. We found the dataset is abnormally distributed. More so, the correlation analysis conducted indicated that the model is not affected by the issue of multicollinearity. The study conducts advanced pretests such as cross-sectional dependence and slope homogenous tests of which the results reveal that the model is affected by spatial effects from among the cross-section units and heterogeneously sloped. Second-generation unit root tests were conducted and it became obvious that the series are only stationary after the first differencing. The cointegration test supports the existence of longrun nexus among the indicators. The main empirical analyses based on CS-ARDL, CCEMG, AMG, and PQR estimators reveal that eco-digitalization, green finance, foreign direct investment proxing Multinational Corporation, and renewable energy significantly drive environmental sustainability by moderating the surge in carbon emissions. On the contrary, nonrenewable energy and population growth hinder the pathway to environmental sustainability due to their positive impacts on carbon emissions.

Asides from Section 1, the other sections make the whole structure of the study slated; thus, Section 2 reviews the relevant studies on the subject matter of the current research, Section 3 models the hypotheses and research objectives with theoretical underpinning, Section 4 focuses on results and discussion of the estimated model, and Section 5 presents the conclusion, emanates policy implications, and limitations for future research opportunities.

2. Literature Review

This section focuses on the assessments of the existing studies relating to the impacts of the key exogenous variables on carbon emissions. Starting with digitalization, Shen et al. [5] estimate the effects of digitalization amidst the presence of green finance, green hydrogen, environmental-related technology, and energy efficiency in the seven most rated economies in hydrogen consumption. The model runs from 1995 to 2019 with subject to cross-section autoregressive distributed lag as the main estimator. The empirical evidence extends its contributions with the consideration of the Common Correlated Effect Mean Group, Augmented Mean Group, and Method of Moment Quantile Regression as robustness to the main estimator. Findings reveal that digitalization significantly drives a sustainable environment by mitigating carbon emissions. Similarly, green hydrogen, energy efficiency, green finance, structural change, and environmental-related technologies support the sustainability agenda. On the flip side, urbanization and natural resources hinder the achievement of the sustainability agenda. Dong et al. [8] investigate how the progress of the digital economy moderates carbon emissions toward achieving carbon neutrality in a panel of sixty countries across the globe from 2008 to 2018. Results uncover that digital economy advancement reduces carbon emission intensity. On the contrary, the development of the digital economy escalates per capita carbon emissions. Moreover, covariates such as financial development, industrial structure, and economic growth mediate between the digital economy and carbon emissions.

The role of green finance in the environmental debates has become a point of empirical attraction in recent times motivated by the possibilities of deducing substantial policy implications for a sustainable environment. For instance, Xiong and Sun [9] evaluate the association of green finance with carbon emissions with a view to explicate the pathway towards achieving environmental sustainability in 34 regions of the Chinese economy. The intervening roles of green innovation, industrial structure, and green investment are equally investigated. Findings from the study reveal that the highlighted exogenous indicators are substantial enough to reduce carbon emissions. Similarly, Cao [10] assesses the relationship between green finance, per capita income, technical innovation, and green energy on green economic performance in E7 countries from 2005 to 2018. Empirical outcomes reveal that green finance, per capita income, green energy, and technological innovation significantly reduce carbon emissions. Ran et al. [11] examine the extent to which green finance substantially drives the pathway towards achieving the joint objectives of carbon peak and carbon neutrality in selected provinces in China from 2007 and 2019. Feedbacks reveal that green finance promotes carbon emission efficiency.

Considering the environmental effects of Multinational Corporations, we explore studies focusing on how FDI impacts the variation in environmental indicators. This is pertinent on the ground that the role of foreign direct investment (FDI) in the environment empirics is well documented. For instance, Wei et al. [12] examine the effects of FDI on carbon emissions in the presence of urbanization for a panel of the Belt and Road Initiative (BRI) region based on annual data from 2000 to 2018. Findings from the analyses reveal the BRI reported a carbon emission rise of 253 million tons with an estimated 3.68% annual growth. Furthermore, it is noted that FDI escalates local emissions. Apergis et al. [13] probe the environmental impacts of FDI in BRICS economies from 1993 to 2012. Feedbacks reveal that diverging effects of FDI in relation to the dimension of selected economies from where FDI flows to the BRICS economy. For example, FDI inflows from Denmark and the United Kingdom provide escalating impacts on carbon emissions to establish the existence of the pollution haven hypothesis. On the contrary, FDI inflows from France, Germany, and Italy moderate the surge in carbon emissions suggesting the validity of the pollution halo effect.

