The Belt and Road Initiative (BRI) was introduced by the Chinese President Xi Jinping in 2013 to revive the historic Silk Road. At its launch, 65 countries joined the initiative, and by 2022, this number had grown to 138 countries (Hussain and Zhou, 2022). Recently, 150 countries have signed a memorandum of understanding with China, making the BRI one of the most significant global infrastructure and economic development projects. The primary goal of the BRI is to enhance regional connectivity across Africa, Asia, and Europe by promoting economic development, infrastructure connectivity, and trade integration. Additionally, the initiative aims to lower tariff barriers, improve technological transfer, and foster financial integration and international cooperation.

Two years after the launch of BRI, global foreign trade experienced significant growth, rising from $38 billion in 1980 (Wang et al., 2017) to $24.59 trillion in 2015 (Du and Lu, 2018). Trade between BRI partner countries and China has also expanded, reaching a cumulative volume of $19.1 trillion, with an average annual growth rate of 6.4% between 2013 and 2022. Moreover, two-way investments between China and BRI countries have exceeded $380 billion (State Council, 2023). Despite these economic gains, concerns have been raised regarding the environmental impact of the BRI. No doubt, increased trade and investment can drive economic expansion, but if not managed effectively, they may also pose serious environmental risks (Zhou et al., 2024). As a result, the BRI has attracted global attention, making its long-term sustainability a critical area of discussion.

Foreign direct investment (FDI), a core channel of economic integration, generally brings threefold advantages for the host country: it generates employment, improves the effective use of natural resources, and enables the transfer of advanced technologies that are often inaccessible through trade or financial flows alone (Loungani and Razin, 2001). Many BRI member countries, primarily developing nations, actively seek FDI to expand economic activities and boost economic growth. However, studies show that countries with flexible environmental regulations are more likely to attract pollution-intensive FDI. As a result, economic progress may be achieved at the cost of environmental deterioration (Zhou et al., 2024; Mehmood, 2022b). Conversely, other studies suggest that FDI can introduce efficient technologies that drive industrialization and economic growth while maintaining a clean environment (Dilanchiev et al., 2024; Voumik and Ridwan, 2023). Given these contrasting perspectives, this study aims to examine whether FDI ultimately benefits or harms host BRI countries.

Furthermore, many countries overuse natural resources to produce goods for trade and stimulate economic growth, often surpassing the Earth’s capacity for natural replenishment. This excessive resource use leads to environmental degradation, such as loss of biodiversity, land degradation, deforestation, and pollution of water and air (Zhou et al., 2024). Research also shows that international trade positively affects carbon emissions, leading to increased environmental harm (Khan et al., 2022b; Jebli and Youssef, 2017). However, it can also enhance production efficiency by enabling countries to specialize based on comparative advantage (Tawiah et al., 2021; Suresh and Tiwari, 2018). Therefore, the effects of trade on the environment are complex and require further investigation.

Financial development is another key driver of economic growth that can expand economic activities (Nuţă et al., 2024) and facilitate investments in renewable energy sources. However, like other developments, financial development is a Faustian bargain, and the threat it poses to the environment should not be ignored (Liu and Chen, 2024; Hussain and Zhou, 2022). Financial development improves economic growth by easing consumer credit, allowing people to increase their demand for automobiles, household gadgets, and other technological products (Hussain and Zhou, 2022). This, in turn, stimulates growth in the transport and manufacturing sectors, leading to higher energy consumption and increased environmental pollution (Shahzad et al., 2017).

The growth strategies of most developing and emerging economies concentrate on infrastructure, construction, and manufacturing. For example, the BRI partner economies, which rely heavily on coal-based technologies for their growth trajectories, are projected to account for half of the global GDP by 2030 (Anwar et al., 2021). Such rapid industrial expansion is expected to drive up coal-based energy demand, with potentially serious consequences for the global environment (Lu et al., 2022). One way to safeguard the global environment, without compromising economic growth, is to invest in renewable energy sources (Lee et al., 2022).

