Environmental implications of regional financial development on air pollution: evidence from European countries

This study focused on the spatial effects of CO2 determinants, using data from 43 European countries between 1996 and 2018. Tests proved the existence of the spatial Durbin model. The findings indicate that the logarithm of GDP per capita, urbanization, and energy use have enhancing effects on CO2 emissions, but that these effects are meaningless for trade openness. Most of the coefficients provided insignificant results, while six financial development indicators were entered into the model in a straight line. The addition of the term "interaction between energy intensity and financial development" improved the effects of all of the components significantly. The phrase's negative and significant coefficient predicts increased energy intensity efficiency as a result of the development of multiple financial production components. The analysis of spatial effects reveals that control variables in adjacent countries have negligible effects on CO2 emissions. Overall, the findings show that CO2 emissions are falling as neighboring countries' productivity and financial development rise. The findings indicate that financial development in neighboring countries has comparable effects on CO2 emissions as domestic financial development. As a result of these findings, European countries have reached a point of financial integration.


Introduction
Natural change is tended to by air pollution, deforestation, depletion of underground water tables, and an increase of global temperature as a result of the extension and accumulation of ozone-depleting substances (GHGs) in the climate (Nourry, 2007;Zhang et al., 2018), and has become a general test that desperately needs to be tended to, bringing in the consideration of environmental sanatorium CO2, methane, and nitrous oxide are common ozone-depleting gases emitted by mechanical operations such as energy-intensive businesses, power plants, and transportation. Carbon dioxide (CO2) emissions account for around 75% of ozone-depleting substance discharges (Abbasi & Riaz, 2016), with global temperatures reaching 1.5 degrees Celsius, which is very high.
The study finds that economic growth stimulates monetary growth, which helps oil premium (Giannetti et al., 2010;Gunasekaran et al., 2014), resulting in increased CO2 emissions (Islam, et al., 2013;Le et al., 2020;Sadorsky, 2010;Tang & Tan, 2015). Moneyrelated development lowers credit targets by lowering acquisition prices, easing liquidity conditions for announced projects, and allowing them to increase yield, allowing monetary respect to grow, resulting in increased energy consumption and CO2 emissions. Furthermore, there is a strong link between monetary growth and environmental degradation, implying that financial development is associated with an initial increase in CO2 emissions, followed by a decline as the economy grows, resulting in an environmental Kuznets curve (EKC) with a switched U-shape (Grossman & Krueger, 1991;Orubu & Omotor, 2011). According to theoretical methods, there are two ways to understand the connection between money-related activities and monetary turn of events: first, the monetary transition territory is triggered by financial development. Furthermore, money-related movement is a critical system for financial development (Goldsmith, 1969;Leahy et al., 2001). Most experts, on the other hand, agree that while financial development is essential for mechanical progress, it also allows businesses and governments to obtain environmentally beneficial inventions. Furthermore, economic growth stimulates interest in energy-productive innovations that are capable of reducing the negative impact of petroleum derivatives, thus enhancing the atmosphere or climate (Acheampong, 2019;Jalil & Feridun, 2011;Tang & Tan, 2015). The financial sector provides investments that are vital for growth but also for adopting energy-conserving, green technologies, research, and development (R&D) (see Amazian et al., 2009;Zagorchev et al., 2011) and is one of the main mechanisms to transfer technology promoting CO2 emissions through the aiding of manufacturing activities. On the other hand, financial development affords sectors and governments to assume environmentally efficient technologies that are capable of reducing CO2 emissions and could motivate firms to adopt environmentally sustainable projects, which reduce carbon emissions (Claessens, 2007;Tamazian and Bhaskara Rao, 2010;Acheampong et al., 2020). Such contradictory effects lead to the influencing dimensions of financial development on environmental quality to be one of the challenging issues in experimental studies.
