Climate change has emerged as one of the most urgent and complex global challenges of the 21st century (Feulner, 2015). A primary driver of this crisis is the persistent reliance on fossil fuels, which has significantly contributed to environmental degradation and an alarming rise in greenhouse gas emissions—particularly carbon emissions (CEM) (Mehmood et al., 2024). Recognizing the escalating threats posed by CEM to sustainability, development, and environmental integrity, policymakers, researchers, and international organizations have intensified their focus on mitigation strategies (Raihan and Tuspekova, 2022).

Efforts to curb climate change remain constrained by the carbon-intensive nature of economic development, energy use, and trade, as well as the embeddedness of emissions in current industrial and institutional systems (Sarkodie, 2021). The United Nations’ Sustainable Development Goal (SDG) 13 explicitly calls for urgent action to combat climate change and its impacts (Department of Economic and Social Affairs, 2025). In this context, reducing CEM while maintaining economic growth has become a critical research and policy imperative (Dong and Yu, 2024; Ren and Zhu, 2025; Zhao et al., 2025).

This study examines the role of green technology (GT), energy intensity (EIN), environmental performance (EPR), and economic growth (GDP) in shaping carbon emissions across G5 economies—Brazil, China, India, Mexico, and South Africa—from 2000 to 2022. The analysis aims to offer actionable insights for effective climate action plans by investigating how these variables interact and influence emissions outcomes in rapidly developing economies. While green investments and innovations are widely promoted as tools for sustainable development, their empirical effectiveness—particularly in emerging economies—remains underexplored (Duan, 2025; Xia et al., 2025; Xu et al., 2025).

In parallel, the deteriorating environmental performance in some regions underscores the importance of institutional reforms (Barra and Falcone, 2024; Feng et al., 2024; Tufail et al., 2021) emphasize that improving fiscal decentralization, lowering the cost of non-renewables, and strengthening institutional quality are crucial to reversing environmental degradation. Meanwhile, evidence suggests that increases in EIN and foreign direct investment (FDI) may inadvertently escalate CEM (Lan et al., 2012). Their findings highlight the spatial spillover effects of EIN reduction targets in China and suggest a nonlinear relationship between GT innovation and CEM mitigation, dependent on innovation thresholds.

This study aligns closely with multiple SDGs. SDG 7 emphasizes the need for affordable and clean energy, and EIN is directly related to energy consumption efficiency—a key indicator of sustainable energy use. Lower EIN reflects improved energy efficiency, contributing to reduced emissions (Tsvetkov et al., 2024). Similarly, SDG 8 (Decent Work and Economic Growth) and SDG 9 (Industry, Innovation, and Infrastructure) are addressed through the analysis of GDP and its relationship with CEM, underscoring the challenge of reconciling economic development with environmental protection. Green technology—such as clean energy innovations and patents for eco-efficient processes—are essential levers in this transition (Dong et al., 2025).

Further, SDG 12 advocates for sustainable consumption and production patterns, which can lower EPR and ultimately reduce CEM through improved industrial practices and resource efficiency (Barbhuiya et al., 2024). The role of international partnerships, as captured under SDG 17, is also critical; collaboration in clean technology development and financial mechanisms is necessary to support low-carbon transitions (Weko and Goldthau, 2022). However, recent SDG progress reports show that G5 countries are falling short on key environmental targets, particularly those related to climate action, clean energy, and sustainable production (Sachs et al., 2022; Apergis, 2016) stress that scaling up investments in renewable energy, restoring natural ecosystems, and advancing GT are pivotal to achieving carbon neutrality.

The G5 nations were selected for this study due to their significant contributions to global economic activity and carbon emissions, as well as their strategic importance in the climate transition. Collectively, these countries represent more than 30% of global carbon emissions and are among the top emerging economies contributing to global GDP growth. Moreover, their participation in key global forums such as the G20 and COP climate negotiations highlights their central role in shaping international environmental policies (Alomair et al., 2025; Jia et al., 2024). These countries—Brazil, China, India, Mexico, and South Africa—represent the world’s leading emerging economies and are simultaneously key stakeholders in climate negotiations. Their rapid industrialization, rising energy demands, and heterogeneous policy responses provide a valuable context for examining the complex trade-offs between growth and environmental sustainability.