The review of the empirical literature indicates that empirical evidence explaining the role of eco-digitalization on environmental sustainability is evolving. The majority of the existing studies focus on digitalization without considering the environmentally sustainable aspect of it. Hence, this provides an extension of the knowledge frontier on the nexus between digitalization and the sustainable environment.

3. Method

3.1. Data, Measurements, and Sources

The underlining data employed for the current analysis on the dual roles of eco-digitalization and green finance in the drive towards a sustainable environment in the MENA economies is exposited in this section. First, longitudinal data for ten selected MENA economies are utilized from 1995 to 2019. For robust empirical outcomes, the intervening roles of multinational corporations (foreign direct investment), renewable, and population growth. The rationale for the period of analysis is influenced by the availability of observations for the selected countries. Specifically, the endogenous variable is the sustainable environment captured by carbon emissions. The exogenous indicators are eco-digitalization, green finance, FDI, renewable energy, and population growth. Further information on the data, sources, and measurements of the various indicators are provided in

Table 1. Description of variables and sources.

Variables	Name	Description	Sources
SUSENV	Sustainable environment	CO2 emissions (metric tons per capita)	WDI
ECODIG	Eco-digitalization	Climate change mitigation in information and communication technologies (ICT)	OECD
GFIN	Green finance	Climate change adaptation technologies	OECD
FDI	Foreign direct investment	Foreign direct investment, net inflows (% of GDP)	WDI
REN	Renewable energy	Renewable energy consumption (% of total final energy consumption)	WDI
NONE	Nonrenewable energy	Fossil fuel energy consumption (% of total)	WDI
ECG	Economic growth	GDP per capita (constant 2015 USD)	WDI
POPG	Population growth	Population growth (annual %)	WDI

World Development Indicator (WDI), and Organization for Economic Cooperation and Development (OECD).

.

3.2. Theoretical Framework and Strategic Modeling

To position the arguments in the current study within the art of empirical evidence on sustainable environment, this study relies on the novel Stochastic Impacts by Regression on Population, Affluence, and Technology (STIRPAT) model propounded by Dietz and Ros [14].

Empirical evidence has highlighted quite substantial benefits of the STIRPAT model of which the ability to account for the non-monotonic impacts of the different environmental factors are apparent [15,16,17].

Conventionally, the STIRPAT model provides strong justification for the effects of population and affluence on the increase and decrease in global greenhouse gas (GHG) emissions [18]. The model takes its strength from three key indicators comprising population (P), affluence (A), and technology (T) which can be modeled; thus,

$$I=\beta P_i^{\vartheta 1}\times A_i^{\vartheta 2}\times T_i^{\vartheta 3}\times {\pi }_i.$$

(1)

Following the mathematical expression in Equation (1), it is important to note that environmental sustainability is denoted by I as a proximate for carbon emissions. Further, $\beta$ is the constant while the exponential determinants of P, T, and A are represented by ${\sigma }_1,{\sigma }_2,{\sigma }_3$, respectively. More so, $\pi$ implies a term for the stochastic variable. The model is expressed in natural log form to estimate the stated nexus as follows;

$$\ln I_{it}={\beta }_0+{\sigma }_1(\ln P_{it})+{\sigma }_2(\ln A_{it})+{\sigma }_2(\ln T_{it})+{\pi }_i.$$

(2)

P in the above model is denoted by population growth (POPG), A is represented by economic growth (ECG) measured by GDP per capita, and T is represented by eco-digitalization [5,19]. It is worth noting that the STIRPAT framework requires modification in line with the objective of an empirical model to accommodate the other factors of the environment that are not specified by the model. Following the preceding narratives, the current study relies on the extant study such as [5,7,20] with adjustments to align with the focus of the current paper IIh is thus slated:

$$\begin{array}{l}SUSEN{V}_{it}={\sigma }_0+{\sigma }_{1}ECODI{G}_{it}+{\sigma }_{2}GFI{N}_{it}+{\sigma }_{3}FD{I}_{it}+{\sigma }_{4}RE{N}_{it}\\ +{\sigma }_{5}NON{E}_{it}+{\sigma }_{6}EC{G}_{it}+{\sigma }_{7}POP{G}_{it}+{\alpha }_{it}\end{array}.$$

(3)

In Equation (3), ENVSUS represents sustainable environment captured carbon emissions. ECODIG denotes eco-digitalization, and GFIN signifies green finance. FDI implies foreign direct investment capturing the environmental impacts of multinational corporations. REN and NONE are representing renewable and nonrenewable energy. ECG signifies an economic group denoting the effects of affluence. POPG is population growth. The disturbance term is captured by $\alpha$, and $t$ signifies the time ranging from 1995 to 2019. The selected 10 MENA economies Algeria, Egypt, Iran, Iraq, Jordan, Morocco, Saudi Arabia, Sudan, Tunisia, and the United Arab Emirates.