Traditional evaluation models often ignore spatial characteristics, resulting in model frustration, and they fail to consider spatial dependence, which leads to model dissatisfaction. As per spatial dependence, one discernment in a set of cross-sectional observations is reliant on other cross-sectional discernments. Furthermore, the growth and ecology of a large nation would have an effect on the environmental ideas of neighboring countries. A high-performing nation will have an effect on its neighboring nations and districts along these lines. Since traditional board econometric approaches ignore spatial relationships and fail to capture the circuitous (impacts caused by nearby) and spatial flood impacts of financial transition on CO2 emissions, spatial econometric models are more valuable and capable (Meng et al., 2017;You & Lv, 2018). We expected a halfway or even unequal evaluation because the vast majority of studies in this field use a standard and clear board information examination that ignores spatial dependence within the data (Zhao et al., 2019;Lv & LI, 2021). Previous studies have focused on the immediate effects of financial development on CO2 outflows, ignoring its flood sway on CO2 emissions-where the flood sway (i.e., indirect influence) refers to the effect caused by neighboring countries' financial shifts.
This investigation adds a number of new obligations to the composition's financial and environmental quality: to the best of the authors' knowledge, only a few analysts discuss the spatial characteristics of CO2 emissions, and this is the first paper to consider a local approach to managing analyses of the impact of financial development on CO2 outpourings, as when inspecting CO 2 emissions and regional financial development, there is a likely spatial dependency, thus neglecting its spillover effect on CO2 emissions, where the spillover effect (i.e., indirect effect) relates to the influence generated by the financial development of neighboring countries on the other hand, most of the studies in this topic use a conventional and basic panel data analysis that has ignored spatial dependence within the data, and these traditional analysis models often oversee spatial characteristics, failing traditional models, and finding by this method would fill a gap in related research and help foster cross-border cooperation on CO2 emissions control among countries. Also, permitting to Anselin (1988), there is a spatial correlation among the units of study due to the gravitational effect, so we anticipated that results in partial or even biased estimation so that in this exploration we examine the relationship among financial development and CO2 outpourings by spatial econometrics approach by entering a board educational collocation. Second, we use a variety of financial movement pointers to capture the various heterogeneous effects of the monetary influence on CO2 emissions.
The following section discusses how the rest of the article is put together: Sect. 2 recalls a writing study on the effect of monetary growth on CO2 emissions; Sect. 3 depicts the evidence test and trial templates used; Sect. 4 presents scientific findings, and Sect. 5 concludes the investigation and provides several policy recommendations.

Literature review
The relationship between monetary changes and CO2 emissions is gaining traction, and this section summarizes some key experimental results on the impact of monetary improvement on CO2 emissions or natural pollution. Furthermore, the earlier tests have an almost similar appearance. It is because they ignored spatial dependency within the data and countries using standard board information and other econometrics methods. We accept that no country is disengaged, so spatial econometric models should be considered where there is spatial dependency across locales to prevent one-sided observational outcomes. Yuxiang and Chen (2011) used GMM to examine the impact of monetary development on fossil-fuel by-products in China, and their results revealed that monetary progress, as determined by the proportion of bank credits to GDP, the proportion of private advances 1 3 to GDP, and the proportion of non-private to GDP, decreases carbon emissions' influence. Jalil and Feridun (2011) used the autoregressive distributed lag (ARDL) approach to find that monetary growth decreases the volume of fossil fuel by-products in Pakistan, as determined by the proportion of liquid liabilities and private region advances to GDP.
