Study on the effect of green financial policies on low-carbon economic development based on evidence from green financial reform and innovation pilot zone

Green finance is key in supporting industries’ green transformation and helping achieve low-carbon economic (LCE) development. This paper constructs an LCE development index using panel data from 30 provinces in China from 2011 to 2020. Based on the establishment of the first five pilot green finance zones in China in 2017 as a quasi-natural experiment, the synthetic control method (SCM) is applied to explore the impact of green finance policies on the level of LCE development and to analyze the mechanism and evaluate the policy effects. The empirical results show that (1) the synthetic analysis unit better fits the development trend before the implementation of the pilot. (2) After the implementation of the pilot reform, the level of LCE development in Zhejiang, Jiangxi, Guangdong, and Guizhou provinces has a more significant enhancement effect, but the enhancement in Xinjiang is not significant, which indicates that the reform effect in Zhejiang, Jiangxi, Guangdong, and Guizhou is significantly better than that in Xinjiang to a certain extent. (3) The samples were statistically significant and passed the placebo and ranking tests. Additionally, this paper analyzes the mechanism of policy effectiveness in terms of sci-tech innovation (STI) and energy consumption structure: green finance as a grip for economic transformation can provide financial support for regional STI and energy consumption structure upgrade and promote the capital flow to green low-energy industries, ultimately achieving sustainable economic development. Based on the above findings, policy insights can be provided for the improvement of green finance pilot regions.


Introduction and literature review
With regard to global warming, China, being the secondlargest economy on the planet with a large proportion of traditional industries and economic development mainly relying on the crude model of growth (Sun et al. 2018), has an urgent need to change its previous development mode, which is highly dependent on natural resources and energy consumption, to a green industrial structure and LCE. Simultaneously, China, as one of the sensitive areas of global climate change, is more affected by climate change, especially extreme weather, than the global average. The "double carbon" target set by China is determined by both the natural environment and the country's social and economic advancement, and an LCE seems to be a better choice to weigh the economic and environmental dilemma. The UK conceived the concept of an LCE for the first time in 2003, which mainly refers to the shift from relying on traditional energy sources for economic development to relying on clean energy sources for economic development, thereby reducing environmental pollution while pursuing economic growth. Before this, the world had already made action toward the transition to an LCE. For instance, the United Nations Framework Convention on Climate Change was established in 1992 to reduce carbon emissions from automobiles, and agreements and policies were developed based on this (Huettner et al. 2010). After this, the world still has not slowed down the pace of LCE transformation, the signing of the Paris Climate Agreement, the emission  (Shimada et al. 2007), Vancouver created the Greenest City 2020 Action Plan, and the 14th 5-year plan formulated by China in 2021 to devise a stable, low-carbon, green, and economic system are all low-carbon actions so as to lower carbon emissions. Building an LCE is not only conducive to the protection of the ecological environment and the energy monument's optimization but also the cultivation of the country's sustainability in competition and the achievement of sustainable development (Yin et al. 2022).
China first started its energy-efficient and reduction of greenhouse gases initiatives in 2006, in the 11th 5-year plan (Zhang & Da 2015), and this plan mainly involves energy restructuring, industrial optimization, circular economy, cutting-edge carbon-free technology, the market for trading carbon emissions, and carbon sink projects.