Considered together, the aforementioned discussions make BRI recipient countries ideal for studying the relationship between economic growth, investment, and environmental sustainability. Although previous research has advanced our understanding of the connections among FDI, trade, and financial development and environmental quality, most studies have examined these factors separately. Although helpful, this fragmented approach leaves important gaps in the literature. This study aims to fill those gaps in four keyways. First, it offers a detailed multi-factor analysis of how FDI, trade, financial development, and renewable energy consumption influence environmental outcomes, particularly within the unique economic and policy contexts of BRI countries. Second, to the best of our knowledge, no previous study has attempted to differentiate between the direct and indirect effects of these macroeconomic variables on environmental quality. This paper addresses that gap by using economic growth as a natural framework to disentangle such effects. Third, unlike other recent studies, this paper employs advanced second-generation econometric models, including panel corrected standard errors (PCSE) and feasible generalized least squares (FGLS) models that account for serial correlation, heterogeneity, and cross-sectional dependence in the panel data, thereby providing more consistent and reliable results compared to the first-generation models. Lastly, the study proposes policy implications based on its empirical findings, which can inform effective policies to promote sustainable economic development and reduce carbon emissions within BRI nations. The rest of the paper is organized as follows: Section 2 reviews the relevant literature. Section 3 provides the theoretical foundation, data, and methodology, while Section 4 discusses the empirical results, and Section 5 concludes the study with policy implications.

In this section, we briefly outline the theoretical relationships between economic growth, FDI, trade, FD, and environmental quality, along with summarizing the associated empirical studies under separate titles.

The deterioration of environmental quality is a pressing global concern, and the world must address its causes and consequences. Environmental degradation threatens not only the ecological balance but also poses significant risks to human wellbeing and economic stability. While the drivers of environmental degradation vary across countries and regions–ranging from unsustainable resource exploitation to rapid economic expansion and weak environmental regulations–however, its impacts are universal and long-term. Many countries and regions have committed to achieve specific carbon emissions reduction targets. For example, the United Kingdom seeks to achieve net-zero carbon emissions by 2050, while China has set a goal of reaching carbon neutrality by 2060 (Akpanke et al., 2023). On the other hand, the European Green Deal projects a 55% decrease in carbon emissions by 2050. Additionally, the Paris Climate Agreement aims to decrease global warming to 2 °C or below, while the United Nations has called for a 7.6% carbon emissions reduction per annum between 2020 and 2030 (Akpanke et al., 2023). In the present study, we adopted a comprehensive mediating analysis approach to examine the relationship between FDI, trade, financial development, renewable energy consumption, and CO₂ emissions with economic growth acting as the mediating factor for the 26 BRI countries from 1995 to 2020. We used CO₂ emissions as a proxy for environmental degradation. Variables in absolute values are transformed into natural logarithms to reduce skewness and facilitate interpretation of the estimated coefficients as elasticities. We employed both FGLS and PCSEs methods to analyze the panel data, as these techniques yield robust results even in the presence of cross-sectional dependence, autocorrelation, and heteroscedasticity. We now proceed to interpret the empirical results and to make sure that the empirical findings are reliable and support the formulation of well-informed policy implications.

As illustrated in Table 8, the results from Model 1 revealed that FDI, TR and FD have a positive and statistically significant influence on economic growth across both FGLS and PCSE estimators. The coefficient for FDI indicates that a one percent increase in FDI boosted economic growth by 0.0172%–0.0882% at a 1% significance level, holding other variables constant. While the coefficient of FD shows that a one unit increase in FD leads to a 0.6093%–2.4423% increase in economic growth, holding other variables constant. Our findings are consistent with recent studies that suggest increased capital inflows and a well-functioning financial sector play a pivotal role in promoting economic growth (Solaymani and Montes, 2024; Nabi et al., 2023; Nguyen, 2022).