According to Al-mulali and Sab (2012), energy use played a significant role in improving both monetary development and monetary turn of events while also generating significant emissions. The results revealed that by increasing energy productivity, countries could improve their energy profitability. To achieve monetary changes and GDP growth in Sub-Saharan African countries, this can be refined through the introduction of oil-investment funds programs, energy stability, and energy base shifting. Furthermore, Shahbaz, Hye, et al. (2013) investigated the association between monetary activities and fossil fuel byproducts in Indonesia using the ARDL and Granger causality studies. Their results showed that monetary reform is expected to result in climate-friendly creativity, which decreases CO2 emissions and improves ecological productivity. In addition, Shahbaz, Solarin, et al. (2013) used the limits checking approach to examine cointegration between monetary events and CO2 outflows in Malaysia and found that there have been substantial run ties between CO2 discharges, monetary events, energy use, and financial development for a long time. It also shows that monetary advancement lowers CO2 emissions. Increased electricity consumption and monetary inflation worsen CO2 pollution. From 1989 to 2011, Ziaei (2015) looked at the effects of monetary pointer stuns (credit market and financial exchange stun) on energy consumption and carbon emissions for 13 European and 12 East Asian-Pacific countries, finding that stock return rate stuns have an impact on energy consumption, particularly in long-skyline scenarios with East Asian-Pacific countries present. Similarly, Al-Mulali et al. (2015) used cointegration tests and the fully modified ordinary least square (FMOLS) approach to find that monetary growth assessed by domestic credit to the private sector raises carbon emissions in 129 countries. Abbasi and Riaz (2016) also used the ARDL and vector autoregression (VAR) techniques to show that monetary development in Pakistan resulted in lower CO2 emissions. To address the monetary turn of events, they used all-out credit, private area credit, bond market capitalization, and exchanged securities.
Plus, et al. (2018) used a metric of homegrown recognition of private space as an intermediary for the monetary growth of BRIC economies and then used the Granger causality test to determine if there was two-path causality between it and CO2 pollution. They have made it so that economic growth contributes to CO2 emissions. By extending homegrown credit to the private sector to repair monetary events, Ahmad et al. (2018) obtained the same outcome. Using the ARDL and error rectification model (ECM) techniques, they discovered that monetary growth animated the fossil fuel by-product in China. From 1990 to 2014, Ehigiamusoe and Lean (2019) examined the impact of monetary change on carbon emissions in 122 nations, finding that monetary progress demolished fossil fuel by-products over time. In either case, industrial growth reduced fossil fuel by-products in highincome countries while growing them in low-and middle-income countries, according to the report. In addition, Charfeddine and Kahia (2019) analyze 24 MENA countries to ascertain the causal relationship between environmentally friendly energy use and monetary activities, as well as CO2 emissions and financial development. The study's findings, which were obtained using the board vector autoregressive approach, indicate that monetary activities and the use of renewable energy have close ties to CO2 emissions. The authors of Wang et al. (2019) investigate the causes of urbanization, monetary activities, population development, and innovation, as well as their links to CO2 emissions. They discovered that fossil fuel by-products had significant positive correlations with all of the variables (urbanization, economic progress, population growth, and innovation). Finally, there has been changing throughout monetary intermediation. Acheampong et al. (2020) used the instrumental variable summed up technique for the second approach to analyze the effect of monetary market growth on the fossil fuel byproduct power in 83 countries from 1980 to 2015. Their results showed that in both industrialized and emerging monetary economies, monetary market growth and its sub-measures, such as money market profundity and proficiency, decrease the impact of fossil fuel by-products. Zhao and Yang (2020) investigate the relationship between monetary development and CO2 outflows at the national level in China using static and dynamic analysis. The unexpected results indicate that territorial monetary advancement has had no effect on CO2 emissions. To research the impact of monetary change on CO2 pollution, Lv and Li (2021) used aboard information spatial econometric technique for 97 countries from 2000 to 2014 and discovered that there is a spatial association between CO2 discharges through nations during this period. They also showed that a country's CO2 emissions can be influenced by its neighbors' monetary progress. Khezri et al. (2021) looked at the spatial association of CO2 determinants. They discovered that as energy efficiency improved, each of the six monetary growth metrics became more important, resulting in an increase in CO2 outflows, despite their unfavorable overflow effects. According to the findings, CO2 outflows decrease as interest and monetary growth in neighboring countries rises.