Research on green finance
Green finance refers to investment and financing that provides environmental benefits (Cai & Guo 2021) and serves as a link between finance and environmental protection concepts, using the power of the financial market to direct money toward the market for environmental protection, assisting the regional industrial economy's optimization ) and adjustment. The establishment of green finance pilot zones as an important GFP can stimulate companies to adopt low-carbon remanufacturing production by increasing the cost of producing carbon (Chang et al. 2017;Cao et al. 2017), prompt them to replace traditional products with green products (Chen et al. 2016), and promote regional LCE development through technological innovation and clean energy use (Wang et al. 2022a, b). The growth of an LCE can be aided by the successful establishment of a green financial market, which will also increase development sustainability (Wara 2007). The effects of green finance policies are currently evaluated by assessing green housing policies' effects on the green building industry (Zhang et al. 2018), the effects of green credit regulations on green research and innovation, and listed firms' corporate performance (Du et al. 2022;Hou et al. 2023;Zhang et al. 2021) carbon dioxide output and the effects of low-carbon city pilot projects (Tu et al. 2022), the impact on green lifestyle changes of residents (Liu & Xu 2022), the effect of experimental carbon trading policies on business innovation in low-carbon technologies (Qi et al. 2021), the effect of low-carbon pilot schemes on the development of green technology (Liu & Xu 2022), and how regional green technology innovation is affected by the creation of Pilot zones for green finance reform and innovation ) and environmental quality effects (Hou et al. 2023;Huang & Zhang 2021).

Research on GF and LCE development
The path of green lending influencing the construction of an LCE mainly includes four aspects: first, injecting capital for the development of LCE, the construction of projects related to low-carbon industry requires a large amount of long term capital, and financial institutions have the natural advantage of capital integration to deliver sufficient funds to promote the growth of low-carbon industry; second, guiding the industry to change to green development path, under the call of low-carbon policy, green finance optimizes the supply of production factors through under the call of low-carbon policy, can reduce the excess capacity of conventional industries and help accelerate modernization and transition by optimizing the structure of supply of producing components (Bergset 2015); optimize the optimal allocation of capital; prompt idle capital to be directed into clean, low-carbon sectors, encouraging businesses to contribute to research and development; and realize the greening of industries; third, green information disclosure will force enterprises to operate with low carbon, and the disclosure of green information will, to a certain extent, restrain the high emission behavior of enterprises, improve their environmental awareness, and regulate their operation, and at the same time, through boosting the city's technical innovation capabilities and reducing industrial pollutant emissions, disclosure of green information can boost the effectiveness of urban green economies (Lin 2022); fourth, by dispersing the risk of carbon-free technology, carbon-free technology development is a new industry, and the development of new industries has certain risks. Green finance can provide solid financial needs for the research, development, and innovation of carbon-free technology and can efficiently reduce the challenges of developing carbon-free technology (Zhao and Xie 2013).
The driving factors of LCE development mainly include environmental regulation and innovation drive . Strengthening the responsibility system of environmental regulation in pilot areas may increase the shortterm cost of pollution mitigation for businesses, thus increasing the generated output and producing a kind of crowding-out effect (Dufour et al. 1998); however, longterm adjustments to the cost structure and resource allocation of businesses will require them to innovate technologically and adapt their current production structure. This innovation not only reduces the cost of reducing pollution for enterprises but also, by utilizing innovative technology, helps cut down on resource usage by enterprises (Rubashkina et al. 2015;Zhao & Sun 2016;Raymond et al. 2015;Ge et al. 2018) and promotes energy utilization efficiency (Wright & Kanudia 2014), thus improving the environmental quality of the region as well as the rate of economic expansion. At the same time, energy transition also contributes to the development of an LCE (Yu et al. 2022), and the higher the proportion of renewable energy, the more it helps the country to achieve carbon peaking at an early date (López-Menéndez et al. 2014). On the other hand, innovation is driven mainly by technological progress to increase the rate of CO 2 emissions as a way to fulfill the twin objectives of carbon reduction and economic growth (Yu et al. 2017;Li et al. 2022a, b).
By combing through the relevant kinds of literature in recent years and organizing them in Table 1, it can be found that the impact of green finance on the development of LCE varies in different regions, so further research and discussion can be conducted for pilot regions that have implemented green finance policies.
The marginal contributions of the article are mainly reflected in the following: (1) In terms of research methodology, the study innovatively uses the synthetic control method to assess the performance of LCE development, which to a certain extent enriches the scope of quantitative research in this field, and (2) based on the literature related to STI and energy consumption structure, the study further analyzes the inner mechanism of policy effects in pilot regions from the perspective of green finance for LCE development, thus providing useful exploration for enhancing LCE development.
The rest of the paper is organized as follows: "Policy background, study design, variable descriptions, and data sources" introduces the background of green finance reform and presents the study design, variable descriptions, and data sources; "Analysis of empirical results" conducts the analysis of the empirical results; "Robustness test" conducts robustness tests based on placebo and ranking tests on the empirical results; "Intrinsic mechanisms analysis" conducts the intrinsic mechanism analysis; and "Main findings and policy implications" summarizes the paper's conclusions and provides policy implications.

