Introduction

The rising concerns about the effects of global warming, driven by increasing greenhouse gas (GHG) emissions, are shared by environmental organizations and international leaders alike. Notably, nearly all GHG emissions are attributable to energy consumption. The link between emissions and economic production has made carbon mitigation a global challenge, despite the widespread acknowledgment of the ecological damage caused by CO2. The reduction in energy consumption and emissions during the Covid-19 pandemic was temporary, as the economic instability of 2021 resulted in a significant 6% increase in CO2 emissions, driven by a 5.8% growth in economic activity and the corresponding rise in power consumption1. Over the past few decades, global efforts to reduce emissions have increased. In this regard, the Paris Agreement set a target to limit temperature rise to 1.6 degrees Celsius, which could have severe negative effects on the planet. The COP-26 environmental accord, however, revised this target to limit the global temperature increase to 1.5 degrees, with a goal of achieving net-zero emissions by 20502.

A study exploring the impact of economic globalization, coal energy, and biomass energy on India’s ecological footprint, while controlling for economic development, suggests potential solutions to these challenges. The results indicate that economic globalization and biomass energy both contribute to better ecological quality. In contrast, India’s expanding economy and reliance on coal power have led to a decline in environmental quality3. The relationship between CO2 emissions and economic growth (EG) underscores the need to identify strategies for promoting sustainable growth, as zero emissions cannot be achieved solely through direct carbon reduction policies. In this context, the idea of fostering modern electronic economies has the potential to promote carbon neutrality by enhancing green digital economies (DE), which are a result of contemporary structural changes driven by advanced network functions and ICT (information and communication technology). These changes have redefined economic mechanisms, increasing production efficiency4. Digitalization can help countries achieve multiple Sustainable Development Goals (SDGs), including SDGs 11, 9, and 13. By enhancing competitiveness, innovation, and effectiveness, digitalization based on contemporary ICT infrastructures supports SDG-9 and reduces environmental pollution5. Moreover, digitalization has encouraged the use of new digital goods and services, while reducing the reliance on traditional, material-intensive goods6. Dematerialization, which leads to resource conservation, lowers emissions and improves environmental quality, thus helping economies achieve SDG-13.

A study by7 suggests that the introduction of digital applications has revolutionized the transportation sector, improving energy efficiency, traffic management, and facilitating travel. These advantages of digitalization can ease traffic congestion, reduce fossil fuel consumption, and lower harmful emissions, making it possible for countries to achieve both SDG-11 and SDG-13. However, digital transformation requires energy-intensive, modern machinery that produces hazardous waste8. Additionally, DE encourages industrial growth and economic opportunities, which can increase pollution, waste, and energy consumption9. In conclusion, DE can either increase or reduce emissions, but it also reshapes economic operations that influence overall economic growth.

Therefore, it is crucial to consider both the direct effects of DE on emissions and its potential indirect effects through GDP growth. Earlier studies on DE and emissions have largely overlooked the varied impacts of DE on emissions via economic expansion. SDG and carbon neutrality goals also require financial resources to support environmental improvements. The growth of these markets is necessary for a nation to expand financially and for businesses to thrive. Financial expansion (FINE) can increase the availability of financing for projects that integrate energy innovation, alternative energy, and modern technologies, which can reduce energy consumption and lower CO2 emissions10.

However, FINE can also stimulate corporate financing, promoting business expansion that increases CO2 emissions and energy consumption. Additionally, people may use loans to improve their standard of living, which can fund construction, travel, and other activities that result in higher CO2 levels and energy consumption11. This study examines the impact of DE and FINE on CO2 emissions in the top ten emitting countries (USA, China, India, South Africa, Japan, Russia, Australia, Germany, UK, and South Korea). To mitigate the rise in global temperatures, it is essential to limit the increase to 1.6 degrees Celsius. Without addressing emissions from the largest polluters, global warming cannot be effectively prevented. Figure 1 illustrates the CO2 emissions trends in these top ten emitting countries.

Fig. 1
figure 1

CO2 emissions levels in top ten emitter countries.