Traditionally, many previous studies utilized traditional methods such as ordinary least squares (OLS) and generalized method of moment (GMM) cannot entirely overcome the problems caused by spatial dependence between attributes by controlling the fixed effects and the fact that spatial dependence was not taken into account (Anselin, 2010). In addition, its spillover effect on CO2 emissions is an indirect effect where refers to the influence caused by nearby countries' financial development Therefore, in response to these limitations, the present study attempts to employ spatial econometric techniques to explore the effect of financial development on CO2 emissions.

Empirical model
In this research, the logarithm of the carbon emission ( lnCO 2 ) is measured to be a purpose of some illuminating variables counting the logarithm of GDP per capita ( lnGDPP ), the rectangle form of GDP per capita ( lnGDPP 2 ), urbanization ( lnURB ), trade openness ( lnOPE ), energy intensity ( lnENER ), and financial development ( lnFD ) so that the experimentally model of CO2 emission model is as follows: When we regard monetary growth as a free factor, the ecological efficiency is turned U-shaped, so the negative coefficient of the squared kind of gross domestic product per capita in the CO2 emanation situation is hypothetically discussed and should be investigated, according to the EKC conjecture. In either case, as the economy develops, the condition of the atmosphere first deteriorates and then changes (Grossman & Krueger, 1995;Lee et al., 2010). Other factors such as urbanization, energy force, and trade openness are (1) often used as illustrative factors for CO2 emissions in the literature (Acheampong, 2019;Chakravarty & Tavoni, 2013;Epule et al., 2012;Kayani et al., 2020;Solarin et al., 2017).
To surface the different accepts of the effectiveness of financial development on CO2 emission from depth, the interaction terms of energy intensity and financial development are entered in the new form of the CO2 emission model of Eq. 2, where (lnFD it × lnENER it ) shows the interaction term, while the coefficient for (lnFD it × lnENER it ) indicates the relationship between financial development and energy use. On the one hand, financial development will reduce CO2 emissions by encouraging firms to adopt environmentally friendly technologies (see Tamazian et al., 2009;Tamazian and Bhaskara Rao, 2010;Zagorchev et al., 2011), but on the other hand, improved financial sector leads to cheaper access to credit for the purchase of new machinery and equipment (Acheampong, 2019;Sadorsky, 2010Sadorsky, , 2011. The following is how energy intensity impacts carbon emissions: Higher energy intensity is supposed to have a positive impact on CO2 emissions because energy intensity is a metric of energy quality, and a higher value of this index equals more CO2 emissions, so coefficient 6 should be positive. However, if the financial market develops to stimulate pro-environmental infrastructure, coefficient 8 is negative, and the energy intensity's initial positive effects are diminishing. The consequences of CO2 emission determinants are investigated using a spatial econometric model, with a focus on financial development metrics. A spatial panel model could have a lagged dependent variable or adopt a spatially autoregressive mechanism in the error word, according to Anselin et al (2008). The spatial Durbin model, which involves spatially lagged independent variables, was developed by LeSage and Pace (2009). The spatial lag model, the spatial error model, and the spatial Durbin model are all written as follows: in which y it represents a dependent variable for cross-sectional unit i = 1, 2, …N at time t = 1, 2, …. T . Also, x it stands for a 1 × K vector of exogenous variables, while represents a K × 1 vector of parameters. It should be noted that ∑ N j=1 w ij y jt accounts for the interaction effects of dependent variables in the adjacent units on the dependent one, w ij denotes element i, j of an N × N matrix of spatial weights, denotes the endogenous interaction effect response parameter, it stands for an error term of independent and identical distribution, c i is a spatial particular effect, and t accounts for the time-period particular effect. A spatial particular effect accounts for all time-invariant space-specific variables, the absence of which would lead to skewed estimates in a typical cross-sectional study. A time-periodspecific effect, on the other hand, accounts for all time-specific effects, the exclusion of which could lead to skewed estimates in common time-series analysis (Baltagi, 2005). Unit i error word in spatial error model Eq. (3) (i.e., u it = ∑ N j=1 w ij u jt + it ) and centered on matrix W and an idiosyncratic component it, is considered to be reliant on the error terms of adjacent units j. Furthermore, LeSage and Pace (2009) suggested that the spatial Durbin model in Eq. (4) be used. It will add individual spatial lag variables to the spatial lag model, where is a K × 1 vector of parameters.