Policy background
Green finance has an important role in promoting the development of green industries and promoting the green transformation and upgrading of industries. Through the implementation of green finance policies, it can promote investment and financing support for energy conservation and environmental protection, clean energy, green transportation, green buildings, and other fields, while limiting the elimination of backward production capacity, increasing the speed of conversion of old and new dynamic energy and guiding the flow of capital into high-tech industries. In June 2017, the People's Bank of China, the CBRC, and seven other ministries and commissions jointly issued the General Plan for the Establishment of Pilot Zones for Green Financial Reform and Innovation, which as of November 2022 has been set up in ten pilot zones for green financial reform and innovation in seven provinces. Each pilot zone for green financial reform and innovation has introduced relevant policies and systems and implementation rules in light of its own actual situation ( Table 2).
The pilot implementation and pilot effects of the pilot zones for green financial reform and innovation are mainly reflected in the following three aspects. Firstly, the pilot zones have made positive explorations and achieved good results in terms of monetary and credit policies, financial supervision policies, financial and tax incentives, and internal management policies of institutions, taking into account the specific local development situation and proposing green finance experiences that can be replicated and extended. Secondly, the pilot zones have relied on the advantages of green financial reform and innovation to actively serve national strategies, continue to promote the construction of local ecological civilization, give full play to the function of financial support for the real economy, promote the transformation of the local green and LCE, and improve the quality of the ecological environment. Thirdly, the pilot zones promote the innovative development of green financial products and services, continuously expand green financial financing channels, promote the research and development and application of green and low-carbon technologies, provide policy support and incentives for the valorization and marketization of ecological products, and provide behavioral incentives for enterprises to implement green development transformation and environmental responsibility, resulting in the steady growth of the local green financial market.