Moreover, the geographical and developmental diversity among G5 countries allows for a more nuanced understanding of how GT, EIN, EPR, and GDP interact with emissions. Brazil and Mexico offer insight into Latin America’s natural resource dynamics and institutional responses. China and India, as Asia’s economic giants, reflect the tensions between industrial growth and green innovation. South Africa, as the leading economy in Africa, exemplifies the challenge of decarbonizing economies heavily reliant on fossil fuels.

Figure 1 illustrates the distribution of CEM across the G5 nations over the study period. China exhibits the widest range and highest median emissions, highlighting its sustained industrial expansion. South Africa also displays a high median with moderate variability. Mexico’s emissions are more stable, while Brazil and India maintain the lowest medians—though Brazil’s emissions show greater variability due to periodic surges. These patterns reflect the developmental trajectories and policy environments unique to each country.

By investigating the effects of GT, EIN, EPR, and GDP on carbon emissions, this study contributes to the understanding of environmental sustainability pathways in emerging economies. It aims to inform policymaking, support international climate goals, and offer a comparative framework for designing green growth strategies in the Global South.

Although earlier works have examined aspects of these relationships individually—such as the link between EIN and CEM (Bashir et al., 2021), or the influence of GT on emissions reduction (Luo et al., 2024; Lv et al., 2024)—these studies typically isolate variables rather than examining their combined and interactive effects. Few studies have integrated GT, EIN, EPR, and GDP into a unified empirical framework to assess their collective impact on CEM. Moreover, emerging literature still lacks the application of dynamic and distribution-sensitive techniques, such as the MMQR method or the Westerlund cointegration test, especially in multi-country studies focusing on emerging economies (Machado and Santos Silva, 2019).

Additionally, longitudinal research examining the evolution of GT’s impact on CEM over extended periods remains scarce. Most studies provide only short-term snapshots or static assessments, failing to capture the structural transformations and policy shifts that occurred between 2000 and 2022—a period marked by considerable regulatory and technological advancements (Chien et al., 2023; Dhayal et al., 2025). These evolving dynamics are crucial to understanding how sustainable transitions unfold in emerging economies.

These research questions collectively aim to highlight the technological advancements of the G5 nations in mitigating CEM, particularly through the application of clean technologies, environmental innovations, and renewable energy patents. The study also evaluates whether economic expansion can occur alongside environmental sustainability, and whether improvements in EPR can support the achievement of Sustainable Development Goals (SDGs).

To achieve these objectives, a comprehensive econometric framework is employed. The Method of Moments Quantile Regression (MMQR) developed by Machado and Santos Silva (2019) is particularly appropriate for this study, as it captures the heterogeneous effects of explanatory variables (GT, EIN, EPR, GDP) across different points in the carbon emissions distribution. This is essential because the G5 countries differ significantly in their emission profiles and policy contexts, which cannot be fully addressed using mean-based models like OLS or traditional fixed effects.

MMQR allows for a nuanced understanding of how each factor affects low-, medium-, and high-emission scenarios, aligning well with our research goal of exploring decoupling between economic growth and emissions at multiple intensity levels. This technique has been increasingly applied in energy-environmental economics, though its use in multi-country G5-focused studies remains limited (Ulucak and Ozcan, 2020).

We complement MMQR with several panel diagnostics to ensure robustness. The slope heterogeneity test (Hashem Pesaran and Yamagata, 2008) addresses structural and institutional differences across countries. The cross-sectional dependence test accounts for potential spillover effects among countries—especially relevant in the globally integrated context of energy and environmental policy. The CIPS unit root test ensures that data are stationary, while the Westerlund cointegration test verifies long-run relationships among variables. This combination of tests enhances both the credibility and depth of the analysis.

