EURASIA: Economics & Business, March 2025
by Andrean Eka Hardana, Zhou Jishun (School of Economics, Huazhong University of Science of Science and Technology, Wuhan, People's Republic of China)
Economic development that continues to pay attention to environmental sustainability has become a global challenge in addressing the increase in CO2 emissions. Thus, the realization of environmental sustainability efforts requires strategic steps through green finance from the dynamics of the economy for the environment. The impact of green finance becomes part of a non-linear investigation into CO2 emissions. The target area of the research is developed countries involved in the Belt and Road Initiative (BRI) over the period from 1990 to 2023. Based on Stochastic Impact through Regression on Population, Prosperity, and Technology (STIRPAT) using regression methods with models such as two-way fixed effects and the PFMOLS Model, applying DK because the panel data is also complemented with cross-sectional (CD). The empirical facts from the analysis conducted for developed countries show a significant relationship between green finance and CO2 emissions, which has an inelastic effect on environmental degradation. However, after reaching a certain threshold, it can gradually reduce environmental degradation. Additionally, there is a long-term equilibrium relationship between all explanatory variables and CO2 emissions. All existing findings serve as evidence of an inverted U-shaped relationship occurring in the developed countries of the BRI region. The existing findings have implications for policy implementation in economic growth to support the use of environmentally friendly energy by adopting and innovating to achieve environmental sustainability.
The global situations in the face of increasing climate change, many countries in the world have set goals to reduce carbon emissions and achieve environmental sustainability. On the other hand, the rapid development of global industrialization has resulted in environmental problems caused by the increasing consumption of fossil energy, which has gradually threatened the sustainable development of the environment. In recent years, the increase in carbon emissions has exceeded the carrying capacity of the environment. A series of climate change phenomena caused by global warming have had a wide impact due to the rapid progress of economic development by the financial sector that ignores the ecosystem (Muhammad et al, 2021).
Empirically, in facing the challenges of increasingly serious environmental degradation, countries are massively building sustainable environmental development policies. So that efforts to reduce carbon emissions through the implementation of policies that favor environmental sustainability as a preventive measure in managing energy consumption from sources with high carbon emissions (Abbasi et al., 2021). It aims to promote economic development and preserve the environment and resources that are essential for human health. However, it should also be noted that traditional environmental policies will restrict business behavior and are considered to cause an economic slowdown (Demirel and Kesidou, 2019). Meanwhile, in its development by policymakers because of its ability to promote economic growth and protect the environment to be a solution step at the same time. Governments around the world can use this concept's adoption and innovation measures aimed at making economies greener (Capasso et al., 2019).
To achieve the goal of building a green economy with carbon neutrality by 2060, it will be necessary to take solutions to transform the low-carbon development process. In this endeavour, green finance plays an important role. This is because green finance encompasses economic activities that support environmental sustainability in a focus on addressing climate change while providing effective and efficient resource utilization (Habib et al., 2024). Such activities involve the provision of financial services for project investment, project operations, and risk management activities across a range of sectors including environmental protection, clean energy conservation in the form of green transportation and green buildings (Khoury et al., 2024).
Green finance is not only an activity like traditional finance, but also the development of environmentally friendly energy conservation that can reduce carbon emissions. In its core function, the performance of green finance is not much different from traditional finance. Both are forms of financing based on market performance to allocate funds according to their terms. However, there are differences between the two, namely that traditional finance is based on the opinions of rational economic people, which can be interpreted as focusing on maximizing interests and prioritizing the goals achieved and minimizing risks. However, the problem is less addressed in resources and the environment (Chen et al, 2023). In contrast, green finance as a basic requirement is to consider the sustainability of environmental sustainability. Of course, the main goal is to achieve sustainable development between the financial industry and its socio-economy (Yuan et al., 2024).
The implementation of green finance in reducing the impact of global climate change in developed countries is generally in line with the provisions of carbon emission guarantors in the green finance market (Sartzetakis, 2021; Banga, 2019). In its application, green finance is also empirically a link between economic development in a region with attention to environmental sustainability, so that developed countries that are members of the Belt and Road Initiative (BRI) can provide a strategic role from the application of green finance to minimize carbon emissions caused by environmental degradation. Therefore, more attention is needed in carrying out further research with a focus on developed countries in the implementation of green finance to reduce carbon emissions.
