Exploring the Influence of Digital Transformation on Clean Energy Transition, Climate Change, and Economic Growth among Selected Oil-Export Countries through the Panel ARDL Approach

Abstract

Amid global imperatives to combat climate change and achieve sustainable economic development, the convergence of digital transformation and the transition to clean energy has emerged as a critical focal point for oil-exporting nations. This study comprehensively investigates the interplay of digital technology, clean energy transition, climate change, and economic growth among selected oil-exporting nations. Drawing upon a diverse set of economic and geographical contexts, this study uses panel data analysis of data from the World Bank’s Economic Indicators and the United Nations Development Program for the period from 2006 to 2020. The results show that digital technology reduces climate change by improving environmental quality, but internet and mobile access have insignificant and negative effects on environmental quality, respectively. Meanwhile, all technology variables negatively impact green energy and economic growth, while the Happy Planet Index and financial development positively impact the green energy transition. This study is important for regulators, producers, and consumers, as it provides a better understanding of the crucial role of digital transformation in sustainable development within oil-export countries. This study’s findings can be used to develop policy recommendations for a low-carbon economy, the promotion of digital transformation through green energy, and the management of climate change.

1. Introduction

Technological advancement is widely recognized as crucial for sustainable economic development. This has led to a surge in interest in innovation and its advancement among both researchers and economic agents, especially in the energy sector, where there is a strong connection between economic and environmental impacts [1,2,3,4]. Therefore, it is necessary to adopt an innovative technical approach in the energy sector to investigate whether technological transformation can help achieve a clean energy transition, reduce climate change, and promote economic growth.

Extensive research has been conducted on the impact of energy consumption on economic and environmental indicators [5,6,7,8,9]. A study by [9] found that a 1% increase in energy consumption harms the environment by 0.18%. Meanwhile, other studies have found that renewable energy use has a negative relationship with carbon emissions [10,11,12,13,14,15,16,17,18,19,20,21]. However, these studies did not consider the economic, social, or environmental dimensions of digital transformation when assessing its potential as a sustainable tool for an economy.

A study by [6] found that technological innovation led to increased energy efficiency and a reduction in energy consumption in industries. Research by [22] stated that environmental policy, clean energy, green innovation, and renewable energy have negatively contributed to seven emerging economies’ attainment of carbon neutrality. Another study, utilizing panel data analysis from 1995 to 2015, explored the dynamic connections between energy transitions, energy consumption, and sustainable economic growth in thirty-eight International Energy Agency countries [23]. However, existing studies often lack reliable and variable methodologies for measuring socioeconomic standards concerning maintaining physical, human, and natural resources of sufficient quality and quantity. As an indicator of environmental impact, we used the Happy Planet Index (HPI) in this study. Previous studies [10,11,12,13,14,15,16,17,18,19,20,21,22] have used carbon emissions to reflect environmental deterioration in different regions. The authors of [23] examined the link between the environmental sustainability index and socioeconomic indicators among selected countries in Asia, while [24] used an environmental degradation index focused solely on pollutants.

Thus, adding the HPI would further add to the uniqueness of this study. The HPI takes a radically different approach, measuring how well countries are doing in creating a good life for their people while sustainably using resources. The HPI is made up of three different indicators: ecological footprint, self-reported life satisfaction, and life expectancy. We anticipate this variable to exhibit a positive correlation with clean energy and a positive or negative correlation with economic growth. Because of digital transformation, the transition from nonrenewable energy to clean energy or environmentally friendly energy can occur [3].

In addition, there is limited empirical evidence on how digital transformation leads to a clean energy transition or how the energy transition affects economic growth and climate change in oil-producing countries. This research gap was also witnessed in a review of the literature by [22]. Therefore, this study investigates the impact of digital transformation on renewable energy, climate change, and economic growth in selected oil-exporting countries. Secondary data were collected from the World Bank and UNDP databases for this study from 2006 to 2020 at the aggregate level because of data availability.

This study makes a significant contribution to the literature on digital transformation by considering all economic, social, and environmental dimensions to ensure that digital transformation is an economically sustainable tool. Moreover, we use a comprehensive panel ARDL empirical investigation to study the intersections of digital transformation, clean energy transition, climate change mitigation, and economic growth in the context of oil-exporting economies. This is a novel and important area of research, as oil-exporting economies are facing unique challenges in transitioning to a clean energy future. Finally, we aim to offer valuable insights and evidence-based recommendations for policymakers, industry stakeholders, and researchers, facilitating informed decision making and the formulation of targeted strategies for sustainable development in the era of digital transformation and clean energy transitions. This is an important contribution, as it can help to ensure that digital transformation is used to promote sustainable development in oil-exporting economies.

In the following sections, a literature review and hypothesis development are outlined, followed by the methods and materials in Section 3. The empirical findings and discussions are presented in Section 4, and the conclusion and policy implications are presented in Section 5.

2. Literature Review and Hypothesis Development

The literature review is divided into two parts for easy comprehension: the first part discusses the relationships between digital transformation, clean energy consumption, and climate change, and the second part discusses the relationships between digital transformation, cleaner energy transitions, and economic growth.

2.1. Relationships between Digital Transformation, Cleaner Energy Transitions, and Climate Change

Global warming and extreme weather events are becoming increasingly prevalent. As a result, green energy sources are becoming increasingly important to the world’s energy portfolio. The global clean energy supply accounted for 12% in 2021 and is expected to increase by 31% and 70% in 2030 and 2050, respectively [25]. A study by [26] suggests that it is technically feasible to achieve a fully renewable energy system by 2050.