Additionally, panel Granger non-causality tests (Dumitrescu and Hurlin, 2012) are used to assess the direction of influence between variables, providing policy-relevant insights on whether, for example, GT and GDP can predict emission trends.

Although our study does not develop a new technique, it offers a novel methodological application by integrating MMQR with cointegration and causality analysis in the context of G5 countries—something not previously done in this combination. This integrative approach allows us to explore both distributional impacts and temporal causality in a single framework, advancing the methodological application of MMQR in comparative environmental research.

Figure 2 illustrates the temporal variation of carbon emissions across G5 countries from 2000 to 2022. China shows a consistent and steep upward trend in emissions until 2020, likely driven by rapid industrialization and economic growth, followed by a visible decline possibly attributable to recent policy shifts or global disruptions. South Africa presents persistently high emission levels with relatively stable trends, except for a mild dip in 2020. Mexico and Brazil exhibit moderate emissions with fluctuations, including observable declines around 2016–2020 in Brazil and in 2020 in Mexico. India demonstrates a steady upward trend, corresponding with its gradual but continuous development trajectory, with a slight decline after 2018 that may reflect the effects of structural reforms.

These country-specific patterns reflect the diversity of economic development pathways and the critical role of industrialization, regulatory interventions, and external shocks in shaping CEM trends. By contextualizing these dynamics within a robust econometric analysis, this study offers valuable insights into sustainable development strategies applicable to the broader Global South.

The remainder of the paper is organized as follows. Section 2 presents the literature review on CEM, GT, EIN, EPR, and GDP, along with the identified research gaps. Section 3 describes the data sources and empirical methodology. Section 4 reports and interprets the empirical findings. Section 5 concludes the study with policy implications and recommendations.

This section explores the theoretical and empirical linkages between green technology (GT), energy intensity (EIN), environmental performance (EPR), economic growth (GDP), and carbon emissions (CEM), with a focus on how these variables interact to support sustainability objectives, particularly in emerging economies. GT has increasingly played a pivotal role in the global reduction of carbon emissions. Green technology (GT)—also referred to in the literature as environmental or clean technologies—refer to innovations and practices that minimize environmental harm by enhancing energy efficiency, reducing emissions, and supporting sustainable resource use. These technologies encompass a range of applications, including renewable energy systems, energy-efficient industrial processes, carbon capture technologies, and low-emission transportation (UNCTAD, 2024). In this study, we use the term green technology (GT) consistently and proxy it through the number of environmental technology patents filed, in line with OECD and WIPO classifications.

The role of GT in mitigating carbon emissions has been emphasized in recent reports. The highlights that clean energy innovation needs to triple by 2030 to meet global climate targets. Similarly, the IPCC Sixth Assessment Report (IPCC, 2025) notes that scalable low-carbon technologies are essential to achieving net-zero transitions. GT has been linked to significant improvements in environmental performance by decoupling economic activity from emissions, particularly in emerging economies (Chen and Yang, 2024; Kashif et al., 2024).

Several empirical studies support the role of GT in emissions mitigation. For instance (Bilgili et al., 2016), found that renewable-based GT significantly reduced emissions in OECD countries, while (Chen et al., 2023) reported similar outcomes in BRICS nations. Despite these insights, there remains limited evidence for multi-country applications involving G5 countries, especially using robust distributional techniques. Our study addresses this gap by incorporating GT into a comprehensive panel framework covering the years 2000–2022.

Rafei et al. (2022) found that economic complexity significantly improves EPR, implying that advanced, diversified economies tend to perform better environmentally. While numerous studies have examined the relationship between economic growth and environmental degradation, many have focused narrowly on renewable energy use, often omitting the broader context of innovation and institutional performance.

Technological innovation in renewable energy depends heavily on R&D, requiring a robust and well-functioning financial system. This study departs from prior research by examining the combined effects of GT, EIN, EPR, and GDP on CEM, with a specific focus on the G5 economies. (Zhang and Vigne (2021) investigated the effect of EPR and CEM disclosures on firm value, using data from SRI-KEHATI index-listed firms. Their findings suggest that while CEM disclosure does not directly affect firm value, EPR and integrated disclosure do have a positive effect.

