North–South Differences and Formation Mechanisms of Green Finance in Chinese Cities

Abstract

This paper discusses the north–south difference in green finance in China and its internal mechanism. The entropy method is used to measure the green finance level in China from 2000 to 2022 through seven dimensions of green credit, green investment, green insurance, green support, green bonds, green funds, and green equity. The Dagum Gini coefficient and variance decomposition are used to reveal the differences and sources from the perspectives of north–south space and structure. The QAP method is utilized to explore formation mechanisms of differences. Research findings show that the overall level of green finance is not high, with the southern level higher than northern one. However, spatial differences in the green finance index have slightly narrowed, with regional differences being the major contributor to uneven distribution. Secondly, the difference in green funds is the main structural source of the difference in green finance, and the effect of the difference in green bonds and green funds on the difference in green finance increases from north to south, while the effect of the difference in green credit and green support decreases from north to south. This study will enrich the assessment indicator system, providing a reference for green finance differential situations and collaborative development strategies.

1. Introduction

With the changes in temperature and the increasing threat to the sustainability of production, green finance is becoming a crucial path of fiscal operation. Green finance achieves sustainable production through products such as green bonds, green insurance, green loans, and sustainable investment. Enterprise production can reduce environmental pollution and prevent drastic changes in temperature. With the increasing global emphasis on the environment, green finance is becoming a policy requirement: at the micro level it is used as a standard and reporting requirement for governance (ESG) enterprises, and at the macro level it emphasizes carbon emission control and sustainable development goals. China is committed to achieving its carbon peak by 2030 and carbon neutrality by 2060. In order to achieve this goal, cities in China are gradually implementing green finance and reducing the degree of environmental damage caused by production.

However, the development of green finance in China has its particularities. First of all, China’s financial system has a certain connection with administrative institutions. Therefore, different cities have different levels of preferential policies and market intervention, and the progress of green finance shows spatial differences. Policy support mainly includes carbon quotas, carbon pricing, pollution trading, and so on. Carbon trading in China is still in the pilot stage and is small in scale. In addition, the government has forced green finance through fines and unannounced inspections. Policy supervision and support are uncertain, which has a certain impact on the development of carbon finance.

Second, China’s financial system is different from north to south. Market factors are more likely to promote green finance. For example, climate change creates the need for green insurance, and the transition to a low-carbon economy creates the need for transitional insurance (Carney, 2015; Chaaya et al., 2019) [1,2]. In consideration of market interests, enterprises will naturally launch more sustainable green financial products (Ma and Qin, 2020) [3]. The economic foundation of southern China is different from that of northern China. Southern cities have better financial systems because of their economic advantages. Southern cities have a higher level of marketization and legalization, and are more inclined to solve problems through market means. Northern cities have emphasized administrative regulation or forced green finance through bank credit. Government-led environmental regulation has a certain guiding effect on green finance, but if it is too strong, it will damage green finance (Liu et al., 2020) [4]. The private sector’s expertise in green finance is irreplaceable for the government sector (Kawabata T, 2020) [5].

Overall, the level of green finance in China has been significantly improved. The main questions of this study are: What are the regional differences in the process of localization of green finance policies in Chinese cities? What structural factors contribute more to the overall difference? The structure of this paper includes the following parts:

(1)

Analysis of the main research hotspots of the green finance literature and proposal of marginal contributions.

(2)

Explanation of the source of the data and the method used.

(3)

Dagum Gini coefficient used to study the spatial differences of green finance.

(4)

Study of the contribution of various parts of green finance to spatial differences and suggestions put forward.

(5)

Recommendations made.

2. Literature Review

Due to the top-down policy characteristics of China’s green finance, its spatial distribution features show different imbalances from those of Western developed countries. Here, from the perspectives of spatial pattern and structural development mechanism, the influencing factors of green finance spatial layout are explained.

2.1. Regional Differences in Green Finance

Green finance in America began in the 1980s, and the Superfund Act was enacted to regulate the sustainability of financial support. Meo et al. (2021) [6] studied the role of green finance in reducing carbon dioxide emission, and empirically proved that the best fiscal solution to reduce emissions was implementing green finance.

Green finance is echoed in different regions, with the BRICS region emphasizing environmentally sustainable green finance policies. Udeagha et al. (2023) [7] found that green finance and fintech can achieve a comprehensive carbon neutrality impact within the region and promote environmental sustainability. In Asia, Azhgaliyeva et al. (2020) [8] determined that two-thirds of the future green demand is concentrated in Asia, with a greater need for green finance. Khan et al. (2022) [9] quantified the green finance footprint of 26 economies in the Asian region and empirically demonstrated that green finance cuts down on carbon emissions and is environmentally benign. However, the evolution of green finance is slow as a result of the imperfect green financial system.

China’s green finance practice comes from top-level design, following the development logic of “top-down” and “from city to countryside”, from the beginning of the pilot to the gradual standardization. Because the urban financial system is exemplary, compared with the rural areas, urban green finance is often practiced first. This logic constitutes a major part of the green finance layout and is the primary reason for differences in the layout of green finance. From the perspective of space, the development efficiency of green finance in the eastern region is higher. The middle reaches of the Yangtze River and the southern coastal areas are in the lead, followed by the southwestern and northwestern provinces. The middle reaches of the Yellow River and the northeastern regions are, comparatively, lagging behind (Xing and Bian, 2023) [10]. Li and Fang (2020) [11] pointed out that resource-depleted cities in the north have a certain market demand for promoting green finance, mainly due to energy pressure and insufficient environmental carrying capacity. Li and Xiao (2021) [12] proved that China’s green finance shows a “diamond-shaped” distribution, respectively, with “Beijing, Shanghai, Shenzhen, Chongqing” as the corner point, and the geographical distribution characteristics of “middle-periphery” between cities are obvious. Yu and Xu (2019) [13] verified the significant spillover effect of green finance in the cities of Guangdong Province, and found that the main reasons for the regional differences are the developing level of the financial system and education. Zhou and Li (2019) [14] verified that the relationship between green finance and high-quality economy is not linear; it is an inverted “U-shaped” nonlinear structure. It is liable to exaggerate the role of green finance without considering spatial correlation. Xue (2012) [15] points out that the international “Equator Principles” cannot be fully promoted in China because the green financial system has just started.