Full size image

The ten emitter countries selected for this research, along with their projected CO2 share for 2021, are considered the top emitters of CO2 globally. Together, these countries accounted for approximately 80% of the world’s CO2 emissions in 202112. Given their significant contribution to environmental degradation, studying this group of countries is essential for achieving carbon neutrality goals by 2050. These countries represent a substantial portion of the global emissions, and their economic growth and energy consumption patterns are key to understanding the broader climate challenge. Sixteen of these countries are ranked among the top ten economies in the world, according to13. Furthermore, these nations are also among the top ten energy consumers globally14. This highlights the close link between high levels of CO2 emissions and economic growth, with the growth process often compromising environmental sustainability.

The decrease in CO2 emissions observed in Italy, Mexico, and New Zealand can be attributed to the use of geothermal energy, which has helped slow environmental degradation. In contrast, geothermal energy plants in the Philippines, Turkey, the United States, and India have contributed to an increase in CO2 emissions. This demonstrates that while geothermal energy has the potential to reduce emissions, its impact on global CO2 levels can vary depending on the context15. Additionally, the findings from Iceland indicate a negative correlation between CO2 emissions and energy intensity, renewable energy, and economic growth (EG), while there is a positive correlation between CO2 emissions and financial development16.

Significantly, previous research has not explored how economic development in the top ten emitters can mitigate the effects of the digital economy (DE). Additionally, the digital economy index used in this study comprises ten components, which capture factors related to digital trade, the societal impacts of digitalization, digital infrastructure, and the societal support that digitalization provides. Finally, unlike prior studies on the relationship between DE and CO2 emissions6, our research highlights the varying effects of DE and financial expansion (FINE) on emissions, using the newly developed MM-QR test17. Most recently, COP28 (2023) underscored the urgency of addressing emissions through comprehensive policies targeting the energy transition, renewable energy adoption, and the promotion of digital and financial solutions. This summit emphasized the transformative potential of digitalization and financial expansion in achieving carbon neutrality, aligning well with the objectives of this study.

In this context, the digital economy (DE) and financial expansion (FINE) have emerged as transformative forces with the potential to drive sustainability. The digital economy fosters innovation, efficiency, and connectivity, while financial expansion provides the capital necessary for green investments and technological advancement. However, their environmental impacts remain complex and multifaceted, as digitalization can simultaneously stimulate economic activities that increase emissions and enable technologies that reduce them. Similarly, while financial development facilitates investment in renewable energy and energy-efficient technologies, it can also intensify resource exploitation in some contexts.

Despite the growing attention to DE and FINE, existing research often overlooks their quantile-specific effects on CO2 emissions, particularly in the world’s largest emitting countries. Most studies rely on aggregate-level analyses that fail to capture the heterogeneous impacts across different levels of emissions. Additionally, the interaction between digitalization and economic growth (EG) remains underexplored, particularly its potential to offset the adverse environmental effects of increasing digital activity. These gaps highlight the need for a more nuanced understanding of how DE and FINE influence environmental outcomes across varying emission levels.

This study aims to fill these gaps by investigating the heterogeneous effects of DE and FINE on CO2 emissions in the top ten carbon-emitting countries from 1990 to 2021. Using the Method of Moments Quantile Regression (MM-QR), the study examines the quantile-specific impacts of DE and FINE on emissions, complemented by Driscoll-Kraay regression for robustness. Additionally, it explores the interactive effect of DE and EG, shedding light on the dual role of digitalization in influencing emissions. By focusing on the leading emitting nations, this research offers critical insights into how digital and financial mechanisms can be optimized to achieve sustainability goals.

The findings of this study contribute to the literature by providing robust evidence on the direct and indirect effects of DE and FINE on emissions. The research underscores the need for targeted policies that balance economic growth, digitalization, and environmental sustainability, offering actionable recommendations for policymakers in high-emission countries.

The novelty of this research lies in its multifaceted approach. First, it evaluates the dual impact of DE on emissions, capturing both direct influences and moderating effects through economic growth (EG). Second, it employs an innovative online economy index comprising ten components, including digital infrastructure, digital trade, and societal impacts. Third, the MM-QR method addresses heterogeneity and distributional variations in CO2 emissions, offering deeper insights than traditional regression techniques. Finally, this research provides actionable policy recommendations for leveraging DE and FINE to mitigate emissions, emphasizing their role in achieving carbon neutrality by 2050. The paper is structured as follows: Sect. 2 reviews the literature, Sect. 3 explains the methodology, Sect. 4 presents empirical analysis and discussion, and Sect. 5 concludes with policy implications and limitations.