3.3. A Priori Expectations

The economic intuitions guiding the nexus between the dependent and independent variables in the present study are explicated in line with the existing studies. Starting with eco-digitalization, ample empirical studies have provided substantial evidence positing that progress in the digital economy serves as a substantial tool for achieving environmental sustainability through the decline in carbon emissions [5,19,21,22]. The economic intuition behind these submissions is based on the ground that digitalization opens up new possibilities for reducing greenhouse gas emissions and optimizing energy use. Emissions generated through power are reduced as a result of the improvement on the grid during electricity generation and the electric vehicles which help reduce transport-related emissions are enhanced through digitalization. Consequent upon the foregoing, we hypothesize an inverse relationship between digitalization and carbon emissions as follows: $\sigma =\frac{\Delta C{O}_2}{\Delta ECODIG}>0$.

The roles of green finance in carbon emission reduction are gaining ground in recent times with arguments focusing on the fact that green projects help mitigate carbon emission surges and promote environmental sustainability. Specifically, an appreciable strand of empirical studies posits that green finance significantly mitigates carbon emissions [7,9,11]. In essence, we expect the coefficient of green finance to negatively relate with carbon emissions as $\sigma =\frac{\Delta C{O}_2}{\Delta GFIN}>0$. The impacts of Multinational Corporation on environmental sustainability are often argued from the viewpoint of technological transfer and spillover in technical diffusion from sustainable practices. On the flip side, carbon-intensive Multinational Corporations often exacerbate the level of environmental degradation due the production activities that do not conform to sustainable practices. This phenomenon often occurs under the concept of the Pollution Haven Hypothesis. The preponderance of existing studies establishes the moderating and inducing roles of Multinational Corporations in carbon emissions [13,23,24]. In view of the preceding, we envisage both negative and positive nexuses between Multinational Corporation (FDI) and carbon emissions as follows: $\sigma =\frac{\Delta C{O}_2}{\Delta FDI}>0$ and $\sigma =\frac{\Delta C{O}_2}{\Delta FDI}<0$.

There is a growing body of empirical evidence for the submission that renewable energy is a viable tool that helps in moderating the surge of GHG emissions and, at the same time, promotes green practices. This suggests why renewable energy has been advanced as a moderating factor of carbon emissions [25,26,27]. Consequently, we hypothesize an inverse association between renewable energy and carbon emission as thus $\sigma =\frac{\Delta C{O}_2}{\Delta REN}<0$. Nonrenewable energy is a significant factor driving the rise in GHG emissions which is empirically documented [25,27,28,29]. As such, a positive relationship is expected between nonrenewable energy and carbon emissions as stated by $\sigma =\frac{\Delta C{O}_2}{\Delta NONE}>0$. The effects of population growth have often been observed to induce significant surges in carbon emissions and this submission is well documented in the literature [30,31]. Hence, a direct association is hypothesized between population growth and carbon emissions, thus $\sigma =\frac{\Delta C{O}_2}{\Delta POPG}>0$.

Consequently, upon the preceding a priori, the null hypotheses to be verified empirically can be stated as

• Eco-digitalization does not significantly impact carbon emissions in MENA countries.
• The impacts of green finance are not significant on carbon emissions in MENA countries.
• Multinational Corporations do not significantly contribute to carbon emissions in MENA countries.
• Renewable and nonrenewable energy are not substantial in influencing carbon emissions in MENA countries.
• Population growth does not significantly impact carbon emissions in MENA countries.

3.4. Estimation Steps

To arrive at the most apt and accurate empirical implications, the conventional estimation procedures are followed in examining the extent to which the moderating roles of eco-digitalization and green finance drive a sustainable environment in the selected MENA countries. A careful search into the recent empirical studies, such as [5,31,32], unveils some prominent steps comprising examining the levels of homogeneity of the slope coefficients, the presence or absence of cross-sectional dependence, the stationarity and long-run natures of the series, assessments of the long-run effects of the exogenous indicators on the outcome variable, and an evaluation of the causality in the estimated model. The preceding steps are presented in Figure 5.