The estimated coefficient of TR is also positively associated with economic growth across both FGLS and PCSE regressions, reflecting that a 1% increase in trade openness raises economic growth by 0.001%–0011%, holding other variables constant. This result indicates that greater integration into global markets enhances economic activity in BRI countries by fostering exports opportunities and facilitating access to foreign markets. The result is consistent with Tawiah et al. (2021) and Suresh and Tiwari (2018) studies that suggest trade openness fosters efficiency gains and capital accumulation, thereby accelerating economic growth. Moreover, similar evidence from Nabi et al. (2023) confirms that trade openness supports economic growth by strengthening comparative advantage and expanding production capacity. Overall, the positive coefficient for trade in Model 1 underscores its role as a critical driver of economic growth in the BRI countries.

In contrast, the coefficient of REC is negative and significant at the 1% level, indicating that a percentage increase in renewable energy use results in between 0.0163% and 0.0194% decrease in economic growth, holding other variables constant. This finding may reflect transitional inefficiencies within the energy sectors of BRI countries. Several recent studies have evidenced similar findings that renewable energy consumption can hinder economic growth, especially when economies are undergoing a transition from fossil fuels consumption to renewable energy sources (Umar et al., 2024; Altıner and Şimşek, 2022; Feng and Zhao, 2022). Maji et al. (2019) argue that renewable energy use lowers total factor productivity, thereby slowing economic growth. Similarly, Solaymani and Montes (2024) attributed this negative relationship between economic growth and renewable energy to the fact that economic growth has been primarily driven by increased reliance on fossil fuel consumption. Additionally, fossil fuels, as conventional energy sources, are less compatible with the integration of renewable technologies. This structural dependence restricts the ability of renewables to contribute meaningfully to economic growth in the short term. Furthermore, the high installation and maintenance costs, along with the limited scalability of renewable energy technologies can also constrain its potential to create economic growth benefits (Dirma et al., 2024). The significant coefficient for FDI, TR, FD, and REC in Model 1 establishes the first stage of the mediated effect.

Considering the estimates from Models 2 and 3 in Table 8, the findings indicate a positive and statistically significant impact of FDI on CO₂ emissions in Model 2. According to the findings, a 1% increase in FDI results in a 0.0159%–0.0706% rise in CO₂ emissions at a 1% significance level, holding other variables constant. However, the coefficient of FDI significantly declines from 0.0159% to 0.0706% (Model 2) to 0.0061%–0.0320% (Model 3) when economic growth is introduced in Model 3, indicating a partial mediation. These results imply that FDI affects CO₂ emissions both directly and indirectly through economic growth, suggesting that as FDI inflow increases in BRI countries, so will carbon emissions. The findings are consistent with Model 1, which shows that FDI significantly stimulates economic growth. Our findings further highlight the dual nature of FDI in BRI countries. First, FDI may help these countries by bringing in capital that boosts output and infrastructure, thereby promoting economic growth. Second, due to flexible environmental rules and regulations aimed at attracting greater FDI, these countries may also facilitate the relocation of pollution-intensive industries. Consequently, our findings support the pollution-haven hypothesis in the context of BRI countries, suggesting that these countries may require stronger environmental regulations to attract more sustainable FDI. Our results are also consistent with other recent studies (Sreenu, 2025; Yousaf et al., 2024; Zhou et al., 2024; Khan et al., 2022a).