This has led to a surge in interest among both researchers and economic agents. There is a large volume of published studies describing the impact of renewable energy on the environment. Renewable energy sources, such as solar and wind power, produce little to no greenhouse gas emissions, making them a clean and sustainable alternative to fossil fuels. A study by [20] found that increasing renewable energy by 1% led to a decline in GHG emissions by 0.88% among EU countries and Ukraine between 2000 and 2016. Another study by [10] found that carbon emissions can be reduced with greater use of renewable energy and forested areas.

The use of renewable energy has a positive impact on the environment. This has been supported by [10] for Peru; [11] for Kazakhstan; [12] for BRICS; [13] for Indonesia, Malaysia, the Philippines, and Thailand; [14] for 25 African nations; [15] for Kenya; [16] for the European Union; [17,18] for the OECD; and [19] for Turkey. However, a study by [27] found dramatic differences between renewable energy use and carbon emissions in different countries. Many studies have tried to determine the causal relationship between carbon emissions and the factors influencing them.

A set of studies [28,29,30,31] stated that technological innovation is well established as the heart of the transition to a lower-carbon economy. According to [32], OECD countries may be able to mitigate environmental degradation problems by integrating structural transformation and renewable energy. However, studies by [30,33] suggest that heavy structural transformation can lead to environmental degradation, while [31] argues that excessive economic transformation can also have this effect. A study by [28] also found that developing countries are more likely to emit greenhouse gases because of the scale effects of greater economic transformation.

Despite the highest energy consumption being in developed countries, their energy innovations balance the system [34,35]. Consequently, implementing renewable energy can lead to more efficient environmental productivity for businesses by combining different types of environmental technology [36].

Therefore, there is no clear conclusion regarding how technology affects energy transformation and climate change. Thus, to investigate the potential of digital transformation for clean energy transition and climate change, this study attempts to develop the following hypotheses:

Hypothesis 1 (H1).

There is a positive association between digital technology and climate change.

Hypothesis 2 (H2).

There is a positive association between digital technology and clean energy transition.

2.2. Relationships between Digital Technology and Clean Energy Transitions in Economic Growth

The transition to a renewable energy system is one of the most important challenges facing humanity today. Digital technology for renewable energy is essential in mitigating climate change and achieving sustainable development. A growing body of literature is investigating the relationship between digital technology, renewable energy transitions, and economic growth. While the evidence is mixed, there is a growing consensus that renewable energy transitions can have a positive impact on economic growth, particularly in the long run. A study by [37] found a direct link between economic growth and energy consumption, as well as a 0.193% decrease in carbon emissions for every 1% increase in renewable energy consumption among selected developing Asian economies from 2000 to 2016. This suggests that economic growth and renewable energy are directly linked in both the short and long run.

Renewable energy consumption and economic growth have also been found to be positively correlated in several other studies. The authors of [38] found that a variety of states, including Germany, Japan, and OECD members, have a positive relationship with renewable energy consumption and economic growth. OECD countries experienced strong economic growth following the adoption of renewable energy during the 2000s [39]. Renewable energy consumption is positively associated with economic growth in Italian regions [40]. A study by [41] shows a direct link between renewable energy use and economic prosperity for 24 countries in the Middle East and North Africa (MENA) region. The findings of [42] suggest that renewable energy can be a major driver of economic growth in sub-Saharan Africa, a region that is heavily reliant on traditional energy sources, which are often expensive and polluting. By investing in renewable energy, sub-Saharan African countries can reduce their reliance on these harmful sources and boost their economies. This study used data from 51 sub-Saharan African countries for the period from 1980 to 2009.

However, not all studies have shown in a positive relationship between renewable energy consumption and economic growth. For example, a set of studies by [38,43,44] found that increases in renewable energy consumption hurt the economic growth of Turkey; 24 European countries; and India, Ukraine, and the US, respectively.

Moreover, a stream of studies by [45,46,47,48] stated that there is no causal relationship between energy consumption and economic growth for 27 European countries; Turkey; and G7 countries.

Using a threshold model encompassing 103 countries from 1995 to 2015, ref. [49] discovered that the influence of renewable energy consumption on economic growth is contingent upon the quantity of renewable energy utilized. This research showed that only when developing countries or non-OECD countries surpass a specific threshold of renewable energy consumption does the impact of such consumption on economic growth become positive and noteworthy. Conversely, in developing countries, a sub-threshold level of renewable energy consumption harms economic growth. While the consumption of renewable energy negatively affects economic growth in advanced nations, it exhibits a positive correlation with economic growth in OECD countries.

The findings of [41] indicate that both renewable energy consumption and financial development exert a moderate influence and can only partially account for variations in economic growth and carbon emissions. These results underscore the limited contributions of both financial development and renewable energy sectors toward fostering environmental quality improvements and facilitating economic growth. This analysis was conducted using data from 24 countries in the MENA region in the period from 1980 to 2015. However, the existing literature does not offer a definitive consensus on these issues.

Meantime, the studies by [50,51] showed that technological innovation has a significant influence on environmental performance that has enhanced sustainability in BRICS countries and Turkey, respectively. Similar results were arrived at in the study by [52], in that renewable technological innovation improves the carbon performance index. This study employed panel data analysis using 25 provinces in China from 2002 to 2015. The study conducted by [53] highlighted the significant roles of technical innovation and renewable energy dimensions in mitigating environmental threats. Additionally, this research found that crude oil price volatility has a notable impact on environmental deterioration, particularly among N11 countries. A parallel conclusion was drawn by [54], reinforcing the influence of technological innovation and renewable energy aspects on environmental dynamics. Furthermore, the investigation by [55] yielded similar findings regarding consumption-based carbon emissions. This study underscored the substantial contributions of technological innovation, energy productivity, and export activities in addressing consumption-based carbon emissions. However, it revealed a positive relationship between gross domestic production and consumption-based carbon emissions in G7 countries.