Literature review

Digitalization is altering traditional business models and impacting various ecological and economic indicators. For example7, identified through data from 275 Chinese cities that the digital economy (DE) encourages sustainable growth and increases the use of clean energy. Similarly18, found that DE affects green development in 281 Chinese cities through several factors, including environmental degradation, technological advancements, human capital, and energy use. In a panel of 72 countries9, observed that DE enhances governance capabilities while supporting the energy transition. However, the effects of DE show heterogeneity at various quantiles. DE promotes the development of green power19, and its impact on renewable energy development in Asian countries is amplified by legal regulation, FINE, and political stability.

An investigation by20 found that DE boosts green productivity in China’s industrial sector, promoting sustainable growth. However, the effects of DE vary across Chinese regions21. demonstrated in another study that DE fosters sustainable development in developing countries. Moreover, industrialization, urbanization, and capital creation promote sustainable growth, while energy intensity and economic sustainability hinder it22. found that internet access and DE promote green development across 28 Chinese provinces. DE and increased internet usage thus encourage sustainable development23. claimed that DE drives ecological advancements and human resource development, both of which contribute to high-quality urban development in China.

In a global panel study24, investigated the effects of DE on energy consumption, observing that DE reduces energy consumption, energy intensity, and emissions. Furthermore, DE fosters innovation, economic growth, and human capital, which contribute to energy savings and, consequently, emission reductions. Similarly25, showed an inverse correlation between internet development and energy consumption in China.

The trend of digitization impacts energy system R&D, economic expansion, and financial growth8. examined the environmental performance of 25 European countries and demonstrated that certain digital transformation techniques improve environmental performance. They noted that although DE may have some negative short-term consequences, its long-term effects are generally positive. A city-level study by26 showed that DE has non-linear effects on environmental quality, initially increasing emissions, then reducing them at higher DE levels. Moreover, DE’s effects on emissions vary geographically, with the dominant effects seen in eastern regions of China, compared to other areas27. also observed that DE helps reduce emissions in China, particularly by impacting the electrical architecture, which affects emissions28. further noted that DE significantly lowers emissions only at elevated DE levels. DE also affects the economy and innovation, both of which impact emission intensity. For instance29, found that DE reduces airborne pollutants in China but also observed substantial variation in its influence.

ICT plays a crucial role in DE’s environmental impact30. demonstrated the value of ICT in reducing emissions across the BRICS nations. They found that higher levels of globalization could intensify pollution. Similarly31, showed that ICT has positive environmental effects in 59 Belt and Road countries, modifying foreign trade and direct investment, thereby reducing emissions both directly and indirectly. The study by32 on Saudi Arabia found that ICT is effective in regulating ecological footprints. In 202133, concluded that ICT can significantly boost economic growth.

However, some studies suggest that certain DE industries, such as digital manufacturing and industrialization, contribute substantially to CO2 emissions19. found that decarbonizing these industries is crucial for enhancing sustainability34. noted that DE increases emissions in a panel of 60 countries, though they did not account for the heterogeneous effects of DE on emissions, using a generalized method of moments (GMM) approach35. showed that ICT increases both CO2 emissions and economic growth. Similarly36, found that ICT boosts electricity consumption, which in turn increases emissions37. demonstrated that in Africa, ICT contributes to higher emissions, and a similar positive correlation was observed between ICT and CO2 in Sub-Saharan African (SSA) countries38. found that ICT increases CO2 levels in G7 countries, and39 noted that ICT raises the ecological footprint (EF) in the Next Eleven (N-11) countries. In Tunisia40, found no significant effect of ICT on emissions.

Financial expansion (FINE) and economic growth can have both positive and negative effects on the environment. On the positive side, many studies argue that FINE can finance renewable energy innovation and modern technologies, which help reduce energy consumption, resource use, and CO2 emissions41. revealed that higher FINE improves environmental quality in China and Malaysia42. found that FINE promotes renewable energy use in India, contributing to a cleaner environment43. showed that FINE reduces CO2 emissions in EU countries, and a similar relationship was observed in 27 countries by44,45 identified a similar trend in earlier studies on FINE’s impact on environmental contamination46. used multiple FINE metrics and found that FINE curbs environmental degradation across a sample of 131 countries47. similarly found that FINE reduces CO2 emissions in developing countries48. concluded that FINE improves environmental conditions in emerging nations.