Green finance can effectively upgrade the industrial structure, and green credit has a significant role in promoting the efficiency of the green economy, but there are still spatiotemporal inconsistencies (Guo et al., 2022) [16]. There are still some barriers to the transformation of green finance into private action due to market failures such as externalities, the lack of a green equity definition system, and the high cost of green credit financing. Therefore, the above literature proves that, despite the existence of certain “natural factors” such as the industrial base and the completeness of the financial system, “top-down” logic forms the innate reasons for the development differences of green finance, and the reality of urban hierarchical management affects the uneven distribution of green finance and determines its development. In reality, city-level management affects the uneven distribution of green finance, which determines the obvious “core-periphery” characteristics. These are reflected in the scale of the seven segments of green finance.

2.2. Mechanisms of Regional Differences in Green Finance

Under the circumstances of green finance, carbon emissions and economic pressure on the ecological environment can be reduced, and economic growth can be sustainable to a certain extent. Therefore, from an endogenous point of view it is inevitable to have a certain market demand for green finance. Mohd et al. (2018) [17] suggested that green finance reduces expenditure in the cost of pollution, which creates incentives for private individuals. Green finance, therefore, is a carbon emission reduction “natural contract” that has the potential to shift to market behavioral choices. The structure of green finance is complex, including green credit, green bonds, green investment, green funds and green equity, green insurance, green support, etc., due to different incentive levels and different levels of development; each has its own reasons.

Green credit is a common form of green finance. Yan et al. (2021) [18] selected 30 provinces as samples according to China’s provincial panel data from 2003 to 2018 and found that green credit supply can effectively increase the proportion of green clean power generation (CPG). However, constrained by scale, green credit is still underdeveloped in a few regions (Ren et al., 2022) [19]. Comparing the comparison of banks under different ownerships, private banks have a key role in green credit promotion (Weber et al., 2019) [20]. Taghizadeh-Hesary et al. (2019) [21] demonstrated that, based on project scale and investment theory considerations, the higher the investment transparency, the higher the likelihood of private participation in green credit.

Green equity is closely related to social responsibility. Despite the nonlinear relationship between green credit and enterprise development, green credit mainly shows a facilitating relationship on the green growth of enterprises (He et al., 2021) [22]. Gilchrist et al. (2021) [23] stated that due to the implementation of green bonds and green loans, etc., enterprises pay more attention to their environmental responsibility. They can not only enhance the shareholders’ value of enterprises, but also enhance the value of stakeholders. Green finance facilitates a close relationship among corporate social responsibility, knowledge capital, and achievement effect. It can also provide views of value and related evidence from the Romanian business climate (Popescu, 2019) [24].

Innovative financial products such as green bonds and debt-for-equity swaps have also contributed significantly to the green economy. Scholars have studied the economic development of six Western Balkan countries and demonstrated that green bonds provide additional financing for the region’s economic sustainability goals (Luksic et al., 2021) [25]. Ferrando et al. (2021) [26], based on the case of Brazil, stated that green bonds can solidify the natural elements of the Global South and achieve sustainable development.

Green investment is closely linked to regional regulatory policies and risk-taking environments. Desalegn et al. (2022) [27] discussed how poor management, an improper balance between risk and return, and inadequate supervision have affected the implementation of green investment. Incentive measures should be actively researched.

The level of promotion of green insurance is low due to the high level of uncertainty in the green economy. Stricker et al. (2022) [28] proposed that green insurance faces issues such as a reconsideration of risk appetite and is in urgent need of reform to support sustainability objectives. Mitchell (2003) [29] proposed that green insurance can be used as a major means to prevent agricultural production reduction and reduce the use of pesticides. However, it is not in line with the rationality of maximizing producers’ market interests. Akomea-Frimpong et al. (2022) [30] mentioned that green insurance, etc., is affected by many aspects such as religion, social inclusion, environment, etc., and these should be taken into account in policymaking. Recalling the recent environmental threats and natural disasters represented by the coronavirus pandemic caused by neocoronavirus, green insurance has become a crucial guarantee for sustainable business development (Streimikiene et al., 2021) [31].

Although the market has a certain endogenous demand for green finance, due to its public product attribute and the externality of connectivity, green finance has market failures and exogenous green support is needed.

Guild et al. (2020) [32] explored how although there is a demand for clean projects in Indonesia’s capital market, misaligned incentives make green support difficult to achieve. Kumar et al. (2022) [33] studied the situation of green support in Pakistan. Research pointed out that the uncertainty of Pakistan’s support policies and the short-termism of finance formed path dependence, which made it difficult for small and medium enterprises to achieve sustainable consumption and production. Based on data from 441 family farms in Shandong Province and Henan Province, China (Yu et al., 2020) [13], there are synergies between green support and industrial upgrading, among which the synergies of the secondary industry are the strongest. Nedopil et al. (2021) [34] studied how the significant issue with green support lies in the provision of product standards, and proposed a standard-setting approach from the three perspectives of output-based, input-based, and already-developed processes. Tuhkanen et al. (2022) [35] held that due to the difference in standards, there is a possibility of “greenwashing” green bonds, which affects the promotion of normal financial products. Hiskes (2017) [36], however, regarded green support as a “new human right” to achieve intergenerational equity, calling for the attainment of pluralistic government support.