Concurrently, the advent of digital transformation has facilitated progress toward a cleaner energy transition, thereby positively impacting overall economic development. These divergent findings highlight the intricate nature of the relationship between digital transformation, renewable energy consumption, and economic growth, which appears to be contingent upon various factors, including the country’s developmental level, the specific type of renewable energy utilized, and the prevailing policy framework. Further research is guaranteed to gain a more comprehensive understanding of this complex relationship. Therefore, this study seeks to propose the following hypothesis:

Hypothesis 3 (H3).

There is a positive association between digital technology, clean energy transition, and economic growth.

3. Methods and Materials

3.1. Theoretical Background, Model Specification, and Variable Selections

Theoretically, economic growth is correlated with carbon emissions. Assuming a market-clearing condition where carbon emissions are closely linked to economic growth within the framework of the standard Marshallian function at time t [56], this model was subsequently enhanced by [8] to incorporate both renewable and nonrenewable energy use. The illustration of this enhancement is depicted in the base model.

CO₂ = f(GDP, EU)

where carbon emission at time t is represented by CO_2t, aggregate gross domestic product is represented by GDP_t, and EU_it represents renewable and nonrenewable energy use.

The current study aims to investigate the influence of digital transformation on renewable energy use, climate change, and economic growth in OPEC countries. Therefore, based on the Marshallian function, our models were developed by incorporating technology as a factor impacting CO₂ emissions in these economies. To represent the environmental impact in our models, we utilize the HPI in place of CO_2t.

Three models were developed to explore the relationship between the dependent variable and independent variables. The first model considers HPI the dependent variable, with PGDP, REN, MOB, INT, ESY, and FD as independent variables. The second model takes REN as the dependent variable, with PGDP, HPI, MOB, INT, ESY, EMP, and FD as independent variables. The third model takes PGDP as the dependent variable, with HPI, REN, MOB, INT, ESY, and EMP as independent variables. These relationships are depicted in the following three models:

HPIit = β0 + β1PGDPit + β2RENit + β3MOBit + β4INTit + β5ESYit + β6FDit + εit	(Model 1)
RENit = β0 + β1PGDPit + β2HPIit + β3MOBit + β4INTit + β5ESYit + β6EMPit + β7FDit + εit	(Model 2)
PGDPit = β0 + β1HPIit + β2RENit + β3MOBit + β4INTit + β5ESYit + β6EMPit + εit	(Model 3)

where HPI is the Happy Planet Index, PGDP is GDP per capita (constant 2015 USD), REN is renewable energy consumption, MOB is mobile cellular subscriptions, INT is individuals using the internet, EYS is expected years of schooling, EMP is labor force participation rate, and FD is financial development indicator.

The previous literature has predominantly employed carbon emissions to assess environmental degradation in various regions. A study conducted by [23] examined the connection between the environmental sustainability index and socioeconomic indicators across selected Asian countries. The authors of [24] emphasized that the environmental degradation index primarily focuses on pollutant levels.

Existing research lacks dependable and versatile methodologies to gauge the socioeconomic well-being of people concerning the maintenance of the quality and quantity of physical, human, and natural resources. Hence, the inclusion of the HPI can enhance the distinctiveness of this study. The HPI offers a markedly different perspective on defining progress. It is computed using the formula “HPI = Life Satisfaction × Life Expectancy/Ecological Footprint” ([57]).

The HPI, introduced by the New Economics Foundation in 2006, is most effectively understood as a gauge of environmental efficiency in promoting well-being within a country. This index enables citizens to lead fulfilling lives without compromising the opportunities of future generations. It comprises three key indicators: ecological footprint, self-reported life satisfaction, and life expectancy. Human well-being is conceptualized as a Happy Life Expectancy. The assessment of human well-being involves an evaluation of the extraction of or imposition upon nature, achieved through the per capita ecological footprint. This measurement involves weighing the carbon footprint and estimating the natural resources needed to sustain a country’s lifestyle. The indicators used for measurement include CO₂ emissions per capita, GDP purchasing power parity (PPP) per capita, exports and imports per capita, population density, and the land required for renewable resources like food and wood products [58].

The HPI represents how efficiently countries convert the Earth’s finite resources into the well-being experienced by their citizens. We anticipate the HPI to exhibit a positive correlation with clean energy and digital technology and a positive or negative correlation with economic growth. Because of digital transformation, the transition from nonrenewable energy to clean energy or environmentally friendly energy can occur [3]. This transition contributes to achieving several vital sustainable development goals, encompassing social and economic development, energy accessibility, energy security, and reducing environmental and health impacts [59].

In the pursuit of sustainable development, [60] underscores the significance of maximizing the utilization of natural resources, prudent investments, technological advancements, and the establishment of institutional frameworks. To measure technological development within a country, in alignment with [60], we employed variables such as “mobile cellular subscriptions” and “individual internet access”, alongside average schooling years, as indicators of digital technology development. These variables reflect social development conducive to technology absorption, selected based on the literature [3,4,61,62]. Tertiary education enrollment denotes social development within the economy (absorption of digital transformation), which serves as the fundamental driver of technological development [62]. Gross domestic product per capita represents the economic status of the country [37,41]. “Domestic credit to the private sector by banks” serves as a financial development indicator [41]. The total labor force (EMP) is an indicator of “human capital”, which refers to the skills, knowledge, experience, and creativity possessed by individuals in a workforce and stands as a crucial determinant of any organization’s success [3,4,8,61].

3.2. Data Definition and Samples

The targeted sample data consists of nine OPEC countries, including Congo, Rep.; Ecuador; Iraq; Iran Islamic Rep.; Kuwait; Nigeria; Qatar; Saudi Arabia; and the United Arab Emirates, encompassing the period from 2006 to 2020. All data were sourced from the World Bank’s “World Development Indicators” and the United Nations Development Program (UNDP) (

Table 1. Data, definitions, and sources.