On the other hand, FINE can also support building projects, industrial initiatives, and travel, all of which can lead to higher CO2 emissions and increased energy use. A 2021 provincial-level study by49 showed that FINE increases CO2 levels in China, while50 found that FINE contributes to higher CO2 levels in India. In developing countries51, showed that FINE worsens ecological quality and increases the ecological footprint. A study in the ASEAN-5 countries by52 found that FINE and economic growth contribute to higher CO2 emissions, negatively affecting the environment. Ahmed et al. (2021b) provided evidence that in Japan, FINE raises the ecological footprint, and50 observed similar trends in BRICS countries53. revealed varied effects of FINE and industrialization on the ecological footprint in N-11 countries.

In summary, the reviewed literature shows that DE can have a range of effects on emissions, both positive and negative, and these effects vary depending on the region and context. Additionally, FINE may have either beneficial or harmful effects on environmental quality, depending on the panel or country. Notably, there is no existing research on the relationship between DE and CO2 emissions in the top ten emitters (USA, China, India, South Africa, Japan, Russia, Australia, Germany, the UK, and South Korea), which together account for nearly 80% of global emissions. Furthermore, most previous studies have employed conventional regression techniques that fail to account for the distributional and heterogeneous effects of DE and FINE on emissions.

Gap in the literature

There exists a gap in the previous literature, as recent studies have yielded varying results regarding CO2 emissions and their relationship with other economic and environmental variables. For instance, the NARDL results reveal a negative and statistically significant relationship between technological innovation and hydroelectricity consumption in Colombia. This suggests that an increase in both variables has a positive impact on the environment in the long run. The frequency domain causality test also found a bidirectional relationship between technological innovation and CO2, but only a one-way relationship between hydroelectricity consumption and CO2 emissions54.

Other empirical studies indicate that economic globalization is linked to ecological degradation in Malaysia in the long term. In contrast, economic complexity, political stability, and energy transition are shown to enhance ecological sustainability in Malaysia over the long run55. Additionally, research findings suggest that an increase in the ecological footprint correlates with economic growth, while a reduction in the ecological footprint is associated with the rise of renewable energy, global trade, and human capital. The synergistic effects of human capital and global trade reduce the ecological footprint, and their combined influence alleviates the ecological footprint in Italy56.

Further research indicated that the efficiencies of coal and natural gas are essential in reducing CO2 emissions. However, the findings also demonstrated that globalization, economic expansion, resource efficiency, social globalization, political globalization, and economic globalization exacerbate CO2 emissions57. Another study employed the Method of Moments Quantile Regression (MMQR) to examine the asymmetric impact of various factors on carbon emissions. The MMQR results indicate that across all quantiles, green resource productivity, renewable energy, and economic globalization reduce CO2 emissions, while economic expansion increases CO2 emissions58. Similarly, research indicates that the relationship between ICT and greenhouse gas (GHG) emissions depends on the level of environmental taxation. In contexts with low environmental taxes, the impact of ICT on GHG emissions is positive but negligible; however, when environmental taxes exceed a certain threshold, the influence of ICT becomes inversely correlated with GHG emissions.

While existing literature has explored various dimensions of CO2 emissions, economic variables, and environmental factors, a valuable gap remains. In response to this gap, the present study aims to examine changes in CO2 emissions in the top ten emitters as a result of DE and FINE. Furthermore, this research investigates the indirect effects of DE on CO2 emissions via economic expansion, recognizing that DE impacts corporate operations and financial procedures that can potentially affect CO2 emissions.

Data description and theoretical framework

The relationships among DE, FINE, EG, PD and CO2 are covered in great detail here in this section. However, choosing an appropriate model that could make it easier to estimate the relationship between DE, FINE, PD, EG and CO2 emissions without adding bias from missing variables is necessary in order to evaluate the impact of DE and FINE on CO2. The popular “Stochastic Impacts by Regression on Population, Affluence, and Technology (STIRPAT) model” is the best option in this respect because it allows59 model to be employed in order to trace the cause of environmental deterioration by taking into account significant determinants like population, technology, and affluence in as well as a few other factors. The model’s equation is expressed as follows,

$$\:{I}_{t}=a{P}_{t}^{b}{A}_{t}^{c}{T}_{t}^{d}{\mu\:}_{t}$$
(1)