In general, green finance studies have focused on developed BRI countries that have undergone a rapid transformation from ecological surplus to ecological deficit, largely due to the region's extraordinary growth over the past decade. In addition, BRI countries account for twenty-one percent of global GDP, as well as accounting for forty-one percent of the world's population with foreign exchange reserves of four trillion US$ (Ibrahim et al, 2023; Udeagha & Muchapondwa, 2023). In general, it can be interpreted as controlling a large part of the global economy. The resulting economic growth also causes the region to consume more than forty percent of global energy, making it a major contributor to global CO2 emissions (Danish and Wang, 2019).
Therefore, this study has an impact on environmental sustainability. To the best of our knowledge, this study is the first to investigate the non-linear effects of green finance on CO2 emissions in developed countries in the BRI region for the period 1990-2023. In addition, this green finance study reviews the EKC hypothesis and focuses on developed countries in the BRI Region, which can provide findings and policy implications for environmental sustainability. Based on these considerations, it motivates us to examine the dynamic impact of green finance in reducing CO2 emissions.
Implications that can be obtained from the study of green finance of developed countries in the BRI Region, especially for policy makers and governments. Efforts to realize environmental sustainability by implementing policies that can encourage the implementation of green finance. This is an effort to reduce the gap in green project investment activities (Bhutta et al., 2022). Based on after the review for the introduction, the next section is a literature review that will review related previous research on the topic under study, then on the research methodology which is explained in detail, the results and discussion that will be interpreted and analysed based on previous research. Finally, the conclusion summarizes the overall research and future directions.
Consistent efforts to preserve a country's environment are well underway, requiring a legal and economic balance that supports resource efficiency in management effectively and efficiently. These activities are part of the next generation getting the benefits of environmental sustainability from the challenges of environmental degradation.
To investigate the impact of green finance on CO2 emissions, we used STIRPAT (Stokastic Impact through Regression on Population, Prosperity, and Technology) developed by Dietz and Rosa (1997).
The STIRPAT model has also been further refined by Donglan et al., (2010). In addition, some researchers have developed this model by adding additional factors (Ma et al, 2017; Niu & Lekse, 2018).
Since our basic model covers the impact of green finance on CO2 emissions, we have modified and expanded the STIRPAT model in the following form: Our extended and modified STRIPAT model includes structural transformation and green finance variables as well as additional factors such as fertility rate, financial development and natural resources, which have been neglected by previous researchers.
Here CO2 emissions measure environmental impact, EI shows energy intensity which is a proxy for technology (T), Economic Growth per capita represents prosperity (A). Based on the studies of (Amna, 2024), (Fuxia, 2023), (Hong-Min 2022), Kwami and Samuel, (2022), for Economic Growth (EG), Green Finance (GF), Natural Resource Rent (NRR), Trade Openness (TO), and Financial Development (FD) variables are additional control variables. Lower energy intensity (EI) signifies greater use of green technologies, greater reliance on cleaner energy, and less consumption of primary energy sources (fossil fuel consumption). Therefore, previous studies have used EI as a proxy for the effect of technology on the environment (Nezahat and Mehmet, 2019; Jean et al, 2021; Daniel et al, 2022).
Previous studies by Oluyomi et al, (2020) and Malika and Samir (2023), have also successfully applied a non-linear version of the non-linear STIRPAT model to test the EKC hypothesis. Therefore, we include a squared EG term in Equation 4 to test the bell-shaped relationship between economic growth and environmental degradation.
Equation 4 tests the EKC hypothesis by including the squared term of EG. We expect an inverted U-shaped relationship between EG and CO2 emissions if the coefficient of EG is positive (>0) and the coefficient of its squared term is negative-lower than zero (<0). Coupled with the EKC hypothesis, the non-linear relationship between energy intensity and CO2 has been examined in equation 5 to test whether there has been a structural shift in the energy structure. In equation 5 for the energy intensity (EI) variable, an inverted U-shaped relationship is expected if the EI coefficient and its squared term are positive (>0) and negative (<0), respectively.