Variables	Explanation	Indicator	Source
Renewable energy, REN	“Renewable energy consumption is the share of renewable energy in total final energy consumption”. (% of total final energy consumption)	Energy transitions	World Bank database
Happy Planet Index, HPI	“A measure of sustainable well-being. It was developed by the New Economics Foundation in 2006 and is calculated based on three factors: ecological footprint, self-reported life satisfaction, and life expectancy”.	Environmental indicator	UNCTAD
Gross domestic product, GDP	“Total monetary or market value of all the finished goods and services produced within a country’s borders in a specific period”.	Economic status of the country	World Bank database
Mobile cellular subscription, MOB	“Mobile cellular telephone subscriptions are subscriptions to a public mobile telephone service that provide access to the PSTN using cellular technology”.	Technological development	World Bank database
Individual internet access, INT	Individuals using the internet (% of the population)	Technological development	World Bank database
Education level, ESY	EYS is expected years of schooling.	Technological development (absorption of digital transformation)	UNDP
EMP	“Employment-to-population ratio is the proportion of a country’s population that is employed”.	Human capital development	World Bank database
Domestic credit to the private sector by banks	“Domestic credit to private sector by banks (% of the GDP)”.	Financial development indicator	World Bank database

). The data for the Happy Planet Index were compiled from the United Nations Conference on Trade and Development (UNCTAD) [57].

3.3. Empirical Analysis Using Panel ARDL (PMG)

Using panel data from nine OPEC countries in the period from 2006 to 2021, this study employs the PMG approach. One advantage of this methodology is its ability to recognize both the long- and short-run relationships between the targeted variables (PGDP, HPI, REN, ENT, MOB, EYS, EMP, and FD). Additionally, the PMG ARDL methodology accounts for cross-sectional dependency, enabling the analysis and comparison of sample countries. The selection of the sample and the duration is predicated on data availability. Preceding the implementation of the panel ARDL, the study conducted several diagnostic tests, including correlation analysis, cross-section dependence testing, and unit root testing. To begin, the study assessed cross-sectional dependence in the data and models, employing various tests, including [63] CD and scaled LM, Breusch and Pagan’s test, and bias-corrected scaled LM tests [64].

3.3.1. Descriptive Analysis and Correlation Analysis

Table 2. Descriptive statistics.

ITEMS	PGDP	REN	HPI	INT	MOB	ESY	EMP	FD
Mean	34,389.94	18.3113	38.6754	45.7996	109.693	12.7756	1.9244	40.6
Median	14,903.52	0.9000	38.9069	41.9400	99.0730	12.8270	1.8300	39.9
Maxi	99,147.29	88.680	60.2019	100.000	212.6390	16.1358	6.0100	138.
Mini	3497.565	0.0000	23.9353	0.9300	21.80617	8.35617	0.2700	2.01
Std.d	31,125.75	30.596	8.47838	32.7239	45.3138	2.03082	1.4439	29.1
Jarque–Bera	15.4531 ***	41.771 ***	4.8355 *	10.8023 ***	4.9952 *	6.93859 **	9.9326 **	10.3 **
Observe	135	135	135	135	135	135	135	135

Source: Authors’ calculations. Note: ***, **, and * denote 1%, 5%, and 10% significance levels, respectively.

displays the results of a descriptive statistical analysis of the targeted variables, demonstrating substantial standard deviations that allow for an investigation of the data series variation. Additionally,

Table 3. Correlation matrix.

ITEMS	PGDP	REN	HPI	INT	MOB	ESY	EMP	FD
PGDP	1.0000
REN	−0.557769	1.0000
HPI	−0.686690	0.0266	1.0000
INT	0.701167	−0.544136	−0.3435	1.0000
MOB	0.575923	−0.463051	−0.2590	0.7833	1.0000
ESY	0.318676	0.696706	0.2689	0.5976	0.559575	1.0000
EMP	−0.299977	−0.420877	0.6111	−0.1049	−0.078664	0.4443	1.0000
FD	0.6873	−0.5643	−0.4313	0.8195	0.6613	0.5934	0.0003	1.0000

Source: Authors’ calculations.

depicts the outcomes of the correlation analysis, revealing that the correlation coefficients among the variables are all below 0.8, suggesting a lack of linear correlation between the regressors. However, a correlation was observed between FD and INT. Given the significance of these variables, they were retained in the models.

3.3.2. Cross-Sectional Dependence Tests

To examine whether the models exhibit cross-section dependence, we conducted the BP LM test, the Pesaran scaled LM test, and the bias-corrected scaled LM test. Across all models, the null hypothesis suggesting the absence of cross-sectional dependence was confirmed at a 1% significance level, as depicted in

Table 4. Cross-sectional dependence tests.

Test	Model 1 HPI Is the Dep.		Model 2 REN Is the Dep.		Model 3 GDP Is the Dep.
	Statistic	Prob.	Statistic	Prob.	Statistic	Prob.
Breusch–Pagan LM	138.7712	0.0000	108..4208	0.0000	200.4823	0.0000
Pesaran scaled LM	12.1117	0.0000	17.0202	0.0000	19.38443	0.0000
Pesaran CD	−0.7222	0.4702	0.1951	0.8453	4.213587	0.0000

Source: Authors’ calculations.

.

3.3.3. Cross-Sectional Dependence Unit Root Tests

As cross-section dependence was observed, this study utilized second-generation unit root tests, namely, the cross-sectional augmented Dickey–Fuller (CADF) and cross-sectional Pesaran–Shin (CIPS) unit root tests [63].

Table 5. Cross-sectional dependence unit root tests.