Synthetic control method (SCM)
In this essay, we use the analytical strategy of Abadie et al. (2010), which views the green finance pilot zone policy as a policy experiment implemented by the government in several provinces. Based on this, we divide the treatment and comparison groups to assess green financing policies' effects on the growth of LCE. The SCM was initially used to examine the effects of policies of the California tobacco control ordinance and later as a policy evaluation of the property tax pilot. Specifically for this paper, the operational steps of the synthetic control method mainly include (1) determining appropriate outcome variables and predictor variables and assigning weights to the control group based on the values of the outcome variables and predictor variables before policy implementation, (2) fitting a counterfactual synthetic control province for the treatment group based on the weights, and (3) comparing the differences in LCE development in comparison to the artificial control group and the treatment group to assess the policy effects of the green finance pilot area.
Assume that panel data can be collected for (K + 1) provinces and LCE development in period T. Assume that a province (pilot zone province) is established as a green finance pilot zone at T 0 (1 ≤ T 0 ≤ T ), and the control group unaffected by the green finance pilot zone strategy is the other K provinces. Let Y 1 it denote the LCE development of province i at time t as a green finance pilot area and Y N it denotes the LCE development in the non-test area. Set is the LCE development of the pilot provinces when the policy is not implemented. D it is a bogus variable for whether it is a trial program for green finance, which takes 1 if region i is established as a pilot area at moment t and 0 otherwise. For non-test provinces, there is Y it = Y N it for the whole period T. The purpose of the study is to estimate the value of it , Y it is the data of LCE development in the test area, which is known, and it is the unobservable Y N it that needs to be estimated. This study employs Abadie's factor model by constructing the "counterfactual" variable estimat- where t is the time trend term, Z i is the (r × 1)-dimensional control variable unaffected by the green finance pilot zone policy, i is the ( F × 1)-dimensional unobservable area fixed effect, t is the (1 × F)-dimensional unobservable common factor, and it is the unobservable short-term shock with a mean of 0. The solution to solve for Y N it is to weigh the control group areas by modeling the treatment group characteristics. Thus, a (K + 1)-dimensional weight vector W * = (w 2 * ,…w k+1 * ) is derived from the predictor variables, satisfying w k ≥ 0 and w 2 + … w k+1 = 1 for any k. For the green finance test area, the potential synthetic control combination is represented by the vector W, and the synthetic control contribution of the control group areas to the green finance pilot area is represented by each w k in the combination, so that the synthetic control's outcome variable is as follows � t t is non-singular, then Eq. (4) holds the following: Abadie et al. (2010) demonstrate that in typical conditions, the side of the Eq. (4) on the left tends to 0 if the time period before the policy is long compared to the time of the implementation of the green finance pilot (Abadie et al. 2010). Therefore, during the pilot period, ∑ k+1 k=2 * k Y kt can be used as an unbiased estimate of Y N it , and thus the estimate of the policy effect it is as follows: In this paper, the five state-approved green finance pilot zones of Zhejiang, Jiangxi, Guangdong, Guizhou, and Xinjiang in 2017 are utilized as the therapy group; the rest of the mainland provinces (except Tibet) are used as the control group; and the other four treatment group provinces are excluded from the control group sample selection in the synthetic control analysis of each treatment group.

Outcome variables
This paper studies the effects of the green finance pilot zone's establishment on the development of an LCE, so the level of LCE development is selected as the outcome variable in the synthetic control method. Drawing on the existing comprehensive assessment indexes on LCE development (Pan et al. 2019), an LCE development evaluation index system in China is constructed from five dimensions of energy consumption and emission, ecological environment, economic construction, technological support, and social development, and 20 specific indicators. The particular indexing scheme is displayed in Table 3 below.
In a bid to boost the precision and scientific nature of the study and lessen the impact of subjective considerations, this paper uses the entropy value method  for the assignment of index weights, which is based on the following principles.
Suppose there are m evaluation objects, n impact factors are selected, and x ij denotes the i th evaluation object, the observed value under the j th impact factor, so that the observed value of each evaluation object will form an n-dimensional vector, noted as X ij = (x i1 ,x i2 ,…x in ).
Since the selected indicators have different magnitudes, the data were standardized in the study to eliminate the magnitudes' effect.
Positive term indicators: Inverse indicators: Define f ij as the weight of the index value of the i th evaluation object under the j th impact factor of matrix X. Then It is also assumed that f ij *lnf ij = 0 when f ij = 0. Another E j is the entropy value of the j th indicator, with Let A j denote the weight of the j th indicator, then we have The weighting factor is the value of the weighting vector A = (A1, A2, A3,igAn)constructed with Aj.

Predictor variables
The synthetic control method in practice needs to fit the situation before the policy implementation with predictor variables, so it needs to identify the motivating elements that affect the development of the LCE. This paper mainly refers Xie et al. (2017). From the perspective of influencing the performance of LCE development, the influencing factors are divided into two aspects. Input factors are selected as six indicators: car ownership, built-up urban area, amount of employees as of the year's conclusion, total fixed asset investment, and total energy consumption. Meanwhile, output factors are selected as two indicators: carbon emissions and gross regional product.

