Does Green Finance Reduce Carbon Emissions? Global Evidence Based on System Generalized Method of Moments

Abstract

Global environmental awareness has increased with the adoption of the sustainable development goals (SDGs) and the Paris Agreement. Global climate change has been a focal point in these international frameworks, with an emphasis on addressing environmental issues and setting specific targets for solutions. Financing plays a vital role in attaining goals established in international frameworks. The limitations of conventional finance have highlighted the need for green finance. There is a scarcity of studies in the literature examining the impact of green finance on carbon emissions, and a definitive conclusion has yet to be reached. This research aims to enhance the current literature by presenting empirical findings on how green finance influences carbon emissions globally. By employing the system generalized method of moments (GMM), developed by Arellano and Bover (1995) and Blundell and Bond (1998), this study investigated how green finance influences carbon emissions. Covering the span of 2017–2022, this study encompassed 48 countries across the globe. Green finance was found to have a negative and statistically significant impact on carbon emissions. Issuing green bonds to represent green finance, when increased by 1%, reduces carbon emissions by −0.012%.

1. Introduction

The adoption of the sustainable development goals (SDGs) in 2015, along with introducing legal obligations on climate change through the Paris Climate Change Agreement, has significantly heightened global environmental consciousness. In 2019, the European Union unveiled the European Green Deal. International frameworks have placed considerable emphasis on addressing environmental issues caused by global climate change and have set specific targets to tackle these problems. The overarching goal of the Paris Climate Change Agreement is to ensure that global warming remains below the threshold of 2 degrees [1]. The European Green Deal aims to achieve carbon neutrality by 2050 and to transform Europe into the first carbon-neutral continent [2].

A crucial aspect of realizing the objectives outlined in international frameworks is the provision of financing. The promotion of green projects is essential for the attainment of set objectives. Despite this, green projects carry more inherent risks than their conventional counterparts. Furthermore, the outcome of these projects is uncertain. This situation has led to a significant disparity between the financial requirements of green projects and the current availability of financing. The insufficiency of conventional finance and financing instruments in addressing this gap has placed green finance and financing tools at the forefront [3]. Green finance, lacking a precise definition, can broadly encompass the financial instruments, approaches, and strategies employed to support environmentally focused projects, policies, and technologies aimed at the sustainable development goals (SDGs) [4,5,6,7]. The role of green finance is crucial in funding projects aimed at reducing carbon emissions and promoting environmental sustainability [8,9,10].

The concept of green finance covers many financial instruments, such as green bonds, green loans, green sukuk, and green securities. It is worth mentioning that green bonds have made significant progress in recent years among these instruments. Unlike conventional bonds, green bonds consider not only risk–return, but also actively contribute to reducing carbon emissions by incorporating environmental impacts. In this aspect, it distinguishes itself from conventional bonds [11]. Furthermore, green bonds impose more stringent disclosure obligations compared to conventional bonds. Therefore, investors have the opportunity to make environmentally friendly investments, as well as the option of pursuing investments with a low level of risk [12]. Although not obligatory, the issuance of green bonds takes into account prominent international standards, such as Green Bond Principles and Climate Bonds Standards [13]. By adhering to international standards, the issuance of green bonds mitigates specific issues, such as transaction costs, commonly encountered in financial market transactions [14]. Implementing these standards helps address the pressing issue of “greenwashing” when resources are shifted to the green economy [15]. Moreover, it promotes transparency and accountability [3].

The focus of this study is to investigate the effects of green finance on carbon emissions in a sample of 48 countries worldwide from 2017 to 2022. Given the global scope of environmental pollution caused by carbon emissions, it is imperative to conduct a comprehensive analysis on a global scale. In cases where T is small, and N is large, the utilization of global data is more suitable as a data source for the application of advanced panel data techniques. This study uses the system GMM estimator, developed by Arellano and Bover [16] and Blundell and Bond [17], known for its effectiveness in small-T and large-N scenarios [18]. Green finance is represented by green bonds, like several studies highlighted in the literature [19,20,21,22,23,24,25]. The STIRPAT theoretical framework is employed to determine the variables. This theoretical framework provides a more consistent approach to selecting variables. The selection of control variables varies among studies exploring this relationship, as the literature shows. Control variables such as GDP per capita, GDP growth rate, foreign direct investments, total population, trade openness, energy consumption, credit market development, and equity market development are included in the model in various studies [20,23,26,27]. As a result, variations can be observed in the obtained results.