We have added the quadratic term of green finance (GF2) to equation (6) to prove the green finance theory, which states that green finance is considered as the main driver of environmental protection and sustainable development. One of the main roles in creating environmental sustainability is through the implementation of the green economy, which also includes budgets used specifically as environmental carrying capacity (Lee, 2020; Muhammad et al, 2021; Zhijuan et al, 2022; Julie, 2023). The expected result is a negative coefficient of GF so that the green finance theory can be accepted (𝛼6 < 0).
We have collected annual data for the period 1990-2023 from the World Development Indicators (WDI) database of the World Bank and the International Renewable Energy Agency (IRENA). As the dependent variable, carbon emissions (metric tons per capita) in our study. Meanwhile, there are Green Finance (GF), Energy Intensity (EI), Economic Growth (EG), Natural Resources Rents (NRR), Trade Openness (TO), Financial Development (FD) as independent variables. This research study uses panel data from 38 developed countries in the BRI Region. This research aims to examine the effect of green finance on CO2 emissions.
The data collection period starts from 1990 and ends in 2023 due to the availability of consistent data on all variables. Before applying panel unit root test and estimation procedure Before applying panel unit root test and estimation procedure, all variables have been transformed into logarithm form to reduce the possibility of econometric problems such as autocorrelation, heteroscedasticity, dimensionality of variables and improve the reliability of estimation (Hossain, 2011; Shahbaz et al., 2012; Wang et al., 2017).
Cross-sectional presence can make empirical results biased, false, and misleading. Previous research generally used Breusch and Pagan (1980) to test cross-sectional dependencies, but this method has some econometric problems. Therefore, Pesaran (2004) introduced stronger tests such as Cross-sectional Dependency (CD) and Langrage Multiplier (LM) to address the shortcomings of previous methods. The following equation presents the CD and LM tests.
The results of the cross-sectional dependency test are given in the table. Almost all show very significant values at a significance rate of 1 percent, which clearly indicates that our data has a cross-sectional dependence problem in terms of green finance and CO2 emissions.
Because our data has cross-sectional dependency problems, the unit root panel of the first generation of tests is inappropriate. Therefore, we have implemented the second generation of test root units proposed by Pesaran (2007) that consider cross-sectional dependence.
In equation (12), the null hypothesis of non-stationary is based on the OLS estimator to determine the order of integration relating to each series. Moreover, the t-statistic of CADF can be mathematically stated in equation.
The more specific case of the above-mentioned generalized form is stated in the following equation but requires simulation for the determination of critical values of Cross-sectional Im Pesaran Statistic (CIPS).
After examining the problem of cross-sectional dependency and unit root, the next step is to determine the co-integration relationship between series by applying the latest techniques from Westerlund (2007), which is a cointegration test based on error correction that allows cross-sectional dependence problems. One of the outstanding features of this method is that it is based on structural dynamics, not residual, and therefore not affected by unobserved factors (Tufail et al., 2021).
In this equation, 𝛼𝑖 determines the speed of adjustment at which short-run fluctuations in the model are restored to long-run equilibrium. Westerlund (2007) has developed four tests to determine co-integration.
If these two tests are statistically significant, we can reject the null hypothesis that there is a cointegrating relation among variables in the whole panel. The remaining two-panel statistics examine the existence of cointegrating relation in at least one country.
Panel co-integration testing only builds co-integration relationships between variables. However, the aim of this research is not only to determine the long-term impact of green finance on environmental degradation, but also to investigate its non-linear impact, which cannot be done by co-integration methods. For this purpose, we have implemented an advanced test of Dynamic Seemingly Unrelated Regression (DSUR) proposed by Nasreen et al., (2018). This technique is very flexible and not only addresses the endogeneity problem but also considers the heterogeneity of samples and cross-sectional dependencies that are not possible with the traditional method of ordinary smallest square (Haseeb et al., 2019). Following studies from Danish et al. (2019), we also apply the method of dynamic smallest quadrant (DOLS), and the smallest modified square regression square, as additional strength tests, which consider cross-sectional dependency and produce a strong standard error (Baloch et al., 2019).
Following the Danish et al. (2019) study, we also applied an AMG developed by Eberhardt and Teal (2010) to analyse the effects of the green finance path on environmental degradation for each country. AMG is An Autoregressive Distributive Lag (ARDL) panel model that is better than the first-generation ARDL panel technique because it allows cross-sectional dependency and sampling heterogeneity at the same time. This method combines a Common Dynamic Effect (CDE) into its two-stage process estimate for addressing problem problems of intersection dependencies. Furthermore, this method has no precondition for non-stationarity and co-integration between variables (Danish et al., 2019). Based on the main features of this AMG, the method is best suited to investigating the impact of green finance at the state level on environmental degradation.