The empirical procedure presented in Figure 5 suggests two angles in the assessment of the dataset and estimators that fit into a regression analysis. For instance, there is a need to conduct a cross-section dependence test and homogenous slope test to ascertain whether the data are affected by unknown common factors or not. The presence of the two tests imply that second-generation estimators in terms of unit root tests such as cross-sectional ADF and cross-sectional IPS unit root units are most appropriate in examining if the series are spurious or not. In the absence of both tests, first-generation estimators such as the conventional ADF and IPS unit root tests are deemed fit for the unit root analysis. The choice of estimators for the subsequent analyses will depend on the adopted line of techniques (first or second generation). Upon verifying the long-run effects, analyses are conducted based on panel quantile regression to extract the distributional effects of the exogenous variables. Similarly, causality tests are conducted and the necessary conclusion, policy implications, and limitations are duly explained.

3.5. Preliminary Analyses

This research explores three windows to present the nature and characteristics of the data utilized in carrying out the empirical verification of the stated objective. At first, summary statistics are employed to explain the average values of each of the indicators over the period of study. Furthermore, normality tests are presented to clarify the extent to which the distribution of the variables is normal or abnormal. More so, the extent to which the model is free from any form of multicollinearity is proven by conducting a correlation matrix analysis. We embark upon these preliminary analyses to aid an intensive understanding of the nature of the various selected environmental indicators in MENA. The values presented in

Table 2. Summary statistics and normality analysis.

	CO2	ECODIG	GFIN	FDI	REN	NORE	ECG	POPG
Mean	6.06	1.27	2.29	2.28	11.17	88.09	8.52	2.49
Median	3.00	1.00	1.50	1.59	2.00	97.57	8.24	2.02
Maximum	30.88	5.15	54.42	23.54	83.61	101.72	11.03	18.13
Minimum	0.18	0.25	0.17	−4.54	0.01	12.97	7.30	−0.85
Std. Dev.	7.25	1.06	4.35	2.96	20.78	22.21	0.97	2.23
Skewness	1.84	2.12	8.16	2.68	2.41	−2.34	1.35	4.17
Kurtosis	5.43	7.14	89.03	16.28	7.53	7.11	3.71	24.99
Jarque-Bera	200.73	361.87	78,906.80	2109.79	450.47	399.29	80.51	5693.73
Probability	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00
Bivariate correlation analysis
CO2	1.00
ECODIG	−0.01	1.00
GFIN	0.19	−0.12	1.00
FDI	−0.06	−0.06	−0.13	1.00
REN	−0.38	−0.03	−0.09	0.09	1.00
NORE	0.27	0.07	0.08	−0.08	−0.46	1.00
ECG	0.97	−0.08	0.21	−0.03	−0.44	0.32	1.00
POPG	0.43	−0.07	−0.06	0.16	−0.07	0.08	0.42	1.00

indicate that carbon emissions average 6.06%, eco-digitalization averages 1.27%, green finance averages 2.29%, and foreign direct investment averages 2.28%. Furthermore, it is evident that renewable energy which averages 11.17% is consumed far less than nonrenewable energy which averages 88.09%, thus suggesting that the majority of the MENA countries are fossil fuels dependent. This submission aligns with empirical evidence that was developed by [33,34]. Economic growth stands at 8.52%, whereas population growth is at an average rate of 2.49% for the period under study.

The dataset employed in the current research appears abnormally distributed following the values presented for the Skewness, Kurtosis, and Jarque–Bera tests. Moreover, the bivariate correlation matrix in Table 2 shows that the stated empirical model is not affected by the econometric issue of multicollinearity. This is proven by the low values observed in the relationships between each of the variables of interest.

4. Results and Discussion

4.1. Results of the Correlation, Interdependence and Slope Coefficient Tests

The correlation analysis across the selected cross-sections in

Table 3. Correlation, interdependence, and slope coefficient tests.