Where the regressors population (P), technology (T), and affluence (A) influence the environmental deterioration (I). In the past literature, economic growth (GDP) depicts affluence while CO2 is extensively used to depict I4,9,60,61,62, claim that human population density can be used as a proxy for environmental effect. The present research uses DE as a technology because the STIRPAT framework allows for different measures of technology. Because this model allows for extra factors that could affect I here. In order to evaluate the indirect effects of DE via EG, the interaction term (EGxDE) is added. The updated model is explained as follows,

$$\:\begin{array}{c}\begin{array}{cc}&\:{\left(LC{O}_{2}\right)}_{\text{it\:}}={\alpha\:}_{0}+{\delta\:}_{1}LE{G}_{it}+{\delta\:}_{2}LD{E}_{it}+{\delta\:}_{3}LFIN{E}_{it}+{\delta\:}_{4}LP{D}_{it}\\\:&\:+{\delta\:}_{5}LEGxD{E}_{it}+{\mu\:}_{it}\end{array}\\\:{\left(LC{O}_{2}\right)}_{\text{it\:}}={\alpha\:}_{0}+{\delta\:}_{1}LE{G}_{it}+{\delta\:}_{2}LD{E}_{it}+{\delta\:}_{3}LFIN{E}_{it}+{\delta\:}_{4}LP{D}_{it}\\\:+{\delta\:}_{5}LEGxD{E}_{it}+{\mu\:}_{it}\end{array}$$
(2)

Where EG, DE, FINE, PD and CO2 represent accordingly, the following: GDP constant 2015 per person, economic growth (EG), digital economy index, financial expansion (financial development index), CO2 emissions refers to the total metric tons of carbon dioxide released from burning fossil fuels and manufacturing processes and population density (individuals per square kilometer). The datasets were converted to their natural logarithms in accordance with earlier research27,63.

The term “digitalization” has not been standardized in previous literature effectively. It is often calculated using a variety of ICT-related variables63. , research from the current literature created a DE index by utilizing information on digital trade, digital infrastructure, the social effects of digitalization, and the social support that comes with it. Each of the four categories that make up this index has several sub-components. As a result, this index is extremely thorough since it gathers data from World Bank, International Telecommunication Union (ITU), and United Nations (UN) reports to represent the many aspects of DE. The digital economy index among the top ten emitters (USA, China, India, South Africa, Japan Russia, Australia, Germany, UK, and South Korea). The CO2 dataset was acquired from OWD (2023), whereas WDI (2022) provided the PD and EG datasets64,65 .We took the IMF data on26. In this study picture 1, describes the variables theoretic framework.

Methodology

There is an exploration of several fundamental dataset aspects prior to the formal evaluation. As an example, the potential cross-sectional dependence (CSD) is examined since the ten countries included in the panel are (USA, China, India, South Africa, Japan Russia, Australia, Germany, UK, and South Korea), significant players in global commerce. Regarding this, different procedures are employed to trace the CSD, such as the Breusch-scaled LM and CD approaches in conjunction with the Pesaran scaled LM and CD techniques Pagan LM examination. Of all these tests, the CD method is the most dependable for shorter-term data that includes additional countries. This examination is stated clearly in the following manner:

$$\:{\text{CDpes}}\:{\text{ = }}\sqrt {\frac{{{\text{2d}}}}{{{\text{s}}({\text{s - 1}})}}\left( {\sum {\:_{{{\text{i = 1}}}}^{{{\text{s - 1}}}} } \:\sum {\:_{{{\text{j = i + 1}}}}^{{\text{s}}} } \:{\text{C\_}}_{{{\text{ij}}}} } \right)}$$
(3)

When pair-wise correlation is described by C _ij, and CDpes applied by CD test. Moreover, d describes particular years, while s indicates the sample’s size. Dependency was shown in the top emitter’s panel by the evidence from each of these tests. Knowing if the data is diverse after that is crucial. The66 methodology is applied in this study, and two well-known tests \(\:{\Delta\:}\) and modified \(\:{\Delta\:}\) are chosen to determine the slope heterogeneity. This is how the delta test is presented.

$$\:\mathop {\Delta \:}\limits^{\sim } = (N)^{{\frac{1}{2}}} (2{\text{~}}K)^{{ - \frac{1}{2}}} \left( {\frac{1}{N}\mathop S\limits^{\sim } - K} \right)$$
(4)

The results of the second test, called the modified\(\:{\Delta\:}\), are displayed above.