The results and discussion section are organized systematically. First, we discuss the results of the panel cointegration test to establish the long-run relationship between CO2 emissions, green finance and other control variables. Second, this research study examines the non-linear effect of green finance on carbon emissions. Finally, there is a review of the description of the empirical results of the panel causality test to identify the intermediate relationship between the variables considered.
The study checks the stationary condition of the concerned variables dependent and independent such as Green Finance (GF), Energy Intensity (EI), Economic Growth (EG), Natural Resources Rents (NRR), Trade Openness (TO), Financial Development (FD) after determining. The presence of Cross-Section Dependence (CSD) and slope heterogeneity test.
The slope coefficient homogeneity and cross-sectional dependence test results from Pesaran (2007) are presented in Table 2. Based on our empirical results, we reject the null hypothesis of slope coefficient homogeneity and accept the alternative hypothesis indicating the existence of diversity and differences between cross-sectional units, in practical assessment terms, the findings for EI, GF, EG, NRR, FD and TO indicate the existence of slope coefficient diversity at various significance levels.
Panel unit-root results have been reported in Table 3. All variables contain unit root at level and become stationary at first difference as stated by values of CADF and CIPS.
As shown in table 3, as well, all variables have increasing trend and seem to exhibit the same order of integration, which is also confirmed by the panel unit results of LLC and IPS. If the variables are cointegrated at first difference, then panel cointegration tests such as Pedroni, Kao and Fisher can be applied to establish long-term association between variables.
Before examining the long-run non-linear effects of green finance on carbon emissions, this study determines the long-run equilibrium derived from the relationship between the variables considered. The results of Kao's panel cointegration test are reported in Table 4. The ADF test values are significant at the 1 percent significance level for developed countries in the BRI Region, indicating a long run cointegration relationship between green finance and CO2 emissions, energy intensity, trade openness, financial development and natural resources. The ADF test is also significant at the 1 percent critical value for developed countries. These results indicate the existence of long-run cointegration among the variables considered for the sub-sample.
In the next analysis, the non-linear impact of the economic growth (EG) variable together with its quadratic on CO2 emissions. In addition, explanatory variables such as energy intensity (EI), green finance (GF), natural resources rent (NRR), financial development (FD), trade openness (TO) are also investigated on CO2 emissions. In testing the EKC hypothesis, three panel regression models are used, namely two-way fixed effects, FMOLS and DK regression. Table 5 reports the results of EG and its explanatory variables.
Based on Table 5, the EG variable has a positive and significant impact on CO2 emissions. A 1 percent increase in EG can have an increasing impact of 0.16-0.27 percent on reducing environmental degradation. In line with the research of Shaari et al, (2020), for upper middle-income countries, economic growth does not significantly affect CO2 emissions. In high-income countries, economic growth can increase CO2 emissions in Canada and the United States, while it does not affect emissions in Poland, Belgium, and Saudi Arabia. Different from the results of the analysis of the squared term of EG, the results found a negative and significant effect on CO2 emissions. This means that for the squared EG variable, a 1 percent increase can lead to about 0.06-0.07 percent decrease in carbon emissions. These results show that there is an inverted U-shape of the effect of the EG variable on CO2 emissions, which further confirms the EKC hypothesis. The results of the study are in line with Khan et al, (2021), In high-income countries, economic growth drives carbon emissions, supporting the environmental Kuznets Curve hypothesis. However, the impact varies, suggesting that while growth may increase emissions, the relationship is not uniform across different income levels.
Based on these results, it can be interpreted that the EG variable initially causes an increase in CO2 emissions, which in turn impacts the decline in environmental quality. However, after reaching a certain threshold, it can reduce environmental degradation. This indicates that changes can be made in environmental sustainability policies. These findings are in line with the study by Yousaf et al. (2022), which shows a U-shaped relationship between economic growth and carbon footprint in high-income countries, indicating that while initial economic growth increases carbon emissions, further growth can lead to reductions through improved ecological efficiency and changes in energy consumption. Then, from the analysis results for all control variables starting from EI, GF, NRR, FD, and TO, positive and significant results were shown.