Bakry et al. (2023) [37] looked at 76 economies and demonstrated that green finance has a restraining effect on carbon emissions. Mohanty et al. (2023) [38] analyzed 1748 papers on green finance. It was found that green finance has made great progress in marketization, and that China has the highest number of research citations in green finance. Abbas et al. (2023) [39] believe that although green finance contributes a lot to the utilization of renewable resources, it is necessary to pay attention to the obstruction of geopolitical risks to green finance. Numan et al. (2023) [40] agree that green finance requires a stable political environment. The newer works in the literature are more focused on the real world issues of green finance development, particularly the risk of political interference.

Overall, the literature involves the differences between urban and rural areas of China’s green finance and the differences caused by city classification. However, there is not much research on the north–south differences. The research in this paper focuses on the following points.

North–south differences in green finance. Firstly, green finance is divided into seven dimensions: green credit, green bonds, green investment, green fund, green equity, green insurance, and green support. The total index and sub-index are measured. The Dagum Gini coefficient is then used to investigate variability within and between regions.

The formation mechanism of north–south variability in green finance. The QAP is used to study the correlation between the seven dimensions and the total index. A regression analysis is carried out to study the mechanism of the north–south difference through regression.

3. Materials and Methods

3.1. Logic for Constructing Green Finance Indicators

Based on the classification standards and practices of Liu and He (2021) [41], data were collected and further updated and mined. The specific indicators involved are as follows.

Green credit. Green credit refers to how, in accordance with the national environmental and economic policies and industrial policies, the banking industry conscientiously implements national policies to provide loan support for energy conservational and green enterprises. Under the pressure of national environmental regulations and the attention of various sectors of society to the environment, it strictly controls the amount of loans provided to enterprises with great energy consumption and high pollution. It has both positive and reverse constraints, and can also promote the improvement of bank competitiveness.

The main procedure is for commercial banks to provide green credit to environmental protection enterprises and limit loans to polluting enterprises, thereby improving green finance. This article uses the proportion of environmental protection project credit to assess the degree of green credit. The specific calculation approach is as follows: Total credit for environmental protection projects in certain province/total credit in the province.

Green investment. This refers to the financial support for other monetary institutions in energy conservation and natural preservation in addition to bank credit and carbon finance. It mainly involves investment related to the management and maintenance of the environment, the protection of resources, and ecology. This research uses the proportion of environmental pollution control investment in GDP to value the degree of green investment.

Green insurance. This is a short-term insurance tool launched for environmental pollution accidents, mainly to hedge environmental risks brought by environmental issues during the green development process. Through the “guarantee, service, compensation” model, timely environmental health examinations are conducted. Professional insurance services are provided and risk prevention and damage compensation are implemented well. An integrated green financial environment risk prevention system is created. The level of green insurance is measured by the promotion of environmental pollution liability insurance.

Green bonds. These are a form of bond that supports ecological and environmental behavior and is conducive to the green development of the securities market. In April 2018, the Shanghai Stock Exchange proposed measures to develop green bonds. In this paper, green bond level is measured using its development degree.

Green support. Green support also needs attention and support from all aspects. Increasing policy guidance and promotion is a variety of support given by government departments to protect the environment. The proportion of fiscal expenditure on environmental protection is applied to assess the green support level.

Green funds. Green funds are dedicated investment funds. They are especially oriented to energy preserving and emission reducing strategies, low-carbon economy, and projects of environmental optimization and transformation. They aim at promoting energy conservation and emission reduction via capital investment. They are characterized by energy savings and environmental protection, with better scientific and technological components and inspiring return prospects. They can also reduce financing cost risks. This paper adopts the green fund proportion to measure its level.

Green equity. Green equity trading mainly includes the carbon emission trade, energy-using right trade, and pollution-discharge right trade, and so on. On the one hand, resource allocation can be optimized through the market to achieve the sustainable development of green rights and interest resources, and then build a green society. On the other hand, it can promote innovation and realize innovation in green technology, green culture, etc., in the process of transaction. The depth of green equity development is used to measure its level.

Table 1. Construction of green finance index systems.

Level 1	Level 2	Level 3	Level 4
Green Finance Index	Green credit	Proportion of environmental protection project credit	Total credit for environmental protection projects in certain provinces/total credit in the province
	Green investment	Proportion of environmental pollution control investment in GDP	Investment in environment pollution control/GDP
	Green insurance	Promotion level of environmental pollution liability insurance	Income from environmental pollution liability insurance/total premium income
	Green bond	Development degree of green bonds	Total amount of green bonds issued/total amount of all bonds issued
	Green support	Proportion of fiscal environmental protection expenditure	Fiscal environmental protection expenditure/fiscal general budget expenditure
	Green fund	Proportion of green funds	Total market value of green funds/total market value of all funds
	Green equity	Depth of green equity development	Total amount of carbon trading, energy use rights trading, emission rights trading/rights market transactions

is for specific indicators.

3.2. Calculation Method and Data Sources

Calculation Method: The index construction method used was an entropy weight method, mainly because the entropy weight method has good monotonicity and constant data information. According to the characteristics of the data, it is more suitable for emphasizing the objective difference to obtain the best weight and remove the subjective influence. At the same time, the robustness test was carried out using the variation method.

Data sources: Websites of the Municipal Bureau of Statistics, Ministry of Science and Technology, People’s Bank of China, and other authoritative institutions, and national and provincial statistical yearbooks, and the environmental status bulletin: China Science and Technology Statistical Yearbook, China Energy Statistical Yearbook, China Finance Yearbook, China Agricultural Statistical Yearbook, China Industrial Statistical Yearbook, China Tertiary Industry Statistical Yearbook, etc.

In the calculation, the data were sorted according to the structure of Liu and He (2021) [41], including 416 cities. Each dimension contains 9683 variable values and dates from 2000 to 2022. See

Table 2. Construction of the green finance index system.