This study provides three valuable contributions to the existing literature. Firstly, the existing literature on green finance primarily examines its effectiveness in renewable energy sources, as indicated by the limited number of studies conducted. While green finance is primarily used to finance renewable energy and energy efficiency initiatives, it also involves financing other environmentally friendly projects. These green projects cover pollution prevention and control, clean transportation, climate change adaptation, eco-efficient and adapted products, the environmentally sustainable management of natural resources and land use, terrestrial and aquatic biodiversity conservation, sustainable water and wastewater management, and green buildings [28,29,30]. The goal of these environmentally friendly initiatives, supported by green financing mechanisms, is to decrease the carbon emissions they release into the environment. Considering this, this research undertook a comprehensive investigation that specifically examined the effect of green finance on carbon emissions. Secondly, studies regarding the influence of green finance on carbon emissions are scarce in the literature, with no definitive findings. This study seeks to contribute value to the current body of literature in the relevant field and shed light on the global implications of green finance on carbon emissions through empirical analysis. Global analyses conducted by researchers [23,24] in the literature have included sub-analyses that specifically concentrate on developed and developing countries. However, when considering the worldwide issuance of green bonds and the implementation of green projects, it becomes apparent that the distinction between conventional developed and developing economies is insufficient for explanatory purposes. As a case in point, countries such as China, traditionally grouped under the developing country classification, display remarkably high levels of green bond issuance. Nonetheless, it is not feasible to carry out this assessment for every developing country. Hence, examining outcomes in developed and developing countries may lead to the adoption of misguided policy alternatives. Therefore, this differentiation will not be included in this study. Thirdly, this study employed the system GMM econometric method. By employing a dynamic model specification, this method effectively tackles the challenge of the endogeneity problem.

The rest of this study is organized as follows. The subsequent section entails the presentation of the literature review. In the third section, the data set and method are discussed. Interpreting the empirical results is the focus of the fourth section of this study. The conclusion section includes a comprehensive evaluation.

2. Literature Review

While there has been a noticeable rise in the number of studies conducted on green finance in recent years, the literature remains quite limited. The absence of a clearly defined green finance definition and the scarcity of data related to green finance are contributing factors. The studies mainly examined the impact of green finance on renewable energy consumption/production or carbon emission.

2.1. Green Finance and Renewable Energy

The impact of green finance on the consumption of renewable energy was examined in various studies. For instance, Dong et al. [31] conducted a study on six Southeast Asian countries (Indonesia, Malaysia, Philippines, Singapore, Thailand, and Vietnam), revealing that green bond issuance had a positive impact on renewable energy consumption from 2016 to 2021. In their study, Peng et al. [32] concluded a positive interdependence between developments in green finance and renewable energy consumption in China.

Several studies have investigated the influence of green bonds on specific renewable energy sources. The impact of green bonds on the consumption of wind, solar, and hydro energy in 15 OECD countries from 2010 to 2020 was analyzed in a study carried out by Wang and Taghizadeh-Hesary [33]. The research findings indicate that the issuance of green bonds positively affects wind and hydro energy consumption, with no significant impact on solar energy consumption. A further examination by Taghizadeh-Hesary et al. [34] focuses on the impact of green bonds on particular renewable energy sources in Japan. Research has shown that issuing green bonds positively impacts the long-term consumption of wind, solar, and hydro energy. Regarding the short-term period, it can be concluded that issuing green bonds has a positive influence on the consumption of solar and hydro energy, while its effect on wind energy consumption is not statistically significant.

Numerous research studies have analyzed the implications of green finance on generating renewable energy. Alharbi et al. [22] provide an example of this phenomenon by studying 44 countries across the globe and concluding that implementing green bonds stimulates renewable energy generation in both the short and long run. Filiz Baştürk [35] investigated this relationship, with a particular emphasis on EU member states. Research has shown that green bonds have a significant and positive effect on the production of renewable energy. The effects of green bonds on renewable energy supply were analyzed in a study conducted by Mavlutova et al. [36] in European OECD countries. The analysis suggested that the rise in green bond issuance leads to an expansion in the renewable energy supply.