The results of the analysis of the explanatory variable for EI indicate that a 1 percent increase in EI results in an approximately 0.57-0.68 percent increase in CO2 emissions. Similar findings were reported by Alola & Joshua (2020), in their study which showed that fossil fuel consumption exacerbates environmental hazards in high and upper-middle-income countries, while the use of renewable energy has a positive impact on environmental quality in these economies, although the effects vary across different income classifications. Meanwhile, Tomić et al. (2022). This study shows that total primary energy supply has a positive impact on CO2 emissions, indicating that higher energy intensity in high-income and upper-middle-income countries can contribute to increased carbon emissions, limiting the prospects for green development.
The Green Finance (GF) variable can increase environmental degradation in countries targeting CO2 emissions. Based on the GF coefficient values, almost all are positive in all models, indicating that for every 1% increase in GF, there is an effect of approximately 1.14-2.30 percent in increasing environmental degradation. The results of this study are in line with Saha et al. (2024), which show that green finance significantly reduces carbon emissions in moderate emissions among high-income and upper-middle-income countries, while the impact is minimal for low-emission countries. However, even major emitters like China and Russia only show marginal effects.
The variable of natural resource rents (NRR) can degrade environmental quality in targeted countries through unsustainable natural resource management aimed at boosting economic growth. The results of this study are in line with Tufail et al. (2021), which state that natural resource rents have a positive impact on carbon emissions in high-income and upper-middle-income countries by increasing emissions. However, this study found that effective resource management, alongside fiscal decentralization, can lead to a reduction in CO2 emissions in these countries. Also supported by Lin et al. (2024) that natural resource rents positively affect CO2 emissions per capita in high-income and high-income countries, as shown by the research findings. This indicates that an increase in resource rents can lead to higher emissions, highlighting the need for effective environmental policies.
Furthermore, financial development and trade openness positively affect CO2 emissions for the overall panel in all six models; a 1 percent increase can cause approximately 0.01-0.70 and 0.01-0.24 percent increases in environmental degradation. These findings are in line with Cao et al. (2021), where financial development and trade openness in high-income OECD countries contribute to an increase in carbon dioxide emissions. This study shows that while globalization and institutional quality help reduce emissions, financial development and economic growth have a negative impact on environmental outcomes.
The examination of the relationship between energy intensity (EI) and CO2 emissions in this study also investigates an inverted U-shaped relationship for the target countries. In Table 6, based on the results presented, it can be seen that there is a non-linear impact of energy intensity on CO2 emissions. From the results of the EI coefficient values, it can be explained that there is an increase in CO2 emissions of about 0.31-0.40 percent due to a 1 percent increase in energy intensity. Unlike the results of the analysis for the square of EI, which shows an effect of about 0.08-0.12 percent reduction in environmental degradation.
Energy intensity positively and significantly influences the improvement of environmental quality in the long term. The results of this test are in line with Tudor et al. (2023). In high-income and high-income countries, structural transformation reduces carbon intensity, while economic growth significantly reduces carbon intensity. This study shows that these countries have effectively decoupled economic growth from carbon intensity, highlighting the importance of renewable energy in this context. As previously tested, as shown in Table 6, we found significant positive effects from other variables such as EG, GF, NRR, FD, and TO on carbon emissions.
Based on the analysis test for the EG variable, it shows a positive and significant effect on CO2 emissions. This can be interpreted that a 1 percent increase in EG can result in an increase of 0.35-0.61 percent in carbon emissions. These findings are supported by Khan et al. (2021), which show that economic growth drives carbon emissions in high and upper-middle-income countries, as evidenced by this research. However, the impact varies, with the Environmental Kuznets Curve hypothesis observed only in high-income countries, indicating a different relationship between growth and emissions across income levels.
The value of the GF coefficient also shows significant and positive results, indicating that a 1 percent increase causes approximately a 0.01-0.07 percent decrease in the quality of the generated environment. The results are consistent with previous research by Zhijuan et al. (2022), which found that green financing significantly reduces CO2 emissions in the long term by increasing the negative emission response to renewable energy investments. It is very important for policymakers to complement renewable energy investments with green financing to ensure environmental sustainability.