Guo et al. (2023) provided further insight, indicating that renewable energy development contributes more substantially to reducing environmental pressure in low-income countries compared to middle-income ones. Ulucak and Ozcan (2020) analyzed OECD countries from 1980 to 2016 using the augmented mean group (AMG) estimator. Their results showed that renewable energy consumption helps mitigate environmental degradation, while dependence on non-renewable sources exacerbates it. These findings emphasize the need for countries to shift toward cleaner energy pathways.

Regarding energy intensity, higher economic activity often entails greater energy consumption, leading to increased emissions. Bashir et al. (2021) found that increased EIN significantly contributes to higher CEM, while (Wen et al., 2022) reaffirmed the critical role of energy efficiency technologies in lowering environmental damage. Additionally (Xu et al., 2022), demonstrated that improved EPR is associated with lower ecological footprints, highlighting its relevance as a sustainability indicator. Liu et al. (2024) emphasized the importance of policy and financial support in fostering GT development, revealing that countries with greater eco-innovation investments tend to experience sharper declines in emissions.

The relationship between GDP and CEM is more complex when green technology is factored in. Luo et al. (2024) explored the causal links among GT, GDP, and CEM, and suggested that the introduction of GT can decouple economic growth from rising emissions. Their findings support the idea that investments in green innovation can enable sustained development without exacerbating environmental pressures.

Overall, the existing literature confirms that GT, EIN, EPR, and GDP all play significant roles in influencing CEM. Advancements in energy efficiency and environmental governance are critical for emission reductions. GT promotes the adoption of cleaner production methods and alternative energy sources, offering promising solutions for climate change mitigation. By enabling the decoupling of carbon emissions from economic growth, green technology can contribute meaningfully to sustainable development. This study aims to expand on this literature by empirically testing these dynamics across the G5 economies using advanced panel econometric techniques. The insights are intended to guide policymakers in designing integrated strategies for climate resilience and inclusive development.

The study includes four main independent variables: green technology (GT), energy intensity (EIN), environmental performance (EPR), and economic growth (GDP), to investigate their effects on carbon emissions (CEM) in the G5 economies from 2000 to 2022.

This particular combination of variables offers a novel multidimensional lens. While individual pairwise combinations (e.g., GT and GDP, EIN and CEM) have been analyzed in prior research (Ulucak and Ozcan, 2020), the joint integration of all four variables using a quantile-based, long-run panel approach is rare, especially in the G5 context.

All variable definitions are presented in Table 1, but are detailed here for clarity:

• Carbon Emissions (CEM): Measured in metric tons per capita (WDI, 2019).

The use of ecological footprint (EF) as a proxy for EPR is based on its ability to capture broad environmental pressure (biocapacity use vs. availability), widely used in studies like (Ha et al., 2022; Sarkodie, 2021). This provides a holistic indicator beyond single-pollutant measures.

The analytical methods employed include Method of Moments Quantile Regression (MMQR), Westerlund cointegration tests, and Dumitrescu-Hurlin causality analysis. MMQR is particularly appropriate for this analysis because it detects how the relationship between predictors and emissions varies at different emission quantiles, reflecting cross-country heterogeneity. This is critical for the G5, whose economies vary widely in emissions levels and development.

Cointegration analysis ensures that the long-term relationships among non-stationary variables are valid, while the panel Granger causality test examines directionality and policy relevance of relationships (e.g., whether GT “causes” lower emissions).

The chosen period, 2000–2022, is based on data availability and policy relevance. This period captures the rollout of major global environmental initiatives (Sun et al., 2022), sharp technology transitions, and rising climate governance in emerging economies. Although data for some variables (e.g., patents or EF) are not yet available for 2023 across all G5 countries, 2022 offers the most recent complete year.

Limiting the analysis to 2000–2022 ensures that the findings reflect the full evolution of green innovation and emission intensity policies in the 21st century, while minimizing inconsistencies caused by incomplete data.