2.2. Green Finance and Carbon Emissions

The existing literature contains scarce research on the influence of green finance on carbon emissions. Given the scarcity of research conducted on this topic, the literature offers no clear consensus on how green finance influences carbon emissions. Multiple studies have proven that green finance effectively decreases carbon emissions. Analysis on a global level has been conducted in some of these studies. In their research, Al Mamun et al. [20] investigated the link between green finance and carbon emissions in 46 countries. The research examining the utilization of green bonds to show green finance concluded that green finance effectively mitigates carbon emissions in both the short and long run. Alamgir and Cheng [26] examined 67 countries that issued green bonds globally between 2007 and 2021. The impact of green bonds on carbon emissions was found to be negative. When the entire period covered in the study was split into two periods, namely 2007–2014 and 2015–2021, it was found that green bonds did not significantly impact carbon emissions prior to 2015. The division of the period from 2007 to 2021 into two distinct periods, pre-2015 and post-2015, was attributed to the decisive impact of the Paris Climate Change Agreement. A worldwide investigation on the relationship in question was undertaken by Saha and Maji [23], encompassing 44 countries throughout 2016–2020. It was found that green bonds negatively affect carbon emissions. According to the study conducted by Shah, Murodova, and Khan [24], which encompassed 29 countries globally, it was concluded that green bonds contribute negatively to carbon emissions.

In some studies, a specific region or group of countries has been examined. For example, in the study by Khan et al. [27], the impact of green finance provided by the ADB (Asian Development Bank) on environmental quality in 26 countries in the Asian region between 2011 and 2019 was investigated. Within the scope of this research, the term “green finance” was explicitly defined as “climate mitigation finance” supplied by ADB. The research established that green finance plays a significant role in decreasing the ecological footprint, which serves as a measure of environmental quality. The study also analyzed the influence of green finance on carbon emissions and asserted that it effectively decreases them. In their research, Meo and Karim [19] examined the association between green finance and carbon emissions in the ten leading countries that promote green finance. The study revealed that green finance had an adverse effect on carbon emissions. Arshad et al. [25] examined the impact of green bonds on carbon dioxide emissions in developing countries and discovered a negative correlation. As stated in the research conducted by Jin et al. [37], the escalation in green bond issuance is expected to make a substantial contribution towards the accomplishment of carbon neutrality targets in 38 OECD countries.

When examining studies conducted within particular countries, the significant influence of China becomes apparent. As a case in point, the study conducted by Xu et al. [14] delved into the period from 2005 to 2019, specifically focusing on China. The study’s findings revealed that implementing green finance leads to a significant reduction in carbon emissions. Liu et al. [38] obtained a similar outcome. Su et al. [39] undertook another investigation on China. This study’s scope spanned 12 years, from 2008 to 2019, and included 30 provinces. Green finance proved to be an effective strategy for mitigating regional carbon emissions. Moreover, it was observed that the eastern regions experienced a noteworthy decline in carbon emissions because of green finance. Ran and Zhang [40] conducted a study encompassing 30 provinces in China. An examination of the period between 2005 and 2020 in this study led to the conclusion that green finance significantly reduced carbon emissions. In their investigation at the city level in China, Xu and Li [41] found evidence that implementing green bonds contributes to declining carbon emissions within cities. Liv et al. [42] explored the influence of green finance on carbon emission intensity and air pollution across 30 Chinese provinces. According to this study, green finance has proven effective in reducing both carbon emission intensity and air pollution. Another study conducted by Ma and Fei [43] in China at the provincial level revealed that green finance effectively decreases carbon emission intensity. Zhao et al. [44] focused on exploring the effects of green finance on corporate carbon emissions in Chinese provinces. The conclusion drawn was that, when examining the situation from a macroeconomic perspective, green finance has the potential to mitigate corporate carbon emission intensity through enhancements in industrial structure and advancements in technological innovation.