	CIPS		CADF
	Level	First Difference	Level	First Difference
PGDP	−2.141	−2.812 ***	−1.730	−2.050
HPI	−2.415 **	−4.165 ***	−1.884	−2.680 ***
REN	−1.562	−3.563 ***	−1.637	−2.477 **
MOB	−1.426	−3.262 ***	−1.646	−2.224 *
INT	−2.131	−2.931 ***	−2.211	−2.335 *
ESY	−1.291	−2.257 *	−1.343	−1.049
EMP	−2.087	−3.653 ***	−1.613	−1.855
FD	0.575	−3.667 ***	−0.702	−2.427 **

Source: Authors’ calculations. Note: ***, **, and * denote 1%, 5%, and 10% significant levels, respectively.

demonstrates that all variables are stationary at I (1) at significance levels of 1%, 5%, and 10%, at least for one of the tests.

3.3.4. Panel Autoregressive Distributed Lag Analysis (PMG) Test

The empirical findings of the panel autoregressive distributed lag analysis (PMG) are presented in

Table 6. Panel ARDA (PMG) tests.

	Model 1		Model 2		Model 3
	Coefficient	t-Statistic	Coefficient	t-Statistic	Coefficient	t-Statistic
Long-Run Equation
PGDP	−0.0002 *	1.9056	−0.0002 ***	−2.5648	-	-
HPI	-	-	0.4574 ***	3.9218	−8.539562	−1.054637
REN	−0.5176 *	−1.9417	-	-	−257.8968 ***	−14.00774
INT	0.0382	1.5459	−0.05384 ***	−2.2627	−36.04164 ***	−3.486599
MOB	−0.0843 ***	−4.0180	−0.0138	−0.9038	−21.98105 ***	−4.398547
ESY	2.1695 ***	5.4442	−2.5228 ***	−3.5687	−1757.590 ***	−12.25489
EMP	-	-	−8.0734 ***	−2.7949	597.8269	1.431022
FD	0.1407 **	2.5133	0.2800 **	3.5753	-	-
Short-Run Equation
COINTEQ01	−0.5928 ***	−5.4134	−0.2124 *	−1.7097	−0.405078 *	−1.964271
D(PGDP)	−0.0008	−0.8794	−0.0002	−0.9730	-	-
D(HPI)	-	-	−0.1492 *	−1.8874	285.4641	1.179213
D(REN)	−22.8945	−0.6858	-	-	2383.576	0.174535
D(INT)	−0.3766	−1.0747	−0.3809	−1.1554	9.938130	0.131026
D(MOB)	0.0677	1.5310	0.0416	−0.9560	−28.71165	−0.691790
D(ESY)	−0.5385	−0.2090	−1.1966	−1.0583	2619.082 *	1.874757
D(EMP)	-	-	−3.7572	−0.5076	11,573.97	0.870498
D(FD)	−0.2316 **	−2.3055	−0.1269 **	−0.9802	-	-
C	15.2124 *	1.7904	17.7873	1.6480	18,159.42	1.109342
@TREND	-	-	-	-	4.489162	0.016299

Source: Authors’ calculations. Note: ***, **, and * denote 1%, 5%, and 10% significant levels, respectively.

. These results show that most of the selected explanatory variables have significant long-run effects on the dependent variables. Also, Table 6 shows that the error correction terms are statistically significant at 1% for model 1 and 10% for models 2 and 3, indicating that a causal relationship exists in at least one direction. The error correction term coefficient of −0.5928 for the first model hints that 59.28% of the disequilibrium from the previous year is corrected back to the long-run equilibrium in the current year. The error correction term coefficients of −0.2124 and −0.405078 for the second and third models, respectively, hint that the adjustment speed is lower for the second and third models.

3.3.5. Dumitrescu–Hurlin Panel Causality Test

Following the PMG models, the final step involved the application of the Dumitrescu–Hurlin panel causality test, a robust method aimed at detecting potential causal relationships between selected variables. The results of the Dumitrescu–Hurlin panel causality test is presented in

Table 7. Dumitrescu–Hurlin panel causality test (used as a robust method).

	W. Stat	Zbar. Stat	Prob
HPI does not homogeneously cause PGDP PGDP does not homogeneously cause HPI	0.79792 2.36782	−0.61614 1.66354	0.5378 0.0962
REN does not homogeneously cause PGDP PGDP does not homogeneously cause REN	2.70596 0.90201	2.15459 −0.46498	0.0312 0.6419
INT does not homogeneously cause PGDP PGDP does not homogeneously cause INT	3.44772 1.12728	3.23170 −0.13787	0.0012 0.8903
MOB does not homogeneously cause PGDP PGDP does not homogeneously cause MOB	2.83841 4.15911	2.34691 4.26475	0.0189 0.0000
ESY does not homogeneously cause PGDP PGDP does not homogeneously cause ESY	2.89547 3.81329	2.42978 3.76257	0.0151 0.0002
EMP does not homogeneously cause PGDP PGDP does not homogeneously cause EMPOR	1.97987 5.11910	1.10020 5.65877	0.2712 0.0000
REN does not homogeneously cause HPI HPI does not homogeneously cause REN	1.32342 2.08474	0.14695 1.25248	0.8831 0.2104	-
INT does not homogeneously cause HPI HPI does not homogeneously cause INT	2.70597 0.90202	2.15459 −0.46498	0.0312 0.6419
MOB does not homogeneously cause HPI HPI does not homogeneously cause MOB	2.38899 0.71467	1.69429 −0.73702	0.0902 0.4611
ESY does not homogeneously cause HPI HPI does not homogeneously cause ESY	2.77198 2.28364	2.25044 1.54132	0.0244 0.1232
INT does not homogeneously cause REN REN does not homogeneously cause INT	4.22324 1.37888	4.35787 0.22748	0.0000 0.8200
MOB does not homogeneously cause REN REN does not homogeneously cause MOB	3.83566 2.25634	3.79505 1.50167	0.0001 0.1332
ESY does not homogeneously cause REN REN does not homogeneously cause ESY	5.19781 4.11930	5.77308 4.20693	0.0000 0.0000
MOB does not homogeneously cause INT INT does not homogeneously cause MOB	0.85604 3.81508	−0.53174 3.76517	0.5949 0.0002
ESY does not homogeneously cause INT INT does not homogeneously cause ESY	6.10743 3.76562	7.09395 3.69335	0.0000 0.0002
ESY does not homogeneously cause MOB MOB does not homogeneously cause ESY	6.29806 3.93053	7.37077 3.93281	0.0000 0.0000
FD does not homogeneously cause GDP GDP does not homogeneously cause FD	3.1154 2.8295	2.7492 2.3340	0.0006 0.0196
FD does not homogeneously cause HPI HPI does not homogeneously cause FD	1.1704 0.8853	−0.0752 −0.4893	0.9400 0.6246	-
FD does not homogeneously cause REN REN does not homogeneously cause FD	3.8544 2.1469	3.8223 1.3427	0.0001 0.1794
FD does not homogeneously cause ESY ESY does not homogeneously cause FD	2.1212 2.2614	1.3054 1.5090	0.1918 0.1313	-
FD does not homogeneously cause EMP EMP does not homogeneously cause FD	2.7378 1.7325	2.2009 0.7409	0.0277 0.4587