Sample selection and descriptive statistics
In this paper, panel data of 30 Chinese provinces (municipalities directly under the Central Government and autonomous regions) from 2011 to 2020 are selected as the initial sample, which five green finance pilot zones established in 2017, namely Zhejiang, Jiangxi, Guangdong, Guizhou, and Xinjiang, are taken as the experimental group, and the rest non-pilot provinces are taken as the control group (limited to data availability, Tibet is excluded from the study). The carbon emission data in this paper are obtained from the CEADs database, and the rest of the materials are obtained from China Statistical Yearbook. To reduce the volatility of the data, all predictor variables are logarithmically processed in this paper. The descriptive statistics of the outcome variable LCE development and the predictor variables are shown in Table 4.

Synthetic control method to synthesize provinces and weights
In this essay, the solution of the weight matrix of the treatment group is carried out by stata17 software, and the weights of different provinces are finally determined by synthetic control analysis as shown in Table 5. From the choice of synthetic provinces, the experimental and artificial provinces have a certain degree of similarity

Fitting results and policy effect evaluation of synthetic control method for each province
Furthermore, this paper conducts synthetic control analysis on the LCE development trend and its policy effect between the pilot green finance area provinces and the synthetic provinces. The left graph shows the LCE development level of the pilot area and the synthetic area, where the solid line represents the LCE development level of the pilot province, and the dashed line represents the LCE development level of the synthetic area. The left side of the dashed line of the time axis is the time period before the policy implementation, and the dashed line of the time axis right side is the time period after the policy implementation. The right graph shows the difference between the LCE development level of the pilot area and the synthetic area, representing the policy implementation effect. The synthetic control analysis is shown in Figs. 1, 2, 3, 4, and 5. The left side of Figs. 1, 2, 3, 4, and 5 shows the LCE development of the real pilot areas and the synthetic areas from 2011 to 2020, and the right side shows the difference between the LCE development of the real areas and the synthetic areas, which represents the assessment of the effect on the LCE development after the policy is proposed. As is evident from the left panel, the LCE development of the real pilot regions and the synthetic regions from 2011 to 2016, before the policy implementation, is well-fitted, which paves the way for the effect assessment after the policy implementation, while in 2017 and later years, the LCE development of these real pilot regions, Zhejiang, Jiangxi, Guangdong, and Guizhou, is significantly higher than that of the synthetic regions, indicating that the policy implementation is positively promoting of LCE development in these regions, which is effective for their LCE development, but does not successfully advance LCE growth in Xinjiang. Meanwhile, according to the difference between the LCE development levels of the pilot provinces and the synthetic objects, it is clear that the beneficial policy effect is enhanced year by year for Zhejiang, Jiangxi, Guangdong, and Guizhou, while the impact on policy is little for Xinjiang. The initial explanation for an arrangement for the Xinjiang Green Finance Pilot Zone not significantly enhancing the development of LCE in Xinjiang is as follows: since green finance is a long-term, complex, and systematic project, it requires the coordination of finance, environmental protection, financial supervision, and other departments to jointly promote it, while Xinjiang is supported by resource-based processing industries, with coal as the main resource consumption, and the division of labor in industry is at the low end of resource and energy raw material output or primary processing, or the low-end link of primary processing (Zhang et al. 2019), limited to the local economic development structure and industrial development stage, insufficient effective carriers to encourage the growth of green finance, and the level of energy usage is much higher than the national level, so it is difficult to promote the transformation and upgrading of industrial greening in the near term. In addition, the difference between the LCE development levels of the real pilot regions and the synthetic regions shown on the right side fluctuates up and down around the horizontal axis during 2011-2016, demonstrating that the difference in LCE development between regions before the policy implementation is not significant, and after 2017, the LCE development of these pilot regions, Zhejiang, Jiangxi, Guangdong, and Guizhou, is considerably higher than the horizontal axis, demonstrating that the implementation of the green finance policy significantly enhances the LCE development in these regions, and at the same time, the policy advantages of these pilot regions gradually increase over time, demonstrating that putting in place green finance policies has a continuously beneficial effect on the LCE development in these regions. The only reason why putting in place green finance policies did not significantly enhance the LCE development in Xinjiang is still that the formation of an LCE model is a protracted,  difficult process, and green finance in Xinjiang lacks long-term effective incentive mechanisms and supporting policies. There is no difference between the market-based benefits of green and non-green financial services, and banks and social capital have insufficient endogenous driving factors for investing in green finance to encourage the growth of green industries in the long term.