$$\:\mathop {\Delta \:}\limits^{\sim } _{{ajd}} = (N)^{{\frac{1}{2}}} \left( {2{\text{~}}K\left( {\frac{{t - K - 1}}{{t + 1}}} \right)^{{ - \frac{1}{2}}} \left( {\frac{1}{N}\mathop S\limits^{\sim } - K} \right)} \right.$$
(5)

The importance of these statistics disproves the homogeneous slope coefficient hypothesis (null hypothesis). The actions of the unit-root are also investigated in the following step. The second-generation CIPS test is used since the top emitter’s panel lacks slope uniformity and independent. We decide the CIPS test to investigate the unit-root development of variables because of its benefits in accounting for heterogeneity and CSD67. The IPS technique is also utilized in combination. The Jarque-Bera test reveals that none of the variables in the basic analysis have a normal distribution. Furthermore, there are issues with heterogeneity and CSD in the datasets. The descriptive examination of the data also shows significant differences between time and scale. In the conditional distribution, the dependent variable likewise exhibits heterogeneous distribution. Because CO2 emissions differ between countries, using MM-QR makes sense for this research in the year 2020.

$$\:{D}_{p}\left(t/{R}_{i,t}\right)={\alpha\:}_{1}+{R}_{i,t}\xi\:+\left({a}_{i}+{Y}_{i,t}\right){U}_{i,t}$$
(6)

Where t shows cross sections, i denotes the quantile fixed effect, a.i. þ Yi; t \ \Ui; t represents the scale coefficient, and Dp represents CO2. Ri, t displays regressors (DE, EG, FINE, PD, and EGxPD). Following that, the effects of FINE and DE on CO2 are examined using the Driscoll-Kraay regression technique. Throughout CSD68, DK regression is quite useful. For both panels, the average errors generated by this examination are of comparable quality. Additionally, panel data with more cross-sections and less time can be used for this test. Heteroscedasticity and autocorrelation issues are taken into consideration by the DK method. Initially the pooled OLS is used in this work to get the DK standard errors.

Results and discussion

The dataset for the top ten emitting countries (USA, China, India, South Africa, Japan, Russia, Australia, Germany, UK, and South Korea) reveals significant variations in CO2 emissions and related indicators, as shown in Table 1. CO2 emissions, measured in tons per person, display substantial deviations and disparities between the lowest and highest levels, reflecting country-level heterogeneity. This variability is mirrored in economic growth (EG), population density (PD), and digital economy (DE) and financial expansion (FINE) indices. The indices, ranging from 0 to 1, highlight disparities in digitalization and financial development across these nations. Furthermore, the Jarque-Bera (J-B) test confirms a non-normal distribution of these variables, consistent with the diverse economic and environmental characteristics of the countries.

Table 1 Descriptive analysis.
Full size table

Table 2 provides evidence of cross-sectional dependence (CSD) among the countries analyzed. The CSD tests demonstrate significant interconnections within the dataset66, which is critical for understanding global carbon emissions dynamics. Additionally, the Δ and Δ-adjusted statistics underline the heterogeneity and structural unpredictability of the data, validating the use of advanced econometric techniques like MM-QR.

Table 2 C-D and S-H tests.
Full size table

Table 3 presented the unit root test analysis and confirms that while FINE is stationary at levels, CO2, EG, DE, and PD are integrated at first difference I(1). The consistency between IPS and CIPS tests reinforces the reliability of the dataset. These findings support the application of quantile-based regression methods, which account for the diverse performance and non-normality of CO2 emissions and their predictors.

Table 3 Panel unit root.
Full size table

The MM-QR results in Table 4 reveal a heterogeneous relationship between EG and CO2 emissions. EG exerts a significant positive effect on CO2 emissions, with this impact diminishing across quantiles. At lower quantiles, where countries are less industrialized, a rise in EG results in a sharp increase in CO2 emissions, consistent with findings for the European Union69 emerging nations60,62,70 for Turkey28, for China71, for Asian countries. However, at higher quantiles, the impact of EG on emissions decreases, indicating that advanced economies mitigate environmental degradation through improved technological innovation, governance, and adoption of low-carbon energy sources72. This diminishing trend underscores the importance of quantile-specific analyses, as it highlights that global warming’s adverse effects are less pronounced in highly industrialized countries.