Variable	Obs.	Mean	Std. Dev.	Min	Max
Green finance	9683	0.276	0.125	0.041	0.671
Green credit	9683	0.041	0.020	0.004	0.133
Green investment	9683	0.010	0.005	0.001	0.036
Green insurance	9683	0.019	0.009	0.002	0.061
Green bond	9683	0.006	0.003	0	0.024
Green support	9683	0.006	0.004	0	0.029
Green fund	9683	0.041	0.021	0.004	0.125
Green equity	9683	0.021	0.012	0.002	0.083

for details.

4. Analysis and Results

4.1. National Green Finance Results and Analysis

From the changes in the composite index, although the overall level is not high, a fluctuating and slow growth trend can be seen. In Figure 1a, the composite index was between 0.178 and 0.374, increasing from 0.178 in 2000 to 0.374 in 2022, with an average of 0.276 and an average growth rate per annum of 4.997%. From the point of each dimension, in Figure 1b, the green credit, green investment, green insurance, green bond, fund support, green fund, and green equity indexes rose steadily in the course of the observation, with growth rates of 5.186%, 5.161%, 5.176%, 5.137%, 5.309%, 4.959%, and 5.206%, respectively. For example, the green credit index increased from 0.026 in 2000 to 0.057 in 2022. Green investment grew from 0.007 in 2000 to 0.014 in 2022. Green insurance increased from 0.012 in 2000 to 0.025 in 2022. Green bonds rose from 0.004 in 2000 to 0.008 in 2022. Green support increased from 0.004 in 2000 to 0.009 in 2022. Green funds increased from 0.027 in 2000 to 0.056 in 2022, and green equity from 0.013 in 2000 to 0.029 in 2022. From the perspective of the growth rate, green funds are a relatively narrow marker for the improvement of the green financial index. The other six indexes show the comparative advantages of the green finance index.

4.2. Results and Analysis of Regional Green Finance

Referring to the division standard of economic geography (with reference to the division criteria of economic geography, northern regions include Beijing, Tianjin, Hebei, Shanxi, Inner Mongolia, Liaoning, Jilin, Heilongjiang, Shandong, Henan, Shaanxi, Gansu, Qinghai, Ningxia, and Xinjiang; southern regions include Shanghai, Jiangsu, Zhejiang, Anhui, Fujian, Jiangxi, Hubei, Hunan, Guangzhou East Province, Guangxi, Hainan Province, Chongqing City, Sichuan Province, Guizhou Province, and Yunnan Province. Due to the lack of data, this study does not include the Tibet Autonomous Region, Hong Kong, Macao, and Taiwan), 30 provinces (excluding Tibet, Hong Kong, Macao, and Taiwan) were divided into southern regions and northern regions to measure and analyze China’s green finance level across seven dimensions and its areal features. Figure 2 reflects the relative superiority and inferiority of green finance and its sub-dimensions in the two major economic zones from 2000 to 2022. Judging from the composite index size, the southern green finance level is higher than the northern level. The average green finance index in southern regions is 0.298, while the mean level of the north is 0.251. As for the successional trend of the green finance index, by 2022, both southern and northern regions showed a growing tendency, with an annual exponential growing rate of 5.00% and 4.99%, respectively, with a relatively large increase. In addition, the northern and southern index evolutional tendencies were similar, showing a relatively stable growth trend until they reached their maximum in 2022, with values of 0.343 and 0.403, respectively.

Green credit index. With regard to the scale of the green credit index, the index of South China is better than that of North China. The developing trend of the green credit index shows that the indexes in the north and the south showed varying degrees of growth. The northern region index was 0.0244 in 2000 and went up to 0.0514 in 2022. The southern index was 0.0283 in 2000 and rose to in 2022. The average growth rates per annum were 5.03% and 5.31%, respectively, indicating that the level of green credit in the north and south of China is improving. Through the study of fluctuation trends of green credit, the fluctuations of two regions are seen to be relatively stable, but due to the shock of the global novel coronavirus epidemic in 2020, the green credit index has declined.

Green investment index. Considering the size of the green investment index, there was a small difference between the green investment index in the north and in the south. From 2000 to 2022, the mean values of the investment index were 0.009 and 0.011, respectively. If only from the average point of view, the southern green index was higher than the northern index. With respect to the change trends of the green investment index, the movement of indices in the north and south regions was similar. During the research period, they showed an expanding trend and the annual average change rates were 4.808% and 5.451%, respectively.

Green insurance index. From the view of the size of the green insurance index, the southern region’s green index was above the northern region’s index, with an average of 0.020 and 0.017, respectively. From the trends of changes in the green insurance index, both major economic zones showed a growth trend. The northern green insurance index increased from 0.011 in 2000 to 0.013 in 2022, while the southern green index increased from 0.013 in 2000 to 0.028 in 2022 with growing percentages per year of 5.007% and 5.310%, respectively.

Green bond index. Green bond sizes in the south were greater than those in the north, with averages of 0.006 and 0.007, separately. From the aspect of development prospects, green bonds in the north and south economic zones in the sample period demonstrated an increasing trend and the average rates each year were 5.036% and 5.216%.

Green support index. China has always pursued new nationwide systems and adhered to the principle of “better integration of effective markets and promising governments”. There was no significant difference in green support between the north and the south, with average values of 0.007 and 0.006 and average annual growth rates of 5.162% and 5.502%. The green support index in the south was relatively large, mainly due to the high level of factor market allocation in the south and fewer obstacles to factor flow compared with the north. Moreover, there are a large number of non-state-owned enterprises in the southern region, with relatively high economic activity.

Green fund index. Comparing by size, the southern green fund index was higher, with an average of 0.045, while the northern green index was relatively lower, with an average of 0.038. The fluctuation trend of the green fund index in two regions was relatively stable, with a mean increasing rate per year of 5.231% and 4.633%, respectively.

Green equity index. The green equity index in south districts was superior to that in north area, with averages of 0.022 and 0.019 and average annual rising rates of 5.288% and 5.107%.