The findings of several studies suggest that green finance has negligible or no effect on carbon emissions. An example of this is the firm-level investigation by Ehlers et al. [45], which yielded no strong evidence suggesting that issuing green bonds contributes to reducing carbon intensity. In their study, Bukvic et al. [46] investigated the link between green bonds and carbon emissions, and found no significant correlation between the two variables. According to Sun’s [47] study on five Asian regions, it was determined that green finance did not exert a substantial impact on carbon dioxide emissions in Central Asia and South Asia, characterized by their heavy reliance on fossil fuels. Though, in the East Asia region, green finance was found to reduce carbon dioxide emissions in the short and long term. Sun et al.’s [48] study on China showed that implementing green finance reduces environmental pollution at the country level. However, an examination of the study at the regional level led to the conclusion that green finance reduced carbon dioxide emissions solely in the eastern and western regions. Findings differ across different regions, as indicated by observations made at the regional level.

Some studies have concluded that green finance affects carbon emissions in the long term, but that no relationship is valid in the short term. The work of Rasoulinezhad and Taghizadeh-Hesary [21] serves as an example of this. The study focused on the ten countries most supportive of green finance; it was determined that while green bonds have a long-term impact on reducing carbon emissions, no short-term causal relationship exists between the two variables.

Preventing the damage caused by carbon dioxide emissions to the environment is a general problem for all economies. The inadequacy of conventional finance and financing tools in solving the carbon emission problem has led to the development of green finance. Green finance is primarily used to finance renewable energy and energy efficiency projects; however, it also involves other environmentally friendly projects (such as clean transport and green buildings). It is crucial to recognize that green projects encompass diverse initiatives to reduce carbon emissions for the environment. Therefore, it is essential to evaluate the effectiveness of green finance on carbon emissions. Within the scope of this context, this study seeks to investigate the hypothesis stated below.

Hypothesis 1 (H1).

Green finance has a negative impact on carbon emissions.

3. Materials and Methods

3.1. Data Set and Model Variables

This study examined the influence of green finance on carbon emissions across 48 countries from 2017 to 2022 (the countries included in the analysis can be found in Appendix A). The inclusion of countries in the analysis was determined primarily based on data availability. According to the IMF’s classification, 27 of the countries within the scope of analysis are developed economies. The other countries included in the analysis, except Bermuda and Macao, are classified as emerging and developing countries by the IMF. These two countries are unclassified in the IMF classification. All variables used in this study, and the sources from which they were obtained, are given in

Table 1. Definitions of used variables in the analysis and source databases.

Variables	Symbol	Source
${CO}_2$ (tonnes per person)	ln${CO}_2$	Our World in Data
Greenbond	lnGB	Refinitiv (18 March 2024)
Real GDP (constant 2015 USD)	lnGDP	World Development Indicators (World Bank)
Population (total)	lnPOP	World Development Indicators (World Bank)

. The logarithmic form was employed for all the variables used.

The STIRPAT approach (stochastic impact of regression on population, affluence, and technology) was employed to determine the variables in this study. The initial version of this model, IPAT (environmental impact generated by population, affluence, and technology), was developed by Ehrlich and Holdren [49]. Dietz and Rosa [50] developed STIRPAT, the stochastic version of the IPAT model.

I = P × A × T

(1)

In the above equation, I represents the environmental impact. The total population is denoted as P, affluence or economic activity per capita as A, and technology as T. The stochastic version of Equation (1) can be shown as follows [50]:

$$I_i=\alpha P_i^b{A}_i^c{T}_i^d{e}_i$$

(2)

The natural logarithm of Equation (2) can be expressed in the following manner:

$$\mathrm{lnI}=\alpha +blnP+clnA+dlnT+e_{it}$$

(3)

In this study, the dependent variable is the environmental impact, measured by per capita carbon emissions (ln${CO}_2$). The issuance amount of the green bond (lnGB) shows the technology in the equation. Real GDP per capita (lnGDP) indicates affluence, while the total population (lnPOP) represents the size of the population. The model estimated in this study is presented below within this context.

$$\ln {CO}_{2it}={\beta }_0+{\beta }_{1}ln{GB}_{it}+{\beta }_2{lnGDP}_{it}+{\beta }_3{lnPOP}_{it}+e_{it}$$

(4)