Table 4 Method of moments quantile regression.
Full size table

The direct effects of DE on CO2 emissions show significant variation across quantiles. At lower quantiles (Q1–Q5), DE has minimal environmental risks, indicating that lower levels of digitalization are less harmful to the environment. However, at higher quantiles (Q6–Q9), DE increases CO2 emissions, suggesting that the energy-intensive nature of digital technologies, including ICT production and usage, contributes to environmental degradation73. These results are consistent with findings for the N-11 countries74 but contrast with studies on China22,75 and Saudi Arabia32.

The indirect effects of DE, as captured by the interaction term (EG×DE), reveal that digitalization can mitigate CO2 emissions by reshaping economic growth (EG). This mitigating effect is significant at higher quantiles, where digital transformation fosters dematerialization, enhances operational efficiency, and promotes sustainable practices76,77. While DE’s direct effects are detrimental, its indirect effects through EG demonstrate its potential to reduce emissions. These findings highlight the dual role of DE, emphasizing the need for policies that enhance digitalization’s positive impacts while minimizing its environmental costs.

FINE exhibits a consistent negative impact on CO2 emissions across all quantiles. This indicates that financial expansion facilitates investments in renewable energy, energy-efficient technologies, and environmental innovations. These results align with studies conducted in the European Union43, 27 countries44, and 131 countries78 but diverge from findings for BRI countries79,80 and China81. The uniformity of FINE’s effects across quantiles underscores its critical role in supporting global CO2 mitigation strategies. PD consistently reduces CO2 emissions across all quantiles, with its mitigating effect becoming more pronounced at higher quantiles. This finding aligns with the notion that denser populations in urbanized settings benefit from economies of scale, energy-efficient infrastructure, and shared transportation systems.

The robustness of the findings is confirmed through Driscoll-Kraay (DK) regression, as shown in Table 5. The DK standard errors account for serial correlation and CSD, with results consistent with MM-QR estimates. Furthermore, the GLS random effects approach supports the pooled OLS findings, confirming that EG and DE directly increase CO2 emissions while their interaction term (EG×DE) reduces emissions.

These findings have important policy implications. First, digital transformation and financial expansion must be leveraged to balance economic growth with environmental sustainability. Second, policymakers should focus on country-specific strategies to address the diverse economic and environmental characteristics of top emitters. Future research should consider disaggregated country-level analyses to complement the quantile-based findings and provide more actionable insights.

Table 5 Driscoll-Kraay test.
Full size table

Conclusion

This study investigates the impact of digital economy (DE) and financial expansion (FINE) on CO2 emissions in the ten largest carbon-emitting countries (USA, China, India, South Africa, Japan, Russia, Australia, Germany, the UK, and South Korea), using data from 1990 to 2021 and advanced econometric techniques like MM-QR and D-K test. The findings reveal a dual role of DE, with direct effects exacerbating CO2 emissions, particularly in high-emission countries, and indirect effects mitigating emissions by fostering economic growth (EG). FINE consistently reduces emissions, while population density exhibits a scaling effect, which becomes more pronounced at higher levels, helping to mitigate emissions effectively.

The results highlight that a uniform policy approach is inadequate, as the diverse socio-economic and environmental challenges of these countries necessitate tailored strategies. For instance, countries like China and India should prioritize sustainable digital infrastructure to offset the direct negative impacts of DE. In contrast, developed economies such as the USA, UK, Australia, and Germany should focus on enhancing the efficiency of digital technologies to amplify DE’s indirect environmental benefits. Countries with less-developed financial systems, such as India, Russia, and South Africa, should emphasize financial inclusion and implement green financing mechanisms to maximize the benefits of FINE.

To address the direct environmental impact of DE, there is an urgent need to develop energy-efficient digital devices that reduce emissions during both production and operation. Recycling facilities for e-waste should be strategically located in urban areas, coupled with consumer incentive programs to return old electronic devices for recycling. Such measures could be reinforced by public awareness campaigns to promote responsible disposal and sustainability practices. Investing in research and development for innovative technologies is equally crucial, ensuring that digitalization fosters economic growth while reducing its environmental footprint. Furthermore, international collaboration is essential, particularly to assist developing nations in adopting advanced digital infrastructure and technologies for sustainable development.

Efforts to mitigate emissions must also include promoting financial expansion to support green initiatives. Financial institutions should provide targeted lending programs for digital transformation and environmentally friendly technologies. Strengthening the role of financial systems in financing green innovation can enhance the sustainability of economic activities while minimizing their environmental costs. This approach also requires fostering partnerships between governments, private sector stakeholders, and international organizations to ensure that financial mechanisms are effectively utilized to support carbon mitigation efforts.