Source: Authors’ calculations ( bidirectional causalities, and unidirectional causalities’, - no causalities).

, illustrating six bidirectional causalities between PGDP and MOB; PGDP and ESY; ESY and REN; ESY and INT; ESY and MOB; and FD and PGDP. Moreover, numerous unidirectional causalities were observed from the independent variables to the dependent variables, specifically, from PGDP to HPI and EMP; from REN and INT to PGDP; from INT, MOB, and ESY to HPI; from INT, MOB, and FD to REN; from INT to MOB; and from FD to EMP. Additionally, the results indicate no causal relationship between REN and HPI; FD and HPI; or FD and ESY. The detected causality between the selected variables confirms the robustness of the panel autoregressive distributed lag analysis (PMG) results.

The findings from the cross-section short-run coefficient test reported in

Table 8. Cross-section short-run coefficient.

	Model (1) COINTEQ01	Model (2) COINTEQ01	Model (3) COINTEQ01
Congo, Rep.	−0.5855 *** (0.0003)	−0.9108 *** (0.0000)	−0.1110 *** (0.0029)
Ecuador	−0.7482 ** (0.0094)	0.0971 *** (0.0003)	−0.2969 *** (0.0001)
Iraq	−0.4900 *** (0.0000)	−0.3101 *** (0.0007)	−1.0067 *** (0.0002)
Iran, Islamic Rep.	−0.5776 *** (0.0002)	−0.0489 *** (0.0000)	−0.9918 *** (0.0000)
Kuwait	−0.0661 *** (0.0031)	0.0008 *** (0.0000)	−0.7991 *** (0.0002)
Nigeria	−1.1544 *** (0.0005)	−0.7610 ** (0.0050)	−0.6526 *** (0.0000)
Qatar	−0.2241 *** (0.0000)	0.0144 *** (0.0000)	0.8086 *** (0.0001)
Saudi Arabia	−0.5811 *** (0.0003)	−0.0005 *** (0.0000)	−0.6817 *** (0.0003)
United Arab Emirates	−0.9084 *** (0.0001)	0.0076 *** (0.0000)	0.5082 *** (0.0000)

Note: ***, and **, and denote 1%, and 5%, significant levels, respectively. Source: Authors’ calculations.

show that all COINTEQ 01 coefficients are negative and significant at 1% or 5% for all countries in model 1; Ecuador, Kuwait, Qatar, and the United Arab Emirates for model 2; and Qatar and the United Arab Emirates for model 3. The COINTEQ 01 coefficients appear with a positive sign. This indicates that a long-run relationship exists between the dependent and explanatory variables. Moreover, the model 1 results show that the convergence rate to the long-run equilibrium is highest in Nigeria (115.44%), followed by the United Arab Emirates (90.84%), and it is lowest in Kuwait (6.61%). The model 2 results show that the convergence rate to the long-run equilibrium is highest in Congo, Rep. (91.08%), followed by Nigeria (76.10%), and it is lowest in Saudi Arabia (0.05%). The model 3 results show that the convergence rate to the long-run equilibrium is highest in Iraq (100.67%), followed by Iran (99.18%), and it is lowest in Congo, Rep. (11.10%).

4. Empirical Findings and Discussion

These findings reveal the significant impact of one of the digital development variables, namely, education, on climate change in model 1. Consequently, the previously posited Hypothesis 1, “H1: There is a positive association between digital technology and climate change”, is affirmed. It can be seen that an increase in one expected year of schooling would improve climate change by 2.169 at a 1% significant level. This signifies that the advancement of digital technology contributes positively to addressing climate change concerns within the OPEC context. Specifically, an escalation in the societal education level plays a pivotal role in reducing climate change or bolstering the HPI, aligning with theoretical expectations. This finding resonates with the perspective highlighted in the work of [53,60], emphasizing the significance of technological advancements in fostering a sustainable environment.

Contrarily, MOB harms climate change. An increase in MOB by 1% would negatively influence climate change by 0.084 at a 1% significant level. These findings align with the research by [65], which projects that digital infrastructure, including data centers, consumes substantial energy and emits a considerable volume of greenhouse gases. Such adverse impacts on the environment could potentially yield negative repercussions for the economy. According to [65], data centers represent one of the leading global energy consumers, contributing to approximately 2% of greenhouse gas emissions worldwide. Moreover, as noted by previous studies [4,61], this could be attributed to the presence of a digital divide within the country. Consequently, the substantial disparities in quality of life and developmental discrepancies during the digital revolution phase pose a significant challenge for OPEC economies.