The issuance amount of green bonds is expected to have a negative impact on the dependent variable of carbon emissions. Green bonds play a significant role in mitigating carbon emissions by supporting environmentally sustainable investment projects. Real GDP per capita, included as a control variable in this study, exhibits both positive and negative associations with carbon emissions. Based on positive impacts, an increase in real GDP per capita results in a greater allocation of financial resources towards green finance instruments. The expansion of green financing instruments plays a vital role in reducing carbon emissions by amplifying funding opportunities for environmentally friendly investments, particularly in the domains of renewable energy and energy efficiency. Further, the rise in real GDP per capita plays a role in the reduction of carbon emissions through the stimulation of market demand for environmentally friendly goods, such as green vehicles and sustainable buildings. Despite this, real GDP per capita increases also contribute negatively to carbon emissions. Additional energy consumption is a consequence of growth in real GDP per capita. This creates a higher demand for fossil resources [47,51]. The assessment of a population, as another control variable, is expected to show a positive impact on carbon emissions with population growth. As the population expands, there is a growing need for fossil energy, leading to a surge in carbon emissions.

3.2. Estimation Method

The system GMM estimator, developed by Arellano and Bover [16] and Blundell and Bond [17], is employed in this study to explain the effect of green finance on carbon emissions. The advantages of the system GMM estimator made it the preferred choice. First, including the lagged dependent variable as an explanatory variable in the panel data equation leads to correlation and endogeneity problems between $y_{i,t-1}$ and ${\epsilon }_{it}$. Inconsistency in OLS estimates arises because of this [52]. If the coefficient of the lagged variable is estimated with pooled OLS, it becomes upward biased. Estimating with the fixed effects estimator results in a downward bias [53]. The system GMM offers a dynamic model specification that effectively tackles the endogeneity problem. Second, the system GMM estimator has asymptotic efficiency gains over the difference GMM. Therefore, using the system GMM estimator improves accuracy and diminishes finite sample bias [54]. Third, the system GMM estimator is more efficient, as it includes additional moment conditions [52,55]. Fourth, the system GMM estimator expands upon the difference GMM estimator developed by Arellano and Bond [56]. By employing this approach, the differencing equations and the level equations are addressed simultaneously [18,57].

$$\begin{gathered} y_{it}=\alpha y_{i,t-1}+{x^{\prime }}_{it}\beta +{\epsilon }_{it} \\ {\epsilon }_{it}={\mu }_i+v_{it} \\ \mathrm{E}({\mu }_i)=\mathrm{E}(v_{it})=\mathrm{E}\left({\mu }_i{v}_{it}\right)=0 \end{gathered}$$

(5)

Here, i is the country, t is the time, $y_{it}$ is the dependent variable, $y_{i,t-1}$ is the lag dependent variable, ${x^{\prime }}_{it}$ is the vector of controls, and ${\epsilon }_{it}$ is the error term. The disturbance term comprises two components. One is fixed effects $({\mu }_i$); the other is idiosyncratic shocks ($v_{it})$. Equation (5) can be written as follows:

$$∆y_{it}=(\alpha -1)y_{i,t-1}+{x^{\prime }}_{it}\beta +{\epsilon }_{it}$$

(6)

4. Results and Discussions

The countries included in the analysis, which have the highest share of global carbon emissions as of 2022, are listed in Figure 1. The green bond issuance amounts of these countries for 2022 are shown in Figure 2. When Figure 2 is examined, it is seen that China made the highest amount of green bond issuance. Germany and the USA followed China. Issuing green bonds is more prevalent among developed countries with high carbon emissions, yet the remarkable green bond issuance by China, an emerging economy, should not be overlooked.

Descriptive statistics for the variables are presented in

Table 2. Descriptive statistics.

Variables	Mean	Std. Dev.	Min	Max	Observations	Groups
ln${CO}_2$	1.707	0.718	−0.630	3.251	288	48
lnGB	19.026	6.883	0	25.293	288	48
lnGDP	10.045	1.031	7.493	11.612	288	48
lnPOP	16.923	1.989	11.059	21.071	288	48

. An analysis of the descriptive statistics reveals that the average carbon emission for the countries within the scope of examination is 1.707 (min: −0.630–max: 3.251). The green bond issuance amount takes the value of 19,026, with an average of (min: 0–max: 25,293). The average real GDP per capita is observed to be 10,045 (minimum: 7493–maximum: 11,612), while the average for the population is 16,923 (minimum: 11,059–maximum: 21,071).