In the case of TR, it shows interesting findings across both Model 2 and 3. In Model 2 under FGLS, the coefficient is positive and statistically significant at 1% significance level. The coefficients indicate that a 1% increase in TR would lead to a 0.0003% increase in CO₂ emissions. However, while statistically significant, its sign changes in PCSE estimation. These findings are noteworthy, given that the environmental impacts of trade openness may operate through three core channels: scale effect, technology effect, and composition effect (Shahzad et al., 2017; Antweiler et al., 2001). The positive coefficient under FGLS suggests a scale effect: higher trade volumes expand production and energy consumption, thereby increasing CO₂ emissions. This aligns with the arguments that greater integration into global markets initially intensifies resource exploitation and industrial output (Khan et al., 2022b; Anwar et al., 2021). In contrast, the negative coefficient under PCSE indicates the possible presence of a technology effect, in which trade facilitates the transfer of cleaner technologies and efficiency gains that mitigate environmental harms (Khan et al., 2022a). In Model 3, when economic growth is introduced as a mediator, trade consistently shows a significant negative relationship with CO₂ emissions across both regressions. Since Model 2 captures the total effects of trade, whereas Model 3 separates the direct effect from the total effects that are mediated by economic growth. The positive scale-related impact of trade is transmitted through its growth-enhancing effect, and this part of the relationship becomes absorbed by the mediator. Consequently, the direct effect of trade represents the technology and composite channels, which contribute to lower emissions by supporting cleaner production processes and encouraging shifts toward less carbon-intensive sectors. This mediated formulation reveals why the coefficient in Model 3 becomes negative and stable under both FGLS and PCSE regressions and indicates that the environmental burden of trade declines when the growth-caused scale effects are separated. In other words, the technology and composition effects dominate the scale effect of trade volume when the growth pathway is controlled for. This highlights that trade openness not only fosters access to environmentally efficient technologies but may also encourage structural reallocation toward less carbon-intensive sectors. Thus, we concluded that trade openness negatively contributes to CO₂ emissions and improves environmental quality in BRI countries. Our findings are consistent with the studies by Zhou et al. (2024) and Tawiah et al. (2021), who reported that trade can help countries achieve efficiency in production by focusing on their comparative advantage, while, at the same time, they can also benefit from green technological transfers to mitigate environmental risks.

Financial development also shows a statistically significant positive impact on CO₂ emissions across both Model 2 and 3. According to Model 2’s findings, a unit increase in FD results in 0.4279%–1.1063% rise in CO₂ emissions (p < 0.01). However, once economic growth is introduced as a mediator in Model 3, the relationship between FD and CO₂ emissions weakens considerably. Under FGLS estimation, the FD coefficient remains positive, but its magnitude significantly declines, while under PCSE the coefficient becomes statistically insignificant, indicating partial mediation. These results further imply that FD affects CO₂ emissions both directly and indirectly through economic growth, suggesting that as FD increases in BRI countries, so will carbon emissions. This finding is supported by the results in Model 1, where FD has a positive impact on economic growth across both estimations. Hence, these findings underscore the dual nature of financial development. It catalyzes economic growth, but may also lead to environmental degradation. These findings further highlight that, in BRI countries with weak environmental regulations and poor enforcement, financial resources are often channeled into pollution-intensive sectors such as manufacturing, construction, and fossil-fuel-based energy production. Several previous studies, including Teklie and Yağmur (2024) for African countries, Shahzad et al. (2023) for the Association of Southeast Asia Nations, and Hussain and Zhou (2022) for BRI countries, have found similar results and concluded significant financial development impacts on CO₂ emissions.

Renewable energy consumption, on the other hand, shows a consistently negative and significant correlation with CO₂ emissions in both the total-effect analysis (Model 2) and mediation analysis (Model 3). It remains a highly significant mitigator of CO₂ emissions across all estimations. Given its negative impact on economic growth (Model 1), the effect of renewable energy on CO₂ emissions appears to be largely direct and independent, even after accounting for economic growth. The estimated coefficient for REC indicates that a 1% increase in renewable energy consumption reduces CO₂ emissions by 0.0382%–0.0434% in Model 2. The result is consistent with Model 3 estimations, indicating that the environmental benefits of renewable energy are structural and not mediated through economic growth. Furthermore, this finding suggests that shifting towards renewable energy sources can significantly reduce CO₂ emissions. Our findings are in line with several recent studies (Al-Zubairi et al., 2024; Yasin et al., 2024; Deka et al., 2023; Khan et al., 2022a), which suggest that renewable energy use can play a crucial role in reducing carbon emissions.