Data
To analyze the effects of CO2 emission determinants and conduct an experimental analysis, data from 43 European countries are compiled from 1996 to 2018. Figure 1 depicts a comparison of CO2 emissions in the countries under consideration. Since all of the factors are in logarithms, the approximate coefficients are elasticity. Table 1 shows a list of the constructed variables used in the study.
We need more exact test measurements to show the spatial impacts. Moran's I is more remarks insights for this reason in Fig. 2. A positive Moran's I addresses to exhibit the spatial outcomes; we need more exact test insights. Figure 2 shows Moran's I am more remarks measurements thus. A positive Moran's I esteem shows the spatial amassing of comparative qualities in the field, while a negative worth demonstrates the spatial collection of non-comparative qualities. Most nations have a positive autocorrelation, while others have a negative; however, the principle autocorrelation, as seen by the fitting line, is positive. As indicated by Moran's I figure, adjoining nations have more practically identical CO2 outflow per capita, monetary organization (FI) development, and monetary market (FM) advancement proportions. To research the effect of CO2 discharge determinants, spatial econometric models are utilized.
The three main sources of our data include International Energy Agency (IEA), the World Development Indicator (WDI), and International Monetary Fund (IMF). Table 2 also provides access to the data's summary statistics which are calculated in EViews software. Furthermore, variables are in natural log. The subscript "i" indicates the country for There are a total of 989 observations. Among all the 11 constructed variables, the highest variation exists in the development of financial market efficiency. The mean value of energy intensity among all included variables is the lowest, whereas the highest average value is for GDP.

Results and discussion
Two autonomous logarithm likelihood (LR) investigations are utilized to inspect the probability of the presence of time span fixed impacts and spatial fixed impacts in the model. The model with synchronous spatial and time-frame fixed impacts is against models with time-frame fixed impacts and additionally models with spatial fixed impacts hence. The model of concurrent spatial and time span fixed impacts is picked if the invalid theory is rejected, and the subsequent model is picked if the invalid speculation is acknowledged. Table 3 shows the LR test insights for each model (3). The test outcomes show that the LR test figures are critical and that the invalid speculation is dismissed for all models. Subsequently, the model of covering spatial and time span fixed impacts is the better model for proceeding onward with the assessment technique in these situations. Another evaluation, seen in Tables 4 and 5, looks at whether using the spatial lag or spatial error in the model with no spatial interaction effects improves the model significantly. LM experiments for a spatially lagged dependent variable and spatial error autoregressive models are used for this purpose, using the residuals of a non-spatial model (Elhorst et al. 2010). The test statistic is based on the Chi-square distribution. The existence of the spatial lagged model and the spatial error model would be verified if the null hypothesis of the LM test is dismissed. We only consider the Lagrange multiplier (LM) statistics for this model since the results of the LR test verified the presence of the model with simultaneous spatial and time-period fixed effects. The test results indicate that the sum of test statistics is meaningful at the 1% level in Tables 4 and 5, indicating that the presence of the spatial lagged in all models and the spatial error for the majority of models is not ruled out. As a result, the inclusion of spatial interaction effects in the model highlights the importance of including such effects in laboratory experiments to explore the factors causing CO2 emissions. Table 6 shows the findings of the Hausman test, which was used to see whether the fixed effects model should be replaced with a random effects model. In this test, the null hypothesis stresses the presence of random effects in the model. The Hausman test results reveal that the presumption of random effects in the spatial lag model is dismissed for all simulations, whereas the presence of fixed effects is verified at a 1% significance stage.