Table 3. Results of the model estimations, with the dependent variable being ln${CO}_2$.

Variable	FE	POLS	System GMM
ln${CO}_2$L1	0.127 * (0.074)	0.634 *** (0.062)	0.422 *** (0.102)
ln${CO}_2$L2	0.038 (0.077)	0.367 *** (0.063)	0.364 *** (0.097)
lnGB	0.001 (0.001)	−0.001 (0.001)	−0.012 ** (0.005)
lnGDP	0.392 *** (0.079)	−0.009 (0.009)	0.215 * (0.126)
lnPOP	−0.322 (0.566)	0.010 *** (0.003)	0.129 ** (0.052)
Constant	23.127 ** (11.360)	−43.246 *** (10.028)	−40.573 *** (13.719)
Diagnostic tests
AR(1) p-value			0.011
AR(2) p-value			0.519
Hansen Test p-value			0.110
Difference in Hansen Test			0.460
Number of Instruments			38
Number of Groups	48	48	48
Observation	192	192	192

Notes: 1. L1 and L2 show the first-order and second-order lag term. 2. The values in parentheses represent the robust standard errors in the system GMM models. *** p < 0.01, ** p < 0.05, * p < 0.10. to limit instrument proliferation “collapse” command used [18].

presents the findings from the one-step system GMM analysis. The results of the fixed effect model (FE) and POLS estimation can also be found in Table 3. System GMM estimation reveals that the lag of the dependent variable falls within the range of estimates from POLS and FE estimators. This situation confirms the accuracy of using the system GMM model. The lags of the first (ln${CO}_2$L1 (lnCO₂L1) and second (ln${CO}_2$L2 lnCO₂L2) order of the dependent variable are between 0 and 1, as expected, and are statistically significant. This shows that current carbon emissions are interconnected with past carbon emissions, exhibiting a dynamic framework.

In order to determine the consistency of the system GMM estimator, it is essential to perform specific diagnostic tests. Within this context, the initial analysis employed the Arellano-Bond AR(2) test to evaluate whether there is a serial correlation in the error term. The AR(2) serial correlation test showed no second-degree serial correlation. Next, the Hansen test of over-identification was conducted to assess the validity of the instrumental variables used. The null hypothesis regarding the over-identified restrictions being valid was not rejected. Finally, the difference in the Hansen test, which tests the validity of additional moment restrictions necessary for system GMM, was not rejected. According to the diagnostic test results, the system GMM estimator was consistent. The results are presented in Table 3. The xtabond2 command, developed by Roodman [18], was used in this study, and the analyses were conducted using Stata 16.

The system GMM estimation findings showed a statistically significant negative impact of green bond issuance on carbon emissions. The obtained outcome supports the research hypothesis. As expected, green finance has shown its effectiveness in reducing carbon emissions. Including real GDP per capita as a control variable in the model reveals a positive and statistically significant impact on carbon emissions. Regarding another control variable, it was found that the total population had a statistically significant and positive effect on carbon emissions. By including GDP per capita and total population as control variables, there was a rise in carbon emissions.

Several studies in the literature have also yielded results that support the findings of this study. For example, the studies conducted by the authors of [20,23,26], which analyzed countries globally, have concluded that green finance plays a role in reducing carbon emissions. Meo and Karim [19], who examined the ten countries with the highest support for green finance, revealed that green finance negatively influences carbon emissions. The research conducted in [14,38,39,40] focused on China and yielded the finding that green finance has a significant effect on mitigating carbon emissions.

Robustness Check

A robustness check was conducted in this study to assess the model’s reliability and the subsequent results. For this purpose, the two-step system GMM estimator was used. Arellano and Bond [56] stated that the standard deviations of the two-step estimation are biased. This study used the one-step system GMM as the fundamental model and employed the system GMM estimator to verify robustness. The findings of the two-step system GMM can be observed in

Table 4. Results of two-step system GMM, with the dependent variable being ln${CO}_2$.