5. Spatial Differences and Formation Mechanisms of Green Finance

5.1. Spatial Differences in Green Finance

5.1.1. Overall Differences in Green Finance

In

Table 3. Green finance and multi-dimensional Gini coefficients.

Year	Green Finance Index	Green Credit	Green Investment	Green Insurance	Green Bond	Green Support	Green Fund	Green Equity
2000	0.2137	0.2486	0.2572	0.2531	0.2761	0.3193	0.2446	0.2823
2001	0.2140	0.2525	0.2636	0.2545	0.2829	0.3244	0.2476	0.2772
2002	0.2144	0.2519	0.2638	0.2539	0.2704	0.3182	0.2540	0.2867
2003	0.2151	0.2548	0.2599	0.2581	0.2747	0.3260	0.2499	0.2795
2004	0.2139	0.2477	0.2619	0.2538	0.2701	0.3134	0.2491	0.2813
2005	0.2154	0.2548	0.2620	0.2572	0.2764	0.3226	0.2467	0.2809
2006	0.2129	0.2487	0.2607	0.2516	0.2754	0.3135	0.2447	0.2796
2007	0.2132	0.2484	0.2629	0.2500	0.2786	0.3092	0.2508	0.2856
2008	0.2136	0.2539	0.2602	0.2555	0.2836	0.3192	0.2521	0.2802
2009	0.2151	0.2623	0.2753	0.2640	0.2842	0.3218	0.2452	0.2868
2010	0.2146	0.2476	0.2619	0.2505	0.2685	0.3001	0.2493	0.2833
2011	0.2121	0.2473	0.2518	0.2516	0.2744	0.3162	0.2495	0.2826
2012	0.2130	0.2450	0.2545	0.2467	0.2687	0.3158	0.2476	0.2894
2013	0.2145	0.2475	0.2601	0.2519	0.2818	0.3040	0.2527	0.2970
2014	0.2151	0.2540	0.2576	0.2587	0.2830	0.3141	0.2503	0.2804
2015	0.2168	0.2538	0.2608	0.2568	0.2744	0.3317	0.2529	0.2939
2016	0.2151	0.2523	0.2593	0.2573	0.2727	0.3194	0.2521	0.2896
2017	0.2135	0.2453	0.2538	0.2477	0.2685	0.3191	0.2450	0.2839
2018	0.2143	0.2452	0.2545	0.2503	0.2685	0.3123	0.2449	0.2886
2019	0.2184	0.2545	0.2642	0.2557	0.2717	0.3280	0.2584	0.2927
2020	0.2152	0.2529	0.2677	0.2561	0.2796	0.3181	0.2513	0.2853
2021	0.2144	0.2425	0.2513	0.2446	0.2724	0.3221	0.2543	0.2921
2022	0.2137	0.2446	0.2559	0.2467	0.2699	0.3273	0.2472	0.2824

, the green financial index Gini coefficient had a relatively stable trend, and it was between 0.212 and 0.218 with a small fluctuation and a small downward trend in the fluctuation. In 2000, the Gini coefficient of the composite index was 0.214. After experiencing the fluctuation of “up down up down”, the Gini coefficient of the composite index declined to 0.213 in 2022. The green credit Gini coefficient was comparatively large, ranging from 0.242 to 0.262, showing a narrowing trend in the observation period, with 0.249 in 2000 and 0.245 in 2022. A decline of 0.074% every year demonstrates that regional differences in green credit were further narrowed. The Gini coefficient of green investment fluctuated greatly, and experienced “up–down–up–down” fluctuations. In 2000, the Gini coefficient of green investment was 0.257, and it then showed an upward trend. It reached 0.262 in 2010. In 2011, the Gini coefficient began to decrease, until it reached 0.252 in 2011, and then began to rise slowly. In 2022, the Gini coefficient was 0.256, with a falling per year value of 0.022%, proving that the regional difference in green investment was further narrowed. The Gini coefficient of green insurance was between 0.245 and 0.264. In 2000, the Gini coefficient of green insurance was 0.253. In 2009, it reached the highest value of 0.264, then decreased among fluctuations. On the whole, the green insurance Gini coefficient revealed a small descending trend, with a yearly declining rate of 0.115%. The green bond Gini coefficient was between 0.268 and 0.284, with an average of 0.276 and a decline 0.103%, showing a downward trend in fluctuations.

The Gini coefficient among green support was better than that of the composite index and its multi-dimensional indicators, ranging from 0.300 to 0.332. The change trend was relatively stable. During the observation period, it showed a slow growth trend of 0.319 in 2000 and 0.327 in 2022, with 0.113% as the annual growing rate. The Gini coefficient of the green fund was between 0.245 and 0.258, rising in constant volatility. In 2000, the Gini coefficient was 0.245, and then it began to show an upward trend in fluctuations. In 2019, it reached the maximum value of 0.258, and then showed a downward trend. Before 2022, the Gini coefficient fell to 0.247. The annual increase during the observation period was 0.048% during the follow-up period. The Gini coefficient of green equity was between 0.277 and 0.297. During the sample period from 1997 to 2020, the difference between green regions narrowed, but there was no overall change. To sum up, if we look at the size of the spatial differences in each dimension, the regional differences from large to small are green support, green equity, green bonds, green investment, green insurance, green credit, green fund, and green finance. From the perspective of the change tendency of spatial differences in all dimensions, regional differences in green finance, green credit, green investment, green insurance, and green bonds showed a narrowing trend, while regional differences in green support, green equity, and green fund showed an expanding trend. It is precisely because of the different trends in the spatial differences of various financial dimensions that the spatial differences exhibited a fluctuating trend.