Urbanization poses another significant challenge, and integrating digital technologies into urban planning can optimize housing systems, energy use, and transportation. Collective transportation systems supported by digital infrastructure can reduce congestion and emissions while improving urban living standards. Population density (PD), when managed effectively with sustainable urban infrastructure, can contribute significantly to carbon mitigation. Digital technologies can enhance energy efficiency in urban areas through innovations such as smart grids, intelligent meters, and modernized cooling and heating systems.

Despite its novel findings, this study acknowledges certain limitations. It captures only a subset of emission predictors, leaving out crucial factors such as green environmental taxation, the role of digital finance, and other relevant policies. Future research should explore these dimensions and examine DE’s role in achieving carbon neutrality through energy efficiency, sustainable urbanization, and transportation. Investigating the impact of DE across different income groups or regions can provide additional insights into sustainable development strategies.

In conclusion, this study underscores the critical role of digital economic growth and financial expansion (FINE) in shaping global CO2 mitigation strategies. However, it emphasizes that one-size-fits-all policies are ineffective for countries with distinct socio-economic and environmental contexts. By adopting tailored measures, these nations can leverage the transformative potential of digitalization and finance to achieve carbon neutrality and support global sustainability goals.

Policy implications

There is a pressing need to establish a policy framework that aligns with the unique socio-economic and environmental characteristics of these nations. Uniform strategies cannot address the diverse challenges faced by these countries. For example, China and India should emphasize developing sustainable digital infrastructures and reducing the energy intensity of their growing digital economies. On the other hand, advanced economies like the USA, UK, Australia, and Germany should focus on improving the efficiency of digital technologies, maximizing their indirect environmental benefits, and ensuring minimal digital waste. Nations with emerging financial systems, such as India, Russia, and South Africa, must prioritize financial inclusion, adopt green financing mechanisms, and foster innovation to unlock FINE’s full potential in emission mitigation.

The direct environmental burden of DE necessitates policies to enhance the energy efficiency of digital devices and manage e-waste effectively. Recycling facilities must be strategically developed in urban centers, complemented by programs incentivizing consumers to return outdated electronics. Promoting public awareness on the environmental consequences of improper e-waste disposal is equally important. Additionally, manufacturers should be encouraged to adopt circular economy principles, such as designing modular electronics and providing incentives for refurbishing devices. Investments in research and development are vital for creating low-energy, high-performance digital technologies that align with sustainability goals.

Financial institutions must play a pivotal role in driving this transition by offering green credit facilities, supporting eco-friendly innovations, and fostering partnerships for sustainable development. Strengthening green financial systems can enable the digital transformation necessary to address climate change effectively. At the same time, governments should collaborate with international organizations to enhance access to digital technologies for developing economies, ensuring an equitable and sustainable digital transition.

Urbanization, coupled with population density (PD), represents another significant opportunity for emission reduction. Policymakers should prioritize integrating digital technologies into urban planning. This includes implementing smart grids, intelligent metering, and automated transportation systems to optimize energy usage, minimize congestion, and promote sustainable urban living. Efficient urban planning that leverages digitalization can significantly reduce emissions, particularly in densely populated regions.

Limitations

Despite its significant contributions, this study acknowledges certain limitations. The analysis excludes other potential emission predictors, such as environmental taxes, renewable energy adoption, and the role of digital finance. Moreover, the findings are derived from data spanning 1990–2021, which may not fully capture recent technological and policy developments. Future research should explore these additional dimensions, focusing on how DE and FINE contribute to achieving carbon neutrality through advancements in energy efficiency, sustainable urbanization, and low-carbon transportation systems. Expanding the analysis to include a wider range of countries, particularly those with varying income levels, can provide more comprehensive insights into the role of digitalization and financial expansion in emission mitigation.

In conclusion, the study highlights the critical role of digital economic growth (EG) and financial expansion (FINE) in shaping global CO2 mitigation strategies. However, the findings also underscore the necessity of adopting tailored, context-specific policy frameworks rather than one-size-fits-all approaches. By addressing the limitations of this study and exploring new dimensions, future research can build on these findings to develop comprehensive strategies that align digital and financial transformations with global sustainability goals.