Meanwhile, the coefficient values of the NRR, FD, and TO variables also show significant and positive results, indicating that a 1 percent increase causes approximately 0.01-0.04, 0.05-0.06, and 0.01-0.02 percent increases in CO2 emissions. These results are supported by previous research by Khan et al. (2024), which shows that dependence on natural resources and trade openness increase per capita carbon emissions, while the development of Fintech can reduce emissions both directly and indirectly. However, the impact of renewable energy consumption on carbon emissions remains statistically inconclusive. In line with previous research, this paper focuses on the new BRICS economies, rather than high-income and high-income countries. It was found that natural resource rents and financial development significantly increase carbon emissions.
The subsequent investigation between the green finance (GF) variable and its quadratic to obtain an inverted U-shaped relationship in all models, namely the two-way fixed effects, PFMOLS Model, and DK. Testing was also conducted on all explanatory variables such as energy intensity (EI), Economic Growth (EG), Natural resource rent (NRR), Financial development (FD), and Trade openness (TO). Based on Table 7, the results of the investigation into the bell-shaped relationship between green finance and other explanatory variables can be determined.
The results of the test between GF and CO2 emissions across all models showed that a 1 percent variation in green finance causes an increase in environmental degradation by approximately 0.64-0.71 percent. However, the analysis of quadratic GF results in a 0.01-0.08 percent increase in environmental quality. Based on these results, it confirms the existence of an inverted U-shaped relationship between green finance and the increase in CO2 emissions for several BRI countries. These results are consistent with the research by Zhijuan et al. (2022), which states that green finance supports decarbonization through capital allocation and project screening, initially increasing CO2 emissions before curbing them after reaching a certain inflection point. This N-shaped relationship indicates that effective green financing can ultimately reduce CO2 emissions related to construction.
In addition, this research also investigates all the explanatory variables in Table 7, and all the current explanatory variables also find a positive and significant impact from EI, EG, NRR, FD, and TO. In general, all the results from these explanatory variables are similar to the results presented in the previous table. The regression results from all models indicate that each 1 percent increase in all explanatory variables causes an increase of 0.45-0.66%, 0.01-0.18%, 2.28-3.75%, 0.01-0.12%, and 0.02-0.06 percent in environmental degradation. The results are consistent with the research by Garda (2022), which focuses on the new BRICS economies, rather than high-income countries. It was found that energy consumption, economic growth, natural resource rents, and financial development have a positive impact on carbon emissions, while trade openness has a significant negative relationship with emissions. From the regression results, it can be interpreted that all explanatory variables have an impact on the decline in environmental quality.
An analysis of the non-linear impact of green finance on the overall panel has been tested for Developed Countries in the BRI group. In Table 8, the results of the non-linear green finance analysis for Developed Countries in the BRI group have been presented. This initial section discusses the results regarding the non-linear effects of energy intensity (EI) on CO2 emissions. From the analysis results, it was found that a 1% fluctuation in EI leads to an impact of approximately 0.316-1.17 percent increase in CO2 emissions, in contrast to the quadratic EI where each 1 percent change results in a decrease of 0.10-0.12 percent for environmental degradation. Thus, these findings support the EKC hypothesis and structural transformation regarding the inverted U-shaped relationship. High consumption of fossil fuels is part of high energy intensity. However, when a certain threshold is reached, there will be a decrease in energy intensity, which is made possible by the adoption and innovation of environmentally friendly technologies. This encourages the improvement of environmental quality in high-income countries. These findings are consistent with our research and align with the results of Sayed et al. (2018) and Eleni (2022). which explains that this study shows that in high-income countries such as the United States and Germany, energy intensity significantly affects CO2 emissions, with a positive contribution from income levels.
The EKC hypothesis testing was also conducted for the economic growth (EG) variable using the three PFMOLS models. The results of the analysis show that a 1 percent fluctuation in EG can increase environmental degradation by 0.79-2.01 percent. Meanwhile, a 1 percent quadratic fluctuation in EG can reduce CO2 emissions by approximately 0.04-0.06 percent. These findings show a bell-shaped relationship between EG and CO2 emissions, confirming the EKC hypothesis in the target countries. In general, people living in high-income countries have increased their self-awareness regarding environmental conditions. Various efforts are made by utilizing the adoption of environmentally friendly technological innovations, so that the community can care about and implement strict regulations. The findings are in line with previous research by Bo and Yayun (2022) and Xiaoyu (2023). This study shows that as economic growth increases, production-based CO2 emissions also rise. In addition, economic growth is also associated with increased emissions, highlighting the challenge of balancing economic development with environmental sustainability.