Finally, we examine two separate hypotheses H 0 ∶ = 0 and H 0 ∶ + = 0 in Eq. (3). The spatial Durbin model is simplified to the spatial lag model if the first  Table 7 for fixed effects models. On both models, the statistical significance of the two experiments, the LR or the Wald test, is important, and the spatial Durbin model cannot be transformed to a spatial error or spatial lag model. As a result, the presence of the spatial lagged independent variable is established, and the spatial Durbin model is used to analyze the estimation results.   Most of the model variables had a major impact on CO2 emissions, according to the coefficients of the model variables in Table 8, and CO2 emissions increased by around 1.45 percent with every percent growth in GDP per capita. However, the GDP per capita coefficient of the squared form is negative, illuminating the EKC hypothesis and resulting in an inverted U-shaped association between GDP growth and CO2 emissions. In addition, each percent increase in urbanization results in a 0.7 percent increase in CO2 emissions. While trade openness has a positive impact on CO2 emissions, most simulations do not find this effect to be significant. The energy intensity has a favorable impact on CO2 emissions, with each percentage increase in energy intensity resulting in an increase in CO2 emissions by around 0.88 percent. The coefficients of the logarithm of financial institution access and financial market efficiency are substantially positive and negative, respectively, when considering various components of the financial development index. The weighted variables of neighboring countries are seen in the lower part of the tables, and their marginal effects are examined in Table 8.
The calculation results of Eq. 2 are provided in Table 9 to understand the impact of financial development spillover effects on energy production. The findings show that all aspects of financial development have significantly important consequences. But for financial market performance, all aspects of financial development have a clear positive impact on CO2 emissions. The addition of the interaction word improved the estimation results significantly, indicating that not using such effects might lead to misleading results. Except for financial market performance, the indirect effects are negative, according to the findings. The spillover effects of the various components of financial development on energy intensity were measured using Eq. 3:     (0.000***) (0.000***) (0.000***) (0.000***) (0.000***) (0.000***) (0.000***)   (0.000***) (0.000***) (0.000***) (0.000***) (0.000***) (0.000***) lnGDPP 2 − 0.084 − 0.063 − 0.029 − 0.034 − 0.036 − 0.028 (0.000***) (0.000***) (0.001***) (0.000***) (0.000***) (0.001***) lnURB 0.505 0.643 0.678 0.677 0.690 0.735 (0.000***) (0.000***) (0.000***) (0.000***) (0.000***) (0.000**  Table 10 shows the direct and spatially indirect effects of all Model B2 variables, as well as the variables specific to Model B3 to B7's financial development. Specific effects measure the influence of independent variables on a special country's dependent variable, while spatially indirect effects measure the effect of independent variables in neighboring countries on a special country's dependent variable. The direct results are significantly different from the approximate values when Tables 9 and 10 are compared. Since the primary effects involve feedback effects from crossing adjacent states and returning to the states themselves, the indirect effects include feedback effects. To investigate the impact of adjacent countries' independent variables on a country's CO2 emissions, we must concentrate on the spatially indirect effects in Table 10. According to the findings, while the control independent variables in neighboring countries have no major effects on CO2 emissions in other countries, the direct and spillover effects of neighboring countries' financial development on CO2 emissions in other countries are significant. The direct effects of financial development in neighboring countries' economies and organizations on a country's CO2 emissions are positive, although spillover effects are negative. In terms of financial market performance, the effects are opposite. The sign of the coefficient is very similar to the non-spatial direct results of countries. The results of neighboring countries' financial development are close to the effects of a country's own internal financial development. Previous literature also studied the effect of financial development on CO2 emissions by traditional econometrics methods (Adom et al., 2018;Tamazian et al., 2009;Zaidi et al., 2019;Zakaria & Bibi, 2019) and concluded that financial development reduced local CO2 emissions and some scholars have highlighted the positive p-value, ***, **, and * show significance at 1%, 5%, and 10% level, respectively influence of financial development on environmental pollution (Sadorsky, 2011;Shahbaz et al., 2016) in all of them, and spatial dependence was not taken into account and estimation which made our research results different where our finding provide new insights into this subject. In addition, a study by Lv and Li (2021) is the only study that is similar to our finding where indicated that there is a spatial correlation between CO2 emissions among countries and a country's CO2 emissions could be influenced by the financial development of its neighbors which same as our finding.