Furthermore, the outcomes of model 2 indicate the notable negative effect of digital technology on clean energy transition in both the long run and short run. Consequently, Hypothesis 2, “H2: There is a positive association between digital technology and clean energy transition”, is refuted. It can be seen that a 1% increase in internet access and education would negatively impact clean energy transition by 0.06 and 2.523 at a 1% significant level, respectively. This suggests that the augmentation of digital technology does not foster a clean energy transition within OPEC countries.

As highlighted by [65,66,67], digital technologies can be effectively employed to curtail energy losses, optimize energy consumption, and enhance grid management. However, given their heavy reliance on fossil fuels and the substantial revenues derived from them, OPEC member states may exhibit a reduced inclination toward investing in such initiatives. Moreover, inadequate technological infrastructure, low digital literacy rates, and insufficient research and development investments pose obstacles to the adoption of digital technologies in numerous OPEC nations. These factors can impede energy efficiency efforts and the transition to cleaner energy sources [68,69,70].

It is imperative for OPEC countries to factor in the geopolitical landscape when formulating energy-related decisions. Clean energy transitions might take a backseat to energy security and geopolitical interests in these nations. Investments in digital technologies supporting fossil fuel extraction and distribution may receive precedence over endeavors aimed at promoting clean energy [67,68,69]. Forecasts by [69] anticipate a surge in energy demand, prompting even fossil fuel giants to embark on the path toward green energy. Therefore, the aforementioned factors could plausibly account for the observed negative and significant impact of digital technology on clean energy transformation within OPEC countries. In addition, the HPI has a positive impact on clean energy transitions as expected. Increasing clean energy use accelerates human and ecological systems on the planet. This finding aligns with the findings of the study by [10]. They found renewable energy leads to environmental quality in Malaysia. Nonetheless, according to the research in [22], renewable energy and green innovations have a notable and adverse effect on carbon emissions in E7 nations, both in the short term and over an extended period.

Moreover, the financial development indicator exhibits a positive inclination toward climate change and clean energy transition in models 1 and 2. This can be expounded as a one percent increase in financial development leading to 0.14 and 0.28 increases in climate change and renewable energy, respectively. Encouragingly, the allocation of funds and investments in clean energy by both public and private financial institutions within OPEC economies could potentially foster advancements in their clean energy transitions. This finding contradicts the observations of [41], suggesting that both financial development and renewable energy sectors were relatively sluggish in contributing to environmental quality improvements and economic growth within the MENA region during the period from 1980 to 2015. Our findings underscore a recent surge in financial support directed toward the clean energy transition process within OPEC economies.

Furthermore, the outcomes of model 2 illustrate a negative relationship between labor, technology variables, and the clean energy transition. These findings diverge from the neoclassical theory of economics, which posits that labor and technological advancements will boost an economy’s output [71].

Finally, the findings demonstrate that digital technology exerts a significant negative impact on the gross domestic product in model 3. Consequently, the previously posited Hypothesis 3, “H3: There is a positive association between digital technology, clean energy transformation, and economic growth”, is rejected. This implies that the proliferation of digital technology and green energy transition does not foster economic growth within OPEC countries. An increase in digital technology variables—internet access, MOB, education, and energy transition—would decrease economic growth by 36.04, 21.98, 1757.59, and 257.9 in OPEC nations, respectively, at a 1% significant level. One possible explanation could be attributed to the job displacement effects associated with digital technology. As machines and software become increasingly proficient, they can automate tasks previously carried out by humans. This could exacerbate inequality within these economies. Notably, inequality has been on the rise across MENA countries in recent years owing to the ascendancy of digital technologies [67]. Additionally, a study conducted by [70] revealed that automation could potentially lead to a reduction of up to 1.9% in US economic growth by 2030. These findings diverge from neoclassical theory, which posits that technological advancements should bolster an economy’s output [71]. This finding is in contrast with the finding of [23]. There, it is stated that the impact of energy transitions on economic growth becomes noteworthy primarily in the long term, while economic sustainability influences economic growth in both the short term and the long term among thirty-eight International Energy Agency countries.

Conversely, an increase in clean energy transition appears to correlate with declines in climate change and economic growth within OPEC economies by 0.5176 and 257.8968, as illustrated in models 1 and 3, respectively. This study implies that OPEC countries have yet to make significant strides in this domain, potentially owing to their heavy reliance on fossil fuels as a cornerstone of economic growth. This finding resonates with prior research conducted by [38,43,44], which found that increased renewable energy consumption adversely affects economic growth in various regions, such as Turkey, 24 European countries, India, Ukraine, and the US. One study [22] stated that a negative relationship between renewable energy, green innovations, environmental tax, and technological innovations in carbon emissions is apparent among China, Turkey, India, Russia, Brazil, Indonesia, and Mexico, and both in short- and long-run clean energy transformation, economic growth is apparent. Furthermore, ref. [49] observed that the impact of renewable energy consumption on economic growth varies depending on the quantity of renewable energy consumed. Notably, developing countries and non-OECD nations can experience significant and positive effects on economic growth only when they surpass a specific threshold of renewable energy consumption. In contrast, in developing countries, renewable energy consumption harms economic growth if it falls below a certain threshold level. Therefore, it can be concluded that OPEC countries have yet to surpass the threshold and must take proactive measures to achieve it.

Furthermore, a 1% increase in economic growth is shown to have an unfavorable impact on climate change and energy transition by 0.0002% in the first and second models within OPEC. Given the heavy reliance of OPEC nations on fossil fuels for energy consumption, these activities yield higher emissions, contributing to adverse environmental repercussions. These findings find support in various existing studies, including [10,11,12,13,14,15,16,17,18,19,20,21].