The current study also verifies an inverted U-shaped relationship between GF and CO2 emissions. Based on the PMOLS regression results in Table 8, it can be explained that a 1% variation in GF significantly increases CO2 emissions by approximately 0.06-1.14 percent, while the quadratic term shows a different result, which is negative and significant, indicating a decrease in CO2 emissions by about 0.02-0.11%. Based on these results, an inverted U-shaped relationship was found between GF and CO2 emissions for high-income countries. From these findings, it shows that GF has a positive impact on environmental quality. Green finance significantly reduces CO2 emissions, especially in high-income countries, by promoting sustainable construction practices and energy-efficient technologies, which ultimately leads to a lower carbon footprint. These results are also confirmed by previous studies by Mei-Hsin (2022) and Zhe et al. (2023) for high-income countries.
In other control variables, all have a positive and significant impact on environmental degradation. The results from the three PFMOLS models reported for NRR indicate that the abundance of natural resources can reduce environmental degradation in upper-middle-income countries. These results are in line with the research by Dong et al. (2022), which states that effective resource utilization can lead to emission reductions, promote environmental quality, and support sustainable economic development through innovative technologies and practices.
In addition, there is also a financial development variable that contributes to reducing environmental degradation. These findings confirm that FDV plays an important role in funding projects for various activities that support the sustainability of environmental conservation. This finding is supported by Duminda et al. (2023), who found a positive effect of financial development on environmental quality. Financial development in institutions negatively impacts CO2 intensity in developed countries, while access to financial institutions also has a negative impact. Additionally, financial stability in the market reduces CO2 intensity in these high-income countries, highlighting the complex relationship. Another effect, according to Bo et al. (2023), of financial development in high-income countries can lead to a reduction in CO2 emissions by facilitating investments in renewable energy and promoting sustainable practices, which ultimately contribute to a decrease in greenhouse gas emissions and improved environmental governance.
Trade openness has a positive impact on the increase in CO2 emissions, as open trade and financial development significantly reduce carbon emissions in high-income countries. The results are similar to the discussion in the previous table. The results of trade liberalization are in line with Yanzi (2023), where in high-income countries, trade openness generally correlates with lower carbon emission efficiency. This indicates a negative relationship between trade openness and carbon emissions in developed regions, including Europe and Oceania, where emissions management is more efficient compared to low-income regions.
The phenomenon of implementing green finance has become part of the challenge of global issues. Over the course of several decades, it has impacted environmental dynamics. Therefore, in this research study, a thorough investigation was conducted on the impact of green finance on CO2 emissions through the STIRPAT model using data from sixty-six countries that represent all variables and are consistent over the period 1990-2023. This study investigates the non-linear impact of energy intensity (EI), Economic Growth (EG), Green Finance (GF), Natural resource rent (NRR), Financial development (FD), and Trade openness (TO) along with their quadratic terms using the model.
In this study, because the panel data is also equipped with cross-sectional (CD), it also applies DK which functions to examine the validity of previous research results. Can provide strong estimates for econometric cases such as serial correlation, heteroskedasticity, and CD. This study also uses the D-H panel causality approach. The panel causality approach is used to determine the unilateral or bilateral causal relationship between variables. Based on the empirical analysis that has been conducted, there is a long-term equilibrium relationship between all explanatory variables and CO2 emissions. In addition, there is also a significant relationship between green finance and CO2 emissions. Based on the results obtained as evidence of the EKC hypothesis in developed countries in the BRI region.
Therefore, to support the EKC hypothesis, a non-linear investigation of energy intensity against CO2 emissions was also conducted. Based on the study conducted at the income level of developed countries, a U-shaped relationship was found between energy intensity and CO2 emissions. This finding emphasizes that there has been a specific adoption of environmentally friendly technological innovations in the long term to reduce environmental degradation.
Original paper, i.e. Figures, Tables, References, and Authors' Contacts available at http://econeurasia.com/issue-2025-03/article_03.pdf