Robustness test
To exclude the chance that the policy implementation promotes the development of an LCE in pilot regions and enhance the validity of the empirical test, the placebo test and the mean square error ranking test are conducted. The two robustness tests are along the same lines, differing only in the selection of the subjects.

Robustness test based on placebo test
The placebo test method refers to the analysis of X. Wang et al. (2022a, b), and the placebo test is conducted by selecting the provinces with the largest matching weights in the test area provinces in the synthetic control analysis as the spurious treatment group. In this paper, Jiangxi and Guizhou were selected for analysis, and the results are displayed in Fig. 6. Figure 6 shows the progression trend of the LCE in Shaanxi, the province with the largest weight in Jiangxi, and Ningxia, the province with the largest weight in Guizhou, respectively, in the synthetic control analysis, assuming their LCE trends when subjected to policy implementation shocks. Again, the solid lines represent the development of the LCE in Shaanxi and Ningxia, respectively, and the dashed lines represent the synthetic case consisting of the remaining control groups. It is obvious from the two graphs that the LCE development levels of the two sham experimental groups are instead not as good as the development trend before the policy implementation, indicating that even provinces with similar economic levels and structural characteristics do not have a significant enhancement effect on their LCE development levels if they do not implement the pilot area green finance policy, verifying the findings of the above study.

Robustness tests based on mean square misranking tests
The mean square error ranking test method refers to Abadie et al. (2010), which assumes that all control group areas are subjected to policy shocks, and then assesses the policy effect by ranking the mean square error RMSPE to analyze whether there is a notable difference between the policy effect in the real test area and the policy effect in the control group areas under virtual shocks. If the RMSPE of the test area after the policy implementation is significantly higher than the RMSPE before the policy implementation, it suggests that the impact of the policy is substantial. Considering this, this paper conducts the mean square error ranking test for the two experimental areas of Guangdong and Jiangxi, and Fig. 7 displays the test findings.
As can be seen from Fig. 7, the mean squared errors of both Guangdong and Jiangxi are significantly higher than those of the control group areas after the policy implementation, and there is a good trend of upward development, indicating that the policy implementation can significantly enhance the LCE development of the test area, which further verifies the previous discussion.