5.1.2. Regional Differences in Green Finance

Regional variations of the green finance composite index. In Figure 3a, the intra-regional Gini coefficient of northern China showed an increasing trend, from 0.241 in 2000 to 0.242 in 2022, ranging from 0.236 to 0.250. In Figure 3b, the southern intra-regional Gini coefficient, between 0.178 and 0.187, had a downward trend, expressing that the difference in the intra-regional green finance level showed a narrowing direction. As a whole, it presented an “N-type” fluctuating trend of “declining–rising–declining–rising”. In 2010, the Gini coefficient reached a maximum of 0.187. In 2018, it decreased to 0.179, and in 2022, it increased to 0.182, showing a downward trend.

(1)

Regional differences in green credit. During the study period, the Gini coefficient of the green credit index decreased slightly in both the north and south parts of China. The Gini coefficient in the north part exhibited “rising–falling–rising–falling”, and the differences within the region accordingly experienced a process of “expanding–reducing–expanding–shrinking”. The Gini coefficient decreased from 0.273 to 0.264 between 2020 and 2022, with a yearly decline of 0.165%. The Gini coefficient in southern provinces presented a condition of “declining–rising–falling”, with the Gini coefficient decreasing from 0.222 in 2000 to 0.221 in 2022. It decreased at a rate of 0.018% each year.

(2)

Regional differences in the green investment index. The Gini coefficient relating to the green investment index in northern China showed a change trend of “up–down–up–down”. It increased from 0.282 in 2000 to 0.300 in 2009, and then began to decline. In 2014, it reached the minimum value of 0.273052, and then began to rise. In 2018, it increased to 0.2931414, and by 2022, the Gini coefficient was 0.275, showing an overall decline. The Gini coefficient of the green investment index in the southern region showed a “decline rise decline” trend, reaching the maximum value of 0.246 in 2020 and declining to 0.232 in 2022.

(3)

Regional differences in the green insurance index. The Gini coefficient concerning the northern green insurance index showed a significant decline from 0.277 in 2000 to 0.264 in 2022, with a yearly declining scope of 0.213%. Southern China showed a small descent of 0.228 to 0.224 from 2000 to 2022, falling 0.079% each year.

(4)

Regional differences in the green bond index. The Gini coefficient regarding the green bond index in the northern region showed a significant decline, while that of the southern region showed a small increase. This shows that the regional differences in green bonds in the north narrowed, while the regional differences in green bonds in the south expanded. The northern Gini coefficient descended between 0.296 and 0.276 from 2000 to 2022, falling at a yearly rate of 0.306%. The southern region coefficient, however, increased from 0.253 in 2000 to 0.257 in 2022, with a growing percentage of 0.076%.

(5)

Regional differences in the green support index. The Gini coefficient of the green support index in the north and south rose constantly, indicating that differences in the green support index within either district further expanded. The Gini coefficient of the index in northern China fluctuated frequently, showing an overall growth in fluctuation from 0.352 in 2000 to 0.359 in 2022, growing yearly at 0.090%. Meanwhile, the southern Gini coefficient fluctuated in an obvious “N” shape, rising from 0.284 in 1997 to 0.312 in 2008 and declining to 0.279 until 2013, and then beginning to rise. In 2019, the Gini coefficient reached 0.320, rising at an annual rate of 0.176%.

(6)

Regional differences in the green fund index. By observing the samples, the Gini coefficient of the northern green fund index could be seen to increase continually, while that of the southern green fund index went downwards. This indicates that regional differences in the northern green fund further expanded, and regional differences in the southern green fund further narrowed. The Gini coefficient of the northern zone increased from 0.268 in 2000 to 0.275 in 2022, growing annually at 0.111%. The Gini coefficient of the southern region declined from 0.221 in 2000 to 0.216 in 2022, declining yearly at −0.091%.

(7)

Regional differences in the green equity index. As samples were being observed, the Gini coefficient of green equity index in the north was increasing, while the southern one was descending. This trend expresses that northern differences in green equity further expanded, and southern differences in green equity further narrowed. The value of the Gini coefficient in northern China increased from 0.306 in 2000 to 0.309 in 2022, going up at a rate of 0.033% every year. The Gini coefficient in the southern provinces declined from 0.259 in 2000 to 0.257 in 2022, decreasing at −0.034% yearly.

To sum up, (1) the Gini coefficient associated with green investment, green credit, and green insurance in the north and south showed downward trends to varying degrees. The Gini coefficient relevant to green bonds in the north showed a slight downward trend. Meanwhile, the green fund index and green equity index in the south had downward trends. All the above illustrate that the intra-regional differences in these indicators have narrowed during the observation time. (2) The Gini coefficient within the green support index was the largest in both the south and the north. This can be interpreted as indicating that the regional variance in the green support index between the north and the south is the largest. (3) From the mean value of the Gini coefficient, the intra-regional differences of green finance and its sub dimension indexes of northern regions are greater than those of the southern parts to varying degrees.

5.1.3. Regional Differences in Green Finance

From the Gini coefficients in

Table 4. Gini coefficients between green finance and multi-dimensional regions.