Indicators	Correlation	Pesaran (2004)
CO2	0.709	6.370 ***
ECODIG	0.855	9.235 ***
GFIN	0.714	8.059 ***
FDI	0.965	10.810 ***
REN	0.669	7.045 ***
NORE	0.858	10.225 ***
ECG	0.957	14.990 ***
POPG	0.773	8.914 ***
Slope Coefficient
Delta	8.672 ***	0.000
Adj. Delta	9.948 ***	0.000

The test values are noted as significant at 1% signified by the ***.

shows a high level of interaction in the panel model specifically with the rates ranging between 67% and 97%. The independence measured by cross-sectional dependence (CSD) tests is examined based on [35]. The outcomes in Table 3 reveal the presence of CSD, thus implying an outright rejection of the null hypothesis as evidenced by the significance of the probability values at the 1% level. Hence, the high correlation rates and presence of CSD specify that macroeconomic indicators in one of the MENA economies have the potential to instigate substantial variation in others. The feedbacks of the slope coefficient in Table 3 reveal substantial evidence to advance the presence of heterogeneous slope coefficient following the significance of the delta and adjusted delta tildes statistics.

The econometric implication of the above results leads to the adoption of second-generation methods in conducting stationarity and long-run tests. Similarly, examining the effects of the exogenous indicators on the outcome variable would be most appropriate through the second-generation estimators [32].

4.2. Stationarity Test Results

The cross-sectional augmented panel unit root ADF (CADF) Pesaran (2003) [36] and IPS (CIPS) Pesaran (2007) [37] tests are used to implement the second-generation unit root technique because the empirical model of the current investigation has cross-sectional dependence and slope heterogeneity. The results in

Table 4. Results of CADF and CIPS stationarity tests.

Variables	CADF (2003)		CIPS (2007)
	Level	First Difference	Level	First Difference
CO2	−1.028	−2.964 *	−1.079	−4.135 ***
ECODIG	−2.118	−4.108 ***	−2.973	−6.964 ***
GFIN	−1.523	−6.420 ***	−0.865	−6.014 ***
FDI	−2.599	−3.464 ***	−1.342	−5.154 ***
REN	−2.019	−2.967 **	−1.530	−4.392 ***
NORE	−2.030	−3.252 ***	−2.285	−4.840 ***
ECG	−2.463	−2.939 **	−1.787	−3.580 ***
POPG	−2.502	−2.817 **	−1.945	−4.743 ***

Values in ***, **, and *, denote significance levels at 1%, 5% and 10%.

show that the series have unit roots that are stationary at the first difference The variables, however, became stationary when the series was subjected to the first difference, which reduced the integration to I (1).

The Westerlund (2007) [38] test results, which are shown in

Table 5. Results Westerlund longrun test.

	Gt	Ga	Pt	Pa
Statistic	−4.065 *	−10.256 ***	−9.778 ***	−10.338 ***

***, and * imply significant level at 1%, and 10%.

, show that the group (Ga and Gt) and panel (Pa and Pt) statistical values do not support the null hypothesis that there is no long-term association leading to the acceptance of the alternative hypothesis positing the existence of long-run relationships among the variables. As a result, we draw the conclusion that eco-digitalization, green finance, Multinational Corporations, renewable and nonrenewable energy, economic growth, and population growth have a long-term impact on environmental sustainability in MENA countries.

4.3. Longrun Results

The feedback from the cointegration test provides the level ground to assess the effects of the exogenous variables on the endogenous variable in the long-run. However, considering the presence of econometric issues such as slope heterogeneity and cross-sectional dependence and the need to derive estimates for the short-run and long-run relationships, we resorted to employing cross-sectionally dependent (CS-ARDL). To further extend the contributions of the study on the extant environmental study, three additional estimators are employed comprising panel quantile regression (PQR), the common correlated effects mean group (CCEMG), and the augmented mean group (AMG).

The results from CS-ARDL, CCEMG, and AMG are presented in

Table 6. Empirical results based on CS-ARDL, CCEMG, and AMG.

	CS-ARDL		CCEMG	AMG
Variables	Short-Run	Longrun
ECODIG	−0.308 ***	−0.227 ***	−0.223 **	−0.069 **
	(0.028)	(0.077)	(0.081)	(0.028)
GFIN	−0.052 ***	−0.029 **	−0.046 ***	−0.044 ***
	(0.019)	(0.013)	(0.015)	(0.013)
FDI	−0.037 **	−0.027 **	−0.021	−0.044 ***
	(0.016)	(0.012)	(0.017)	(0.013)
REN	−0.395 ***	−0.354 ***	−0.191 ***	−0.062
	(0.079)	(0.073)	(0.053)	(0.057)
NONE	0.094 **	0.053 **	0.186 ***	0.141 **
	(0.043)	(0.023)	(0.055)	(0.065)
ECG	3.064 *	1.433 **	2.465 ***	3.886 **
	(0.591)	(0.608)	(0.765)	(1.638)
POPG	0.141 **	0.129 ***	0.122 **	0.088
	(0.058)	(0.041)	(0.049)	(0.051)
ECM (−1)	−0.861 ***
	(0.151)
_cons	2.501	1.432	0.252	(1.768)
	(4.291)	(2.079)	(1.554)	(0.287)
RMSE	0.032	0.026	0.022	0.037
Observations	240	240	250	250