Regarding the impact of economic growth on CO₂ emissions in Model 3, the results demonstrated a positive and statistically significant relationship between economic growth and CO₂ emissions, confirming its role as a key mediator. This result establishes the second stage of the mediated effect. This condition indicates that the relationship between CO₂ emissions and the explanatory variables diminishes once economic growth is controlled for. The reduction in magnitude, rather than the complete disappearance, of the coefficients implies that economic growth absorbs part of the environmental impact of the explanatory variables. Therefore, satisfying the first three conditions provides evidence of partial mediation, suggesting the coexistence of both direct and growth mediated effects on CO₂ emissions. The significant coefficient of lnEG shows that a 1% increase in economic growth corresponds to a 0.4376%–0.5228% rise in CO₂ emissions, holding other variables constant. Thus, economic growth is the primary driver of CO₂ emissions in the sample of BRI countries, suggesting that the panel countries may not adequately consider their environmental quality in their growth strategies. Hence, the finding underscores the urgent need for policy initiatives to decouple economic growth from environmental degradation, thereby pursuing sustainable economic development and addressing environmental issues. The result is in line with the other recent studies (Al-Zubairi et al., 2024; Wang et al., 2024; Khan et al., 2022b; Nawaz et al., 2020) that also found a similar positive relationship between economic growth and CO₂ emissions. We also assessed the robustness of our results by re-estimating the empirical findings using Dynamic Ordinary Least Squares (DOLS) and Fully Modified Ordinary Least Squares (FMOLS), given in Table 9. All the results generally matched the FGLS and PCSE estimations in Table 8. Although there might be slight variations in the magnitude and statistical significance of the coefficients, the overall trends and implications remain consistent. Overall, our results underscore the urgent need for policy measures to decouple economic growth from environmental degradation, ensuring that economic progress in the panel BRI countries is not achieved at the expense of environmental quality.

Finally, we employed the Dumitrescu and Hurlin (2012) test to test the short-run association among variables. This approach comprises Z-bar and W-bar statistics. Table 10 presents the results of the Dumitrescu and Hurlin (D-H) causality test. The results show that bidirectional D-H causality runs between lnFDI and lnEG, between TR and lnEG, between FD and lnEG, between REC and lnEG. Similarly, a bidirectional D-H causality association is found between lnFDI and lnCE, between TR and lnCE, between FD and lnCE, between REC and lnCE, and between lnEG and lnCE. These significant relationships observed for lnFDI, TR, FD, and REC with respect to both lnEG and lnCE, suggesting that FDI, international trade, financial development, and renewable energy consumption contribute to the changes in economic growth and CO₂ emissions. Hence, these findings highlight a strong interdependence among economic, energy, and environmental indicators.

Global commerce (FDI and trade) and local financial development, when handled carefully, have been the primary drivers of economic growth in recent history. However, because most developing countries may not be sufficiently cautious about their environments, both global commerce and local financial development, if driven by profit maximizers, may lead to deteriorating environmental quality in such countries. Most BRI countries have experienced a surge in both FDI and trade flows recently, and, being predominantly developing and emerging economies that rely on coal-based technologies, their environmental sustainability may be at risk. Hence, this study examined how economic growth mediates the effects of financial and trade-related macroeconomic variables on CO₂ emissions in 26 BRI countries from 1995 to 2020. The primary contribution of this study lies in its mediation framework, as it introduces economic growth as an endogenous channel. This approach is grounded in the EKC and the pollution haven–halo hypotheses.

Employing robust FGLS and PCSE estimators, the findings revealed interesting and complex results. These findings provide policymakers in BRI countries with important insights. They highlight the need for a multidimensional policy approach that considers both the direct and indirect effects of macroeconomic variables on environmental quality. First, this study found significant cross-sectional dependence amongst the sample BRI countries. This indicates strong economic and environmental interdependence. Being tied to one another, BRI countries could benefit from improved environmental quality if cooperation to share best environmentally friendly practices is enhanced and cross-border environmental issues are jointly resolved. Second, FDI increases CO₂ emissions both directly and indirectly. This evidence supports the pollution haven hypothesis for the host BRI countries, suggesting that investment inflows may contribute to environmental degradation. Therefore, BRI countries need to consider environmental quality when evaluating long-term investments. This could include a proper mechanism to ensure that investment inflows align with their environmental goals.