Year	Green Finance Index	Green Credit	Green Investment	Green Insurance	Green Bond	Green Support	Green Fund	Green Equity
2000	0.2203	0.2535	0.2609	0.2577	0.2812	0.3269	0.2478	0.2854
2001	0.2213	0.2587	0.2694	0.2609	0.2894	0.3280	0.2544	0.2820
2002	0.2217	0.2564	0.2685	0.2582	0.2753	0.3206	0.2600	0.2919
2003	0.2217	0.2608	0.2652	0.2640	0.2813	0.3303	0.2547	0.2849
2004	0.2213	0.2526	0.2665	0.2580	0.2743	0.3184	0.2554	0.2865
2005	0.2225	0.2623	0.2705	0.2642	0.2827	0.3282	0.2538	0.2875
2006	0.2203	0.2557	0.2666	0.2583	0.2805	0.3232	0.2520	0.2859
2007	0.2214	0.2543	0.2696	0.2554	0.2837	0.3153	0.2580	0.2948
2008	0.2209	0.2591	0.2652	0.2604	0.2906	0.3226	0.2597	0.2866
2009	0.2221	0.2734	0.2853	0.2757	0.2924	0.3340	0.2517	0.2928
2010	0.2217	0.2534	0.2678	0.2567	0.2726	0.3035	0.2549	0.2896
2011	0.2197	0.2551	0.2582	0.2594	0.2819	0.3238	0.2549	0.2875
2012	0.2199	0.2509	0.2603	0.2526	0.2726	0.3213	0.2545	0.2965
2013	0.2220	0.2536	0.2661	0.2578	0.2889	0.3083	0.2610	0.3049
2014	0.2223	0.2588	0.2619	0.2638	0.2883	0.3191	0.2563	0.2863
2015	0.2240	0.2589	0.2651	0.2619	0.2793	0.3370	0.2581	0.2988
2016	0.2227	0.2612	0.2682	0.2659	0.2807	0.3264	0.2594	0.2980
2017	0.2210	0.2530	0.2612	0.2547	0.2763	0.3270	0.2512	0.2891
2018	0.2221	0.2521	0.2613	0.2573	0.2744	0.3173	0.2524	0.2940
2019	0.2258	0.2596	0.2696	0.2611	0.2776	0.3299	0.2644	0.2982
2020	0.2228	0.2591	0.2732	0.2625	0.2876	0.3221	0.2573	0.2894
2021	0.2225	0.2473	0.2555	0.2493	0.2768	0.3294	0.2597	0.2958
2022	0.2203	0.2505	0.2623	0.2527	0.2747	0.3329	0.2535	0.2858

, by 2022, the north–south differences from the largest to the smallest are green support index, green equity index, green bond index, green investment index, green insurance index, green credit index, green fund index, and green finance index. Among them, the regional differences of indices concerned with green finance, green credit, and green bonds all showed a downward trend, while those relating to green investment, green insurance, green support, green fund, and green equity had an uptrend. Specifically, the regional Gini coefficient of the green financial composite index changed slightly, having an average value of 0.226. The Gini coefficient of the green credit index between the north and south regions decreased from 0.254 in 2000 to 0.250 in 2022, declining annually at 0.055%. The north–south Gini coefficient of the green investment index increased from 0.261 in 2000 to 0.262 in 2022, going down every year at a speed of 0.025%. The Gini coefficients between the south and north of the green support index, green fund index, and green rights index went up from 0.327, 0.248, and 0.285 in the year 2000 to 0.333, 0.254, and 0.286 in 2022, going upwards, respectively, at the rate of 0.083%, 0.105%, and 0.007%. The green insurance index decreased from 0.281 in the year 2000 to 0.275 in 2022, decreasing annually at a rate of 0.105%.

To sum up, the regional differences among green investment index, green support index, green fund index and green equity index were the main reasons for the formation of regional differences between the north and south in green finance. The green credit index, green insurance index, and green bond index showed downwards trends between the north and south regions. The differences in green investment index, green support index, green fund index, and green equity index between the north and south regions showed upward trends.

5.1.4. Sources of the Green Financial Regional Differences

Table 5. Gini coefficient between green finance and multi-dimensional regions.

Year	Green Finance Index			Green Credit			Green Investment			Green Insurance
	Gw	Gnb	Gt	Gw	Gnb	Gt	Gw	Gnb	Gt	Gw	Gnb	Gt
2000	0.4872	0.1853	0.3275	0.4925	0.1485	0.3590	0.4949	0.1152	0.3899	0.4933	0.1434	0.3633
2001	0.4856	0.1948	0.3195	0.4902	0.1547	0.3551	0.4914	0.1464	0.3622	0.4900	0.1677	0.3423
2002	0.4857	0.1966	0.3178	0.4933	0.1380	0.3687	0.4935	0.1405	0.3659	0.4938	0.1340	0.3723
2003	0.4873	0.1882	0.3245	0.4908	0.1670	0.3422	0.4923	0.1580	0.3497	0.4911	0.1690	0.3399
2004	0.4854	0.1984	0.3162	0.4925	0.1476	0.3599	0.4935	0.1370	0.3696	0.4940	0.1360	0.3701
2005	0.4862	0.1931	0.3207	0.4878	0.1678	0.3444	0.4864	0.1663	0.3473	0.4890	0.1603	0.3507
2006	0.4852	0.2003	0.3145	0.4887	0.1824	0.3289	0.4912	0.1534	0.3554	0.4895	0.1750	0.3355
2007	0.4836	0.2162	0.3003	0.4908	0.1660	0.3433	0.4901	0.1742	0.3357	0.4917	0.1578	0.3505
2008	0.4854	0.1968	0.3178	0.4922	0.1465	0.3613	0.4927	0.1425	0.3649	0.4928	0.1367	0.3705
2009	0.4863	0.1952	0.3185	0.4820	0.2234	0.2946	0.4851	0.2044	0.3105	0.4812	0.2223	0.2966
2010	0.4861	0.1981	0.3158	0.4908	0.1666	0.3426	0.4912	0.1629	0.3459	0.4903	0.1748	0.3348
2011	0.4848	0.2028	0.3123	0.4870	0.1849	0.3281	0.4899	0.1630	0.3470	0.4872	0.1824	0.3303
2012	0.4864	0.1955	0.3180	0.4904	0.1595	0.3500	0.4912	0.1644	0.3444	0.4905	0.1650	0.3445
2013	0.4851	0.2006	0.3144	0.4901	0.1654	0.3445	0.4910	0.1580	0.3510	0.4908	0.1632	0.3459
2014	0.4859	0.1960	0.3181	0.4931	0.1617	0.3452	0.4942	0.1518	0.3540	0.4928	0.1677	0.3396
2015	0.4861	0.1964	0.3175	0.4925	0.1520	0.3555	0.4939	0.1343	0.3717	0.4926	0.1472	0.3602
2016	0.4851	0.2061	0.3088	0.4853	0.2044	0.3103	0.4859	0.1975	0.3166	0.4864	0.1999	0.3137
2017	0.4851	0.2036	0.3113	0.4870	0.1859	0.3271	0.4882	0.1794	0.3324	0.4885	0.1798	0.3317
2018	0.4844	0.2091	0.3066	0.4884	0.1612	0.3504	0.4891	0.1389	0.3720	0.4884	0.1576	0.3540
2019	0.4856	0.1953	0.3191	0.4923	0.1469	0.3608	0.4920	0.1438	0.3641	0.4919	0.1548	0.3533
2020	0.4852	0.1985	0.3163	0.4904	0.1855	0.3240	0.4924	0.1601	0.3475	0.4904	0.1891	0.3205
2021	0.4837	0.2112	0.3051	0.4922	0.1318	0.3761	0.4936	0.1033	0.4032	0.4923	0.1184	0.3893
2022	0.4871	0.1870	0.3258	0.4906	0.1795	0.3299	0.4903	0.1799	0.3298	0.4906	0.1783	0.3312