***, **, and * imply significant level at 1%, 5% and 10%.

. Starting with the impacts of eco-digitalization (ECODIG), it is evident that this indicator significantly mitigates carbon emissions in the short run and long-run. Specifically, a proportionate increase in the adoption of digitalization in economic activities will lead to a substantial decrease in carbon emissions, thus flattering the emission curve toward the sustainability of the environment. The economic intuition of these outcomes and their empirical regularity can be argued from three angles. Digitalization reduces the engagement of manpower in production activities and services thereby reducing the carbon footprint generated by human activities. For instance, the emergence of automated teller machines (ATMs) has reduced the role of human efforts in the banking hall and improved financial services. Similarly, the wide adoption of email has reduced poster services involving human involvement. Second, digitalization in building and housing systems has led to the utilization of smart heating and cooling systems, connected appliances, and energy management systems which have increased comfort while using less energy. Third, digitalization has led to maximizing the use of resources and energy, enhancing supply chain management, and enabling product customization which has contributed significantly to driving a sustainable environment. This result is in consonant with previous studies such as [5,21,39] that reported substantial moderating impacts of digitalization on carbon emissions and others.

Furthermore, the findings show that green finance (GFIN) mitigates carbon emissions in the short-run and long-run by exerting negative and significant impacts on it. This thus suggests that a percentage rise in the rate of investment and execution of green capital projects will lead to a substantial decline in carbon emissions. This result conforms to economic intuition on the ground that implementing developmental projects is complicated in terms of huge finance and the ensuing environmental effects. However, the engagements of green finance through green bonds and carbon market instruments may be a potent tool to stimulate green development in the MENA economies which promotes economic growth without compromising environmental sustainability. Furthermore, green finance enhances pilot initiatives with favorable financial, social, and environmental effects. In fact, long-term financial gains from green finance investments in the green economy are possible. Consequently, promoting green finance substantially leads the way toward achieving environmental sustainability. Ample empirical findings have provided evidence to support the arguments that green finance is a useful tool for promoting environmental sustainability through the reduction in the stock of carbon emissions [7,40].

The environmental impacts of multinational cooperation captured by FDI are provided in Table 6. The outcomes reveal that FDI drives a sustainable environment by significantly reducing carbon emissions by 037% and 027% in the short run and long-run, respectively. This result is an intuition for the MENA countries on the ground that the operation of the majority of the multination corporations in the region is usually guided by global standards and practices that often conform with the environmental sustainability agenda. The preceding narratives can be argued under the much-celebrated pollution halo hypothesis (PHAH) which posits that multinational corporations transfer greener technology through FDI to support environmental sustainability in the host country [13]. The reported moderating role of FDI on carbon emissions in the preceding results complements previous submissions by [41,42,43].

The feedbacks on the nexus between renewable energy and carbon emission in Table 6 show that renewable drives a sustainable environment in MENA economies as evidenced by the negative and significant coefficients in the short-run and long-run. Intuitively, a percentage rise in renewable energy consumption will reduce carbon emissions in the MENA countries. The environmental effects of renewable energy justifying its criticality in the drives towards sustainability can be advanced on the ground that it enhances the production of energy without emitting any greenhouse gases, while also lowering some types of air pollution. In addition, it increases energy supply diversity and lowers reliance on imported fuels. The empirical outcomes in this study agree well with [23,33] that posit the existence of an inverse relationship between the two. The various submissions in favor of renewable energy in environment empirics justify why it is perceived as a key determining factor of the carbon neutrality targets set by many nations [29].

On the flip side, the examined effects of nonrenewable energy unveil the exacerbating effects of fossil fuels on the environment by inducing significant surges in carbon emissions in the MENA countries. The result is quite in line with the tune of events in developing nations like MENA where the majority of the economic activities depend hugely on fossil fuel consumption. An appreciable strand of extant studies provides empirical evidence supporting the inducing impacts of nonrenewable energy on carbon emissions [27,44]. In specific terms, diversifying away from reliance on fossil fuels becomes highly pertinent for the countries in the MENA region going by the carbon-inducing impacts of the energy resources. Furthermore, the fact that developing nations such as those in MENA are more affected by global warming suggests the urgency of policy measures that will help mitigate the emission surge and salvage the people and environment toward sustainable life and environment.