Third, our findings indicate that FD indirectly drives CO₂ emissions through its impact on economic growth. This finding highlights the need to align financial sector expansion with environmental objectives. Governments and financial institutions should strengthen green financial architecture. Mechanisms such as green bonds, environment quality-linked lending, and climate risk assessments are essential. Further, redirecting capital toward sustainable investments requires a well-regulated and climate-conscious financial system. Hence, regulatory reforms aimed at enhancing the environmental accountability of financial flows are essential to shift capital flow toward sustainable investments projects.

Fourth, empirical evidence on trade suggests that trade openness does not directly influence CO₂ emissions. Instead, its effect is transmitted indirectly through economic growth. This suggests that the environmental benefits of trade are not inherent but contingent upon structural and policy factors. Therefore, policy interventions should not aim to restrict trade. Rather, they should focus on managing the nature and quality of trade-led growth. Furthermore, trade policy should be restructured to align with national climate goals by incorporating environmental impact assessments of trade flows. By managing the composition and quality of trade-driven growth, BRI countries can ensure that economic integration contributes to long-term environmental preservation and sustainable development.

Fifth, the brighter side of our study is that renewable energy consumption reduces CO₂ emissions, independent of the economic growth pathway. This indicates that increased use of renewable energy plays a significant role in reducing environmental degradation in BRI countries. Policymakers should foster both public and private investment in renewable energy infrastructure, to achieve a low-carbon future and promote environmental sustainability. Although the short-term impact of renewable energy on economic growth may appear negative, the long-term benefits are substantial. These include environmental protection, energy security, and sustainable development. Therefore, renewable energy should remain a strategic priority in the policy agendas of BRI countries.

Sixth, the findings confirmed that economic growth plays a significant mediating role in the relationship between FDI, FD, trade, and CO₂ emissions. Economic growth acts as a transmission mechanism through which these macroeconomic variables indirectly affect environmental quality. While economic progress is vital for development, its environmental cost–manifested through increased CO₂ emissions–must not be overlooked. Therefore, BRI countries must shift toward sustainable growth models that minimize environmental harm while sustaining developmental momentum. This includes integrating environmental criteria into macroeconomic planning, incentivizing resource-efficient industries, and promoting the adoption of low-emission technologies. Growth strategies should be supported by robust institutional frameworks that align investments and production with long-term environmental goals. By addressing the indirect environmental impacts of macroeconomic expansion, policymakers can decouple economic growth from CO₂ emissions. This can foster a development path that is both economically inclusive and environmentally sustainable.

Several limitations in this study may be addressed in future research. First, the study is limited to 26 BRI countries. Although these countries represent key emerging and transition economies from BRI partners, excluding other BRI participants may limit the generalizability of the results. In addition, the dataset covers data from 1995 period up to 2020, which is determined based on the availability of consistent data for the variables of interest across the selected BRI member countries. While the latest data is available for a few individual countries, post-2020 data for certain variables remains unavailable or incomplete for most BRI economies. This limits our ability to construct a balanced, methodologically reliable panel dataset beyond 2020. Given the substantial variation in economic structures, policy frameworks, and environmental regulations across the BRI countries, future research should expand both the dataset to include a broader range of BRI countries and the time period once harmonized post-2020 data become accessible. This would enhance empirical robustness and allow for more generalizability of the findings, while also capturing post-2020 developments. Second, the current study does not account for key structural and institutional variables. Factors such as technological innovation, institutional quality, and regulatory enforcement are not included. These may influence how FDI, trade, and financial development affect CO₂ emissions through economic growth. Their absence limits the study’s explanatory power. Including such moderating or control variables would strengthen the analytical depth of future studies and provide more targeted policy implications in BRI countries.