5. Conclusions and Policy Implications

With global warming and extreme weather fluctuations becoming increasingly prevalent [72], the integration of green energy sources into the global energy portfolio has gained significant importance. In this context, OPEC nations must establish clear policies prioritizing green energy transitions, thereby facilitating the advancement of zero carbon emissions by 2050. This strategic approach would not only contribute to effective climate change management but also promote export diversification within the OPEC community.

Expanding on our research focus, our study delves into the impact of digital technology on green energy transition, climate change, and the economic growth of OPEC economies using a comprehensive panel data approach. Our findings suggest that, while digital technology can mitigate climate change by enhancing environmental quality, aspects such as internet and mobile access may yield insignificant or even negative effects on environmental quality, primarily because of the substantial energy consumption and emission of pollutants associated with technology infrastructure. Furthermore, our analysis indicates the negative influence of all technology variables on clean energy transformation and economic growth while also revealing a positive relationship between climate change and financial development and green energy transition. Although digital technology may serve as a driving force in global clean energy transitions, OPEC nations should exercise caution owing to unique challenges and considerations, including economic dependence on fossil fuels, technological limitations, and geopolitical factors influencing their energy policies. Hence, the effective adoption of digital technologies in the realm of OPEC countries necessitates tailored policies and international cooperation to address these complex challenges adequately.

Our findings carry significant policy implications. Firstly, digital technologies can contribute to energy conservation in buildings, homes, and industrial facilities, leading to substantial energy savings. Secondly, the implementation of a well-developed technological infrastructure is crucial for enhancing energy efficiency and transitioning to cleaner energy sources, wherein a high-speed and reliable internet connection becomes essential for the deployment of smart grids and smart meters. Notably, the scarcity of necessary infrastructure can impede the widespread deployment of these solutions in OPEC countries. Thirdly, fostering a skilled workforce capable of implementing digital solutions for energy efficiency improvements and the adoption of cleaner energy sources remains crucial. However, low levels of digital literacy may pose challenges in training workers within OPEC countries for the effective utilization of these solutions. Additionally, the development and deployment of energy efficiency and clean energy solutions might necessitate the involvement of foreign entities, leaving OPEC countries vulnerable to price fluctuations and supply disruptions.

Our findings can provide valuable insights for regulators, producers, and consumers alike. With clean energy companies experiencing a surge in acquisitions, there is a growing global shift toward clean energy, projected to account for 95% of electricity generation by 2050 [73]. Amid this transition, OPEC must prioritize the early implementation of export diversification strategies for long-term development, even as the demand for fossil fuels remains high. Export diversification aids countries in diminishing their reliance on fossil fuel exports, thereby fostering greater economic resilience against external shocks. Simultaneously, clean energy transition can open new export avenues for countries in renewable energy, energy efficiency technologies, and related sectors [74,75]. Governments can significantly facilitate both export diversification and clean energy transition by investing in education and skills training, promoting research and development, offering financial incentives to encourage exports, reducing trade barriers [74], investing in renewable energy infrastructure, and setting ambitious targets for the deployment of renewable energy and improvements in energy efficiency. The United Arab Emirates has emerged as a significant exporter of tourism, financial services, and manufacturing products, while Saudi Arabia is actively diversifying its economy, with a particular emphasis on sectors such as tourism, entertainment, and logistics, in alignment with their essential export diversification efforts. This serves as evidence that major fossil fuel producers are initiating endeavors toward achieving export diversification.

However, oil-dependent countries still grapple with various challenges in achieving export diversification, including the high costs associated with developing new export industries, trade barriers, labor skill shortages in soft skills, and the necessity of infrastructure investment and research and development.

In addition, OPEC nations find themselves at a pivotal juncture. To ensure a prosperous future in the rapidly changing energy landscape, officials are encouraged to adopt comprehensive strategies. This includes openly acknowledging and addressing the financial implications of carbon emissions and exploring potential taxation models. Moreover, continuous investment and support for research and development efforts focused on clean energy technologies are essential. Creating conducive policies and market incentives to facilitate the commercialization of low-carbon-emission technologies is crucial. Gradually phasing out subsidies for nonrenewable energy sources to encourage a transition toward cleaner alternatives is recommended.

Collaborating on technology transfer initiatives is vital, which will foster knowledge sharing and accelerate the adoption of sustainable solutions in developing nations. Additionally, establishing a forward-thinking green trade policy that promotes sustainable development, fosters partnerships, and creates new economic opportunities is essential. Embracing these proactive measures will not only help OPEC nations navigate the transition to a low-carbon future but also position them as leaders in the development and deployment of cutting-edge clean energy technologies. This bold and responsible approach will secure their long-term prosperity and contribute to a healthier planet for all.

Our study assumes significance by offering a comprehensive understanding of the impact of digital technology on climate change, green energy, and economic growth within the OPEC context. By providing empirical insights into the intricate relationships between digital transformation, the adoption of clean energy, and responses to climate change, our research contributes to the existing literature on sustainable development in the era of technological advancement and energy transition. Furthermore, our findings emphasize the need for policymakers to address the challenges identified in our study effectively.

However, certain limitations in our study must be acknowledged. While we concentrate on assessing the impact of digital technology on climate change, clean energy transformation, and economic growth, further research is necessary to evaluate the feasibility and potential impact of digital technology in driving clean energy transitions within the specific context of OPEC nations. Moreover, researchers could consider using the same model, taking logarithms with different sample distributions and different groupings for future research. Moreover, exploring the disparities in digital technology’s role in facilitating clean energy transitions among various OPEC nations could be a promising avenue for future investigation. Additionally, a comparative analysis can be performed using cross-sectional data by grouping countries in terms of regional or income distribution.