The role of economic growth in driving the surge in carbon emissions is evident from the reported parameters with the coefficient positively and significantly affecting carbon emissions. This implies that a percentage rise in economic growth leads to a corresponding increase in carbon emissions. The results support previous studies such as [31,45]. The effects of population growth on carbon emissions are reported to be positive and statistically significant, suggesting that a substantial increase in population leads to a significant rise in carbon emissions. These results support previous studies such as [30,31,46] that posit the existence of a direct relationship between population and carbon emissions.

The extent to which the disequilibrium caused in the short run can be corrected is evident in the error correction model (ECM) which posits the existence of an 86% rate of adjustments. This rate looks high enough to bring about a corrective measure within the span of a year. For a detailed viewing of the empirical outcomes at a glance, the various directions of positive and negative impacts are presented in Figure 6.

4.4. Disintegration of the Exogenous Effects Using Panel Quantile Regression

The current research extends the frontier of knowledge on the sustainable environment in MENA by estimating the disintegrated effects of eco-digitalization, green finance, Multinational Corporations, renewable and nonrenewable energy, economic growth, and population growth on the sustainable environment based on the novel quantile regression method. The choice of panel quantile regression hinges on the fact this estimator is effective in dissecting the impacts of each exogenous indicator in varying quantiles. Specifically, we employ nine-level quantile distributional effects from the 10th to the 90th quantiles categorized into lower, middle, and higher quantiles. The results as presented in

Table 7. Results of the panel quantile regression.

Indicators	Dependent Indicator: Carbon Emissons
	10th	20th	30th	40th	50th	60th	70th	80th	90th
ECODIG	−0.516 ***	−0.524 ***	−0.533 ***	−0.551 ***	−0.632 ***	−0.677 ***	−0.692 ***	−0.674 ***	−0.932 ***
	(0.095)	(0.096)	(0.103)	(0.09)	(0.1)	(0.119)	(0.157)	(0.191)	(0.301)
GFIN	−0.051	−0.043	−0.038 **	−0.036 **	.034 *	−0.028 ***	−0.024 ***	−0.022 ***	−0.015 **
	(0.029)	(0.024)	(0.016)	(0.013)	(0.015)	(0.008)	(0.004)	(0.009)	(0.006)
FDI	−0.068 **	−0.065 **	−0.072 *	−0.068 **	−0.075 **	−0.085 *	−0.108 *	−0.166 **	−0.203 *
	(0.034)	(0.028)	(0.037)	(0.032)	(0.036)	(0.043)	(0.056)	(0.069)	(0.108)
REN	−0.038	−0.048	−0.061 **	−0.065 ***	−0.073 ***	−0.079 ***	−0.087 **	−0.095 **	−0.124 ***
	(0.023)	(0.023)	(0.024)	(0.021)	(0.024)	(0.029)	(0.037)	(0.045)	(0.072)
NORE	0.037 ***	0.045 **	0.052 **	0.067 ***	0.079 ***	0.108 ***	0.145 ***	0.167 ***	0.174 ***
	(0.012)	(0.022)	(0.022)	(0.019)	(0.021)	(0.025)	(0.033)	(0.04)	(0.063)
ECG	6.224 ***	6.63 ***	6.578 ***	7.184 ***	7.407 ***	7.626 ***	7.73 ***	8.04 ***	7.498 ***
	(0.151)	(0.152)	(0.164)	(0.143)	(0.159)	(0.189)	(0.25)	(0.304)	(0.479)
POPG	0.031	0.027	0.153 ***	0.161 ***	0.182 ***	0.195 ***	0.202 **	0.211 *	0.494 ***
	(0.054)	(0.054)	(0.058)	(0.051)	(0.056)	(0.067)	(0.089)	(0.108)	(0.170)
_cons	−47.916 ***	−50.159 ***	−49.79 ***	−57.144 ***	−58.995 ***	−62.153 ***	−63.899 ***	−68.06 ***	−59.104 ***
	(2.825)	(2.836)	(3.053)	(2.663)	(2.966)	(3.519)	(4.664)	(5.676)	(8.937)
Observations	250	250	250	250	250	250	250	250	250

***, **, and * imply significant level at 1%, 5% and 10%.

provide substantial evidence to advance the moderating effects of eco-digitalization on carbon emissions from the 10th to the 90th quantiles. This implies that the initial flagship of environmentally friendly digitalization significantly reduces carbon emissions and a further advancement in digitalization further drives a sustainable environment through the reduction in the stock of carbon emissions. It is interesting to note that the environmental impacts of ECODIG are on increasing returns giving the rising magnitudes from the lower to higher quantiles.