Intrinsic mechanisms analysis
Some scholars have analyzed the impact of STI (Luo & Zhang 2022;Meng et al. 2023;Razumovskaya et al. 2021;Guangming et al. 2022;He & Liu 2022;Xu et al. 2023) and energy consumption structure on LCE from different perspectives, and in order to verify whether the implementation of the green finance pilot zone policy will enhance the development of low-carbon economy through the enhancement of STI and optimization of energy consumption structure, then promote the development of LCE, this article refers to Hayes' method of testing the mediating effect (Hayes 2009) and conducts an empirical test on the mesomeric effect of STI and energy consumption structure and constructs the following recursive equation.
where Y it represents the level of LCE development, and post × treat it is the policy dummy variable of the green finance pilot zone. It is a result of the provinces' dummy variables in the pilot zone and the pilot zone's implementation time for the green finance reform. If province i has established a green finance pilot zone at time t, the variable is 1; otherwise, it is 0. M it represents the mediating variable STI and energy consumption structure, and the indicator of STI is replaced by R&D input intensity (regional R&D expenditure/regional GDP). The energy consumption structure indicator is replaced by the share of coal consumption in total energy consumption. The coefficient 1 in Eq. (11) represents the total effect; the coefficient 2 in Eq. (13) represents the direct effect; and the product of in Eq. (12) and θ in Eq. (13) ( × ) represents the indirect impact. The sum of the direct and indirect effects is the overall effect, and the precise estimation outcomes are displayed in Tables 6 and 7.
From model (1) in Table 6, the coefficient 1 is significant at a 1% confidence level, so the main effect is significant, demonstrating that the implementation of the green financial reform pilot zone policy can significantly improve the level of LCE development; from models (2) and (3), the coefficients and theta are significant at 1% and 0.1% confidence levels, accordingly, showing that the mediating effect of STI exists, combined with the coefficient 2 is significant at a 5% confidence level, which indicates that the mediating effect of STI is the partially mediating effect. Specifically, the rate of STI in the model (2) is overwhelmingly positive, proving that the establishment of the pilot green financial reform zone has a positive impact on STI, i.e., the implementation of the policy can significantly improve the level of STI in the pilot area; meanwhile, both 2 and in the model (3) are significantly positive, indicating that the establishment of the green low-carbon pilot zone can improve the level of STI in the pilot area through improving the LCE development.
From model (1) in Table 7, it can be seen that the coefficient is significant at a 1% confidence level, so the main effect is significant, indicating that the establishment of  green finance pilot areas can significantly improve the development of LCE; from the model (2) and model (3), it can be seen that both the coefficient and coefficient are significant at 1% confidence level, indicating that the mediating effect of energy consumption structure exists, and then combined with the significant at 5% confidence level. The coefficients and coefficients are significant at 1% confidence level, indicating that the mediating effect of energy consumption structure exists. Specifically, the coefficient of energy consumption structure in the model (2) is significantly negative, indicating that the establishment of green finance pilot areas has a negative effect on energy consumption structure, i.e., the implementation of the policy can significantly reduce the share of coal consumption in total energy consumption in the pilot areas; meanwhile, the neutral sum in the model (3) is significantly negative, indicating that the establishment of green finance pilot areas can, by increasing the reduction of the share of coal consumption promote low-carbon production, and thus improve the development of LCE.

Main findings and policy implications
The "double carbon" target can be attained in part by developing an LCE. In this paper, the introduction of the pilot green finance zone policy in 2017 is used as an external shock, and the panel data of LCE development at the provincial level from 2011 to 2020 is selected as the sample. The five provinces that joined the pilot green finance zone in 2017 are used as the treatment group. The SCM is used to analyze the influence of the establishment of green finance pilot zones on regional LCE development. The study's findings demonstrate that putting in place green finance policies can significantly enhance the LCE development of the pilot areas; the mechanism analysis shows that the implementation of green finance policies can achieve the promotion of regional LCE development by improving the level of STI and optimizing the energy consumption structure in pilot regions. The results of this study provide empirical evidence for evaluating the green finance reform pilot area policy and enhancing regional LCE development. For policy implications, they are divided into the following aspects.
First, improve the efficient development mechanism of green finance in the green finance pilot program for reform and innovation. Timely practical use of green financing policies, efforts to achieve simultaneous development of green finance in the pilot areas based on consideration of regional differences and local resource and environmental characteristics, and unified standards for LCE development enable regions to learn from one another and promote cooperation in the same dimension.
Secondly, integrate STI and optimization of energy consumption structure into the planning and management of the pilot zone. Financial technology can maintain financial stability (Muganyi et al. 2022) and better serve economic development; optimizing energy consumption structure can better adapt to green and sustainable development. Therefore, the government should strengthen the incentive policies and subsidies for enterprise STI and expand the STI effect of green financial policies.
Thirdly, establish the information sharing mechanism of LCE and green financing. The establishment of an information-sharing mechanism can greatly minimize the harm caused by information asymmetry regarding the shift to an LCE, strengthen the review of green information disclosure of enterprises and financial institutions, and improve the low-carbon awareness of enterprises.
Fourth, pay attention to technology development and strengthen talent cultivation. Talents are the fundamental driving force of development, and regions should introduce corresponding policies for the introduction of talents, deepen cooperation with universities and research institutions to make up for the shortage of green finance professionals, and promote the development of green finance and an LCE in the region.