Variable	System GMM Two-Step
ln${CO}_2$L1	0.418 *** (0.107)
ln${CO}_2$L2	0.366 *** (0.108)
lnGB	−0.013 ** (0.005)
lnGDP	0.230 * (0.134)
lnPOP	0.129 ** (0.051)
Constant	−37.910 ** (14.889)
Diagnostic tests
AR(1) p-value	0.016
AR(2) p-value	0.594
Hansen Test p-value	0.110
Difference in Hansen Test	0.460
Number of Instruments	38
Number of Groups	48
Observation	192

Notes: 1. L1 and L2 show the first-order and second-order lag term. 2. The values in parentheses represent the robust standard errors in the two-step system GMM models. *** p < 0.01, ** p < 0.05, * p < 0.10.

.

The outcomes of the two-step system GMM estimator were similar to the basic model estimated using the one-step system GMM. Issuing green bonds has a significant and negative statistical effect on carbon emissions. Including real GDP per capita and total population as control variables showed a positive and statistically significant impact on carbon emissions. When analyzing the results of the diagnostic tests conducted to assess the consistency of the two-step system GMM estimator, it becomes apparent that the AR(2) serial correlation test yields no evidence of second-degree serial correlation. Based on the outcomes of the Hansen test of over-identification, which aimed to assess the credibility of the instrumental variables utilized, the null hypothesis regarding the over-identified restrictions being valid was not rejected. In addition, the difference in the Hansen test, which tests the validity of additional moment restrictions that are necessary for system GMM, was not rejected.

5. Conclusions

The reduction of carbon emissions is essential in order to prevent environmental harm caused by global climate change. Because of this rationale, the primary focus of various economies is the shift towards a low-carbon economy. Financing plays a vital role in facilitating this transition. Given the limitations of conventional financing methods, the increasing popularity of green finance highlights its importance.

The primary focus of this study was to analyze the worldwide effects of green finance on carbon emissions, using the system GMM method proposed by Arellano and Bover [16] and Blundell and Bond [17]. This study encompassed 48 countries and revealed a statistically significant negative effect of green finance on carbon emissions. Including control variables such as real GDP per capita and total population in the model showed a statistically significant positive correlation with carbon emissions. In order to ensure the robustness of the findings, a two-step system GMM estimator was employed. The two-step system GMM results showed consistency with the outcomes derived from the basic model, specifically the one-step system GMM model. The impact of issuing green bonds on carbon emissions was both significant and negative.

This study’s outcomes have noteworthy policy implications. In the first place, it contributes to the realization of the Paris Climate Change Agreement’s aim to keep global warming below 2 degrees, a commitment embraced by various countries across the globe. In addition, it serves as a critical factor in supporting the attainment of net zero carbon targets for countries such as Turkey (with a target of 2053) and China (with a target of 2060), as well as regions like the EU. Hence, the continuation of government support for green bonds is essential, alongside introducing additional support mechanisms. In the second place, the ongoing significance of efforts to address global climate change will sustain the need for financial instruments that promote environmentally friendly initiatives, such as green bonds. This will play an essential role in deepening the green finance market, which has not yet reached sufficient depth, especially in emerging markets. In this context, government regulations will also contribute to deepening the green finance market.

Besides the advantages that green bonds provide worldwide, they also have other advantages, as Azhgaliyeva et al. [30] have stated. Green bonds provide various benefits for issuers, buyers, and governments. From a green bond issuer’s perspective, it enhances the institution’s reputation and enables them to achieve their sustainability targets successfully. The perspective of green bond buyers aligns with meeting the demands of investors who prioritize addressing the problem of global warming. When evaluated from a government perspective, it contributes to sustainable growth and energy supply. When considering the various benefits offered by green bonds, a fundamental financing instrument in green finance, it becomes apparent that their position among financing instruments will evolve from a newly introduced instrument to a market structure with increasing depth, rather than remaining confined to a narrow market.

Limitations and Future Research

This study also contains some limitations. First, the absence of a definitive definition of green finance has led to a lack of consensus among scholars on representing it. Green bonds are a commonly used means of symbolizing green finance. The growing body of research will lead to a consensus on this matter. Second, since green bond issuance amount data from 48 countries were available during the period examined, the analysis was conducted for 48 countries. In the coming years, this analysis will be expanded to cover more diverse countries, including those not present in the current analysis because of their lack of green bond issuance. Third, instead of approaching countries from a macro viewpoint, analysis can be performed at a micro level, mainly focusing on individual countries or firms.