The distributional effects of green finance are noted to significantly moderate carbon emissions from the 30th to 90th quantiles. This implies that the initial phase of the green project might not be effective due to the time frame needed because of the capital nature of it. Consequently, it is intuitional to see GFIN drive a sustainable environment from the 30th quantiles to the 90th quantiles. Moreover, Multinational Corporations through FDI drive a sustainable environment by significantly mitigating carbon emissions from the 10th to 90th quantiles. The disintegrated impact effects of renewable energy are significant from the 30th to 90th quantiles. These effects are negatively significant suggesting a substantial reduction in carbon emissions in the presence of renewable energy consumption. On the contrary, nonrenewable energy positively and significantly drives the surge in carbon emissions, thus implying a hindrance to the sustainable environment. The distributional effects of economic growth are observed to exacerbate the MENA environment across the nine quantiles. This implies that every stage of production activities contributes significantly to the surge in carbon emissions. This is intuitional considering the fact that the MENA economies are fossil fuel dependent and still at the primary and manufacturing stages of development. Population growth deteriorates the environment from the 30th quantile across the 90th suggesting that the effects of population on the environment only become apparent with a persistent increase in population that causes a burden on nature.

5. Conclusions, Policy Insights, and Limitations

The persistent surge in global warming constitutes one of the greatest challenges of the current era despite frantic efforts to subdue the devastating effects stemming from the environmental menace. Within this line of thought, it is noted that developing countries that the complications on the environment and humanity’s peaceful co-existence seem highly devastating in developing regions like MENA. Consequently, the present study examines the extent to which eco-digitalization, green finance, multinational corporations, renewable and renewable energy, economic growth, and population growth drive or drag environmental sustainability in a panel of selected MENA countries from 1995 to 2019. The empirical model is theoretically guided by the novel STIRPAT model and estimated based on advanced second-generation estimators comprising CS-ARDL, CCEMG, AMG, and PQR. Feedbacks from the analyses reveal that eco-digitalization, green finance, renewable energy, and FDI significantly reduce carbon emissions, thus promoting the drive towards a sustainable environment. Conversely, nonrenewable energy, economic growth, and population growth deter sustainable environment agenda.

Based on the reported empirical outcomes, the following policies are recommended for embracing a sustainable environment in MENA.

First, the moderating effects of eco-digitalization imply that investing more toward digitalizing various economic activities could be much more effective in reducing the exacerbating role of economic activities on the environment. Furthermore, there seems to be more of paper works in the majority of economic activities and transactions in the economy should be digitalized. Consequently, the government should make policies that will enhance a significant transition to the digitalization of the economy.

Second, the roles of green finance are observed to significantly drive a sustainable economy by reducing carbon emissions. It is thus suggested that government should implement policies that will redirect the focus of government capital expenditure toward green projects. This is particularly important for MENA economies since the concept of green finance is just evolving. Hence, promoting green projects will serve the objective of reducing unemployment rates and carbon emission surges.

The roles of multinational corporations prove to support the pollution halo hypothesis for the MENA economies as evident from the moderating effects of FDI on carbon emissions. To sustain moderating impacts of Multinational Corporations in the MENA region, regulatory policies must be firmly established to ensure foreign firms do not take countries in the region as dumping grounds for environmentally harmful products and services.

The moderating role of renewable energy can be sustained through the encouragement of investment in renewable energy projects and the withdrawal of subsidies on fossil fuels which can be used to promote renewable energy toward a carbon-neutral environment.

The present study provides relevant and substantial empirical evidence to advance the tripartite roles of eco-digitalization, green finance, and Multinational Corporations in the sustainable environment in the MENA region. However, the study did not cover the effects of the selected exogenous indicators on other environmental outcomes such as ecological footprint and PM2.5 air pollution. Furthermore, analyses on a large group of the region such as the Sub-Saharan African region and other intergovernmental organizations such as G7, G20, and E7 can equally be explored.