Conclusion and policy implications
The spatial influence of CO2 emission determinants was studied in this analysis, which used data from 43 European countries from 1996 to 2018. Diagnostic experiments in the model selection process contributed to the selection of the spatial Durbin model, resulting in skewed estimation and harmed performance outcomes due to the model's failure to account for spatial effects. According to the findings, the impact of economic development on CO2 emissions is inverted U-shaped, confirming the EKC hypothesis. Except for trade transparency, the findings demonstrate that the factors have positive substantial effects on CO2 emissions. The impact of six financial development metrics in three separate equations was investigated in this article. Except for the coefficients of the logarithm of financial institution access and financial market performance, which emphasize substantially positive and negative, respectively, the liner results of most components of financial development are meaningless. As the relationship terms are entered, the majority of financial development indices have a favorable and important impact on CO2 emissions due to increased energy quality, but the spillover results on how energy intensity effects are negative. Financial sector performance suffers as a result of such efficacy. In a separate study, this article looks at how factors in adjacent countries influence CO2 pollution in a country of origin and using the spatial effects of variables and shows the negligible effects of GDP growth and other control variables on CO2 emissions in a neighboring region. Indicators of financial development, on the other hand, show major consequences. Financial development in neighboring countries has comparable effects on CO2 emissions as domestic financial development, according to the data. Such findings point to a form of financial convergence and financial transition in European countries, implying that reducing greenhouse gas emissions necessitates any level of international policy convergence.
The results provide policy recommendations for environmental management. The expansion of energy efficiency should be considered as a basis for the development of environmentally friendly finance. Although structural changes and optimal processes reduce CO2 emissions by improving energy efficiency (Wang et al., 2019), the results of the present study indicate that the flow of financial resources toward green technologies investment can increase the beneficial environmental dimensions of such policies. For example, the development of technologies such as carbon capture and utilization (CCUS) and carbon capture and storage (CCS) with its direct effects on increasing energy efficiency (Raza et al., 2019) should be considered by policymakers. The mere drafting of international agreements such as the Paris Agreement cannot be a guarantee to reduce greenhouse gas emissions, because countries are unable to develop energy efficiency due to the lack of internal mechanisms due to its technology-dependent dimensions. Clean technologies between countries should be considered a global issue and the context of such transfers should be considered by policymakers. However, such a transfer seems to have an internal automated mechanism in European countries. Because the effects of financial development overflow in countries, with the expansion of energy efficiency, lead to the movement of clean innovations between countries and ultimately reduce air pollution in neighboring countries. In a way, regional financial convergence itself provides the basis for regional movements to expand energy efficiency, an issue that could highlight the need to pay attention to the environmental dimensions of financial convergence in other countries.

Author contributions NA.
Funding: This research was supported by the National Natural Science Foundation of China under (Grant No. 1911818).
Availability of data and material Data used in this research are taken from the world bank website available at: https:// data. world bank. org/ indic ator. All results reported in this research are carried out on R-environment, a user-friendly statistical analysis tool with help of PLM-package available online under https:// cran.rproje ct. org/ web/ packa ges/ plm/ vigne ttes/ plmpa ckage. html.
Method: Spatial Durbin Econometrics Model.

Declarations
Ethics approval This article does not contain any studies with human participants or animals performed by any of the authors.