MMQR technique regarding this research is given in Equation 1. Q C E M τ ∣ G T , E P R , E I N , G D P = β 0 τ + β 1 τ · G T + β 2 τ · E P R + β 3 τ · E I N + β 4 τ · G D P + ϵ τ (1)

Where, Q C C E τ ∣ · illustrates the τ -th quantile of conditional distribution of CEM.

β i τ (for i = 0 , 1 , … , 4 ) shows the quantile-specific coefficients that vary with τ , and observes the various effects of independent variables at different points of CEM distribution.

Error term is denoted by ϵ τ for the quantile τ -th.

It takes into consideration the heterogeneous impacts across several quantiles, the MMQR technique has the advantage of offering a comprehensive knowledge of how each independent variable affects the dependent variable (CEM) throughout its distribution, as opposed to only at the mean. This is especially helpful for studies on CEM, as different factors may have various impacts at different levels of CCE metrics. C E M i t = α i + β 1 i G T i t + β 2 i E PR i t + β 3 i E I N i t + β 4 i G D P i t (2)

Equation 2 depicts the summarization of Slope heterogeneity (Pesaran, 2006) test.

Where, α i is the individual-specific intercept, β k i are the slope coefficients that can be varied across individuals and error term is denoted by ϵ i t .

However, to find out if panel data exhibits cointegration, the (Westerlund and Edgerton, 2008) test has also been conducted before MMQR analysis. A long-term equilibrium relationship between the dependent variable and one or more independent variables is indicated by cointegration. It is based on the error correction model depicted in Equation 3. Where y i t manifests the dependent variable, x i t denotes the independent variable and Δ represents the first difference. α i is the adjustment coefficient whereas, β i ′ is the vector of long-run coefficients. Four distinct test statistics are created by Westerlund to evaluate cointegration in panel data: G t G α P t , and P α . The purpose of these test statistics, which come from the panel error correction model (ECM), is to ascertain whether the variables in the panel have an equilibrium connection over the long-run. Δ y i t = α i y i , t − 1 − β i ′ x i , t − 1 + ∑ j = 1 p   γ i j Δ y i , t − j + ∑ j = 0 q   δ i j Δ x i , t − j + ϵ i t (3)

Equation 4 given below is supposed to be used when incorporating the Westerlund test in the situation where CEM is the dependent variable and GT, EPR, EIN and GDP are the independent variables: C E M i t = α i + β 1 G T i t + β 2 E I N i t + β 3 E P R i t + β 4 G D P i t + ϵ i t (4)

Dependent variable for the cross-section i at time t is C C E i t . Individual-specific intercept has been shown by α i . While β 1 , β 2 , β 3 , β 4 are the slope coefficients and ϵ i t is the error term.

An extensive and reliable method for managing complicated data is provided by econometric analysis, which makes use of matrix correlation, slope heterogeneity, cross-sectional dependence tests such as the CIPS unit root test, Westerlund cointegration analysis, and MMQR. Figure 4 is showing the flowchart of test techniques. Moreover, Matrix correlation offers a comprehensive understanding of interdependencies by assisting in the identification of correlations between various variables. Slope heterogeneity improves the precision of model estimations by acknowledging that various cross-sectional units may display a range of behavioural tendencies. Besides, the validity of panel data results is ensured by cross-sectional dependence tests, including the CIPS unit root test, which address potential correlations between units. Strong techniques for identifying long-term equilibrium relationships are provided by Westerlund’s cointegration tests, even in the presence of cross-sectional dependence or structural fractures. Furthermore, MMQR offers a refined comprehension of the correlation between variables across several quantiles, hence identifying heterogeneity in the response variable distribution and facilitating more customized policy implications. When combined, these techniques improve the econometric findings’ depth, dependability, and applicability.

While MMQR is suitable for capturing distributional effects across quantiles, we also employed complementary techniques such as Westerlund cointegration and Dumitrescu-Hurlin causality tests to verify long-run equilibrium and directionality. These methods, when used together, reduce model misspecification risk and enhance the robustness of findings. Moreover, the dataset passed key diagnostics such as cross-sectional dependence and slope heterogeneity tests, confirming the appropriateness of applying advanced panel estimators over simpler pooled or fixed effects regressions.

This section presents the empirical results of the study. Table 2 provides the descriptive statistics for the study variables, while Table 3 displays the correlation matrix. The descriptive statistics indicate variability across countries and time, which underscores the heterogeneity in economic and environmental profiles within the G5 nations.

The skewness and kurtosis as indicated by the Jarque-Bera statistics show that most variables deviate from normality, supporting the use of non-parametric quantile regression methods.

Table 3 presents the correlation matrix among the key variables. Carbon emissions (CEM) are strongly and positively correlated with energy intensity (EIN) at 0.7032, indicating that countries with higher energy intensity tend to emit more carbon. In contrast, CEM is moderately and negatively correlated with environmental performance (EPR) at −0.3746, suggesting that improved environmental governance is associated with lower emissions. Additionally, green technology (GT) are positively associated with both EPR (0.5443) and GDP (0.4503), underscoring their dual environmental and economic relevance.

The strong statistical association between environmental performance (EPR) and carbon emissions may reflect overlapping constructs, particularly since ecological footprint per capita captures a broad spectrum of environmental pressures—land use, carbon absorption, and resource demand. However, ecological footprint remains one of the most comprehensive and globally comparable metrics, used extensively in environmental sustainability research (Li et al., 2023). It reflects not only emissions but also the systemic capacity of a country to sustain its ecological demand. Nonetheless, we caution that part of the correlation observed may stem from structural overlap with the dependent variable and encourage future research to triangulate EPR with multidimensional indices, such as the Environmental Performance Index (EPI) or planetary boundaries framework.

Table 4 presents the results of the cross-sectional dependence (CSD) test. All variables show significant cross-sectional dependence, with p-values below 0.01. This implies that shocks or policy changes in one G5 country are likely to influence the others, validating the need to account for interdependencies in panel estimations.

Table 5 reports the slope heterogeneity test results. Both the Delta and Adjusted Delta statistics are highly significant (p < 0.01), confirming that the slope coefficients vary across the G5 countries. This heterogeneity reinforces the choice of quantile regression to capture differential effects across emission levels and nations.

The slope heterogeneity test (Hashem Pesaran and Yamagata, 2008) confirmed statistically significant differences in slope coefficients across countries. This suggests that the relationships between variables—particularly between GT and CEM or EIN and CEM—are not uniform across G5 economies. For instance, the effect of green technology may be more pronounced in technologically advanced countries like China, while energy intensity may matter more in South Africa’s coal-heavy energy mix. These heterogeneous dynamics imply that policy implications must be country-specific and caution should be exercised when generalizing findings to other economies without accounting for contextual factors such as industrial structure, institutional capacity, and policy enforcement.

Table 6 provides the results of the CIPS unit root test. The findings reveal that all variables are non-stationary at levels [I (0)] but become stationary after first differencing [I (1)]. This justifies the use of cointegration analysis to examine long-term relationships.

The results of the Westerlund cointegration test are shown in Table 7. Three of the four test statistics (Gt, Ga, Pt, and Pa) indicate significant cointegration, particularly Gt and Pt, which have p-values of 0.000. This confirms the existence of a stable long-term equilibrium relationship among CEM, GT, EIN, EPR, and GDP.

The preliminary diagnostics confirm that the panel data are characterized by cross-sectional dependence and slope heterogeneity, and the variables are integrated of order one with evidence of cointegration. These results validate the application of MMQR for estimating the differential impact of the independent variables on CEM across its conditional distribution, which will be discussed in the subsequent section.

The Method of Moments Quantile Regression (MMQR) results are reported in Table 8, which captures the differential effects of independent variables—green technology (GT), environmental performance (EPR), energy intensity (EIN), and economic growth (GDP)—on carbon emissions (CEM) across various quantiles (0.25, 0.50, 0.75, and 0.90). This approach provides a more nuanced understanding of the emission dynamics across low-, medium-, and high-emission scenarios.

The MMQR results indicate that green technology have a consistently negative and statistically significant effect on CEM across all quantiles. The strongest effect is observed in the lowest quantile (−0.150), suggesting that green technology adoption is especially effective in reducing emissions in countries or periods with relatively lower emission levels. This highlights the essential role of GT in decarbonization efforts across varying environmental contexts.

Conversely, EPR shows a positive and significant relationship with CEM across all quantiles, although the effect diminishes as emissions increase. This counterintuitive result suggests that current EPR indicators may not be fully capturing the effectiveness of environmental policies, or that higher ecological footprints still coexist with rising emissions in some G5 countries. It reflects the possible gap between policy design and actual implementation.

Energy intensity (EIN) is also positively associated with CEM across all quantiles, with the highest effect seen at the 0.90 quantile (0.035). This finding reinforces the well-established link between inefficient energy use and increased emissions and underscores the urgent need for energy efficiency improvements, especially in high-emission contexts.

Economic growth (GDP), although showing a negative relationship with CEM, does not reach statistical significance at any quantile. This suggests that while some decoupling between growth and emissions may be occurring, it is not yet strong or consistent enough across the G5 countries to yield robust results in this model.

Figure 5 visually presents the quantile-specific impacts of GT, EPR, EIN, and GDP on CEM, with marker symbols indicating the level of statistical significance. The horizontal dashed line represents the zero-impact threshold, clearly distinguishing variables with mitigating versus exacerbating effects on emissions.

To validate the robustness of the MMQR findings, FMOLS and DOLS estimations were performed. Table 9 summarizes the results.

The robustness checks confirm the primary findings from the MMQR analysis. Both FMOLS and DOLS show that energy intensity significantly increases carbon emissions, reaffirming the need to reduce energy consumption per unit of economic output. Similarly, environmental performance remains positively associated with emissions, although the magnitude is slightly lower in the DOLS estimation. This again raises questions about the adequacy of current environmental performance metrics in driving emission reductions.

GDP shows a statistically significant negative relationship with emissions in both models, indicating that structural economic improvements may facilitate lower emissions in the long run. Lastly, green technology continue to exhibit a negative association with CEM, with stronger effects observed under the DOLS estimation, underscoring the strategic importance of investing in clean technologies for long-term climate mitigation.

Figure 6 illustrates the relationships among variables using DOLS and FMOLS trends, further reinforcing the directional consistency of the results across methodologies.

In summary, the empirical results suggest that while energy intensity and environmental performance continue to exert upward pressure on emissions, green technology play a decisive role in emissions reduction. The influence of GDP is more nuanced and context-dependent, suggesting a potential shift toward green growth. These findings underscore the need for more targeted and integrated policies that not only promote clean technology adoption but also enhance the effectiveness of environmental performance measures and energy-use efficiency in emerging economies.

To further investigate the directionality of relationships among the variables, the (Dumitrescu and Hurlin, 2012) Panel Non-Causality Test was conducted. The results, presented in Table 10, reveal statistically significant causal effects from green technology (GT), energy intensity (EIN), environmental performance (EPR), and economic growth (GDP) toward carbon emissions (CEM). In contrast, no evidence was found for reverse causality from CEM to any of these predictors, suggesting a unidirectional causality pattern.

Specifically, GT Granger-causes CEM with strong significance (p < 0.01), confirming that the expansion of green technology significantly contributes to emissions mitigation across the G5 countries. Likewise, GDP shows a significant causal relationship with CEM, indicating that economic activities influence emissions trends—likely reflecting industrial structure, energy usage patterns, and environmental policies. The causal impact of EIN on CEM is also robust, supporting the argument that energy intensity is a key driver of emissions. Finally, EPR exhibits a weaker but still statistically significant causal relationship with emissions, implying that environmental performance—although imperfect—may influence emission dynamics to some extent.

Overall, these results underscore the importance of proactive policy interventions that promote green technology deployment, improve energy efficiency, and strengthen environmental performance frameworks to effectively reduce carbon emissions in emerging economies.