illustrates different sources of the green finance composite index. The contribution of inter-regional differences is in two parts: inter-regional net excess variability and inter-regional hypervariable density. The inter-regional difference is the uppermost source of the imbalanced distribution of green finance, and the average contribution was 51.44%. Among the inter-regional differences between the north and the south, the inter-regional excess net value went upwards with the fluctuation. The fluctuation range was small, ranging from 18.53% to 21.62%, and the average contributing rate was 19.85%. The inter-regional hypervariable density contributed to the decreasing trend in fluctuations, ranging from 30.03% to 32.75%, at a contributing rate of 31.59%. The contribution of inter-regional hypervariable density difference far exceeded that of the inter-regional hypervariable net value. This means that the cross-overlapping problem among different regions was the primary reason for the unbalanced distribution of green finance in the country. The regional difference contribution rate within the regions ranged from 48.36% to 48.73%, an average of 48.56%, which was relatively stable. This suggests that regional differences are of importance to the uneven distribution of green finance. There exists a significant gap in the comprehensive index of green finance among provinces within each region, resulting in uneven development within the region.

Table 6. Sources and contributions of regional differences in green finance and sub-dimensions.

Year	Green Bond			Green Support			Green Fund			Green Equity
	Gw	Gnb	Gt	Gw	Gnb	Gt	Gw	Gnb	Gt	Gw	Gnb	Gt
2000	0.4934	0.1459	0.3606	0.4910	0.1492	0.3598	0.4954	0.1024	0.4023	0.4965	0.0882	0.4153
2001	0.4910	0.1413	0.3677	0.4963	0.0853	0.4184	0.4888	0.1608	0.3504	0.4938	0.1316	0.3745
2002	0.4933	0.1456	0.3611	0.4982	0.0891	0.4127	0.4909	0.1674	0.3417	0.4936	0.1508	0.3556
2003	0.4906	0.1679	0.3415	0.4958	0.1170	0.3872	0.4926	0.1353	0.3721	0.4927	0.1392	0.3681
2004	0.4945	0.1373	0.3683	0.4947	0.1394	0.3659	0.4898	0.1587	0.3515	0.4929	0.1116	0.3955
2005	0.4910	0.1465	0.3626	0.4940	0.1490	0.3570	0.4882	0.1689	0.3429	0.4907	0.1443	0.3650
2006	0.4934	0.1566	0.3500	0.4880	0.1909	0.3211	0.4877	0.1800	0.3323	0.4913	0.1448	0.3639
2007	0.4933	0.1343	0.3723	0.4927	0.1386	0.3688	0.4885	0.1969	0.3146	0.4874	0.2185	0.2941
2008	0.4904	0.1640	0.3456	0.4968	0.1088	0.3943	0.4878	0.1929	0.3193	0.4914	0.1708	0.3377
2009	0.4886	0.1796	0.3319	0.4847	0.2160	0.2993	0.4894	0.1729	0.3377	0.4921	0.1476	0.3603
2010	0.4948	0.1484	0.3568	0.4964	0.1095	0.3940	0.4912	0.1602	0.3486	0.4918	0.1656	0.3427
2011	0.4891	0.1677	0.3431	0.4909	0.1670	0.3421	0.4914	0.1327	0.3759	0.4935	0.1189	0.3875
2012	0.4949	0.1177	0.3875	0.4935	0.1175	0.3890	0.4888	0.1966	0.3146	0.4909	0.1956	0.3135
2013	0.4903	0.1675	0.3423	0.4951	0.1106	0.3943	0.4866	0.2087	0.3047	0.4899	0.1988	0.3113
2014	0.4933	0.1493	0.3574	0.4944	0.1150	0.3906	0.4906	0.1658	0.3435	0.4920	0.1312	0.3768
2015	0.4937	0.1585	0.3477	0.4946	0.1264	0.3790	0.4921	0.1492	0.3587	0.4942	0.1312	0.3747
2016	0.4883	0.1946	0.3171	0.4919	0.1595	0.3486	0.4883	0.1907	0.3211	0.4885	0.1791	0.3323
2017	0.4884	0.1809	0.3307	0.4903	0.1440	0.3657	0.4899	0.1735	0.3366	0.4931	0.1193	0.3876
2018	0.4916	0.1567	0.3517	0.4946	0.1417	0.3637	0.4876	0.1936	0.3188	0.4930	0.1345	0.3725
2019	0.4917	0.1490	0.3593	0.4990	0.0821	0.4189	0.4908	0.1570	0.3522	0.4930	0.1296	0.3774
2020	0.4888	0.1930	0.3183	0.4962	0.1319	0.3719	0.4907	0.1721	0.3371	0.4950	0.1120	0.3930
2021	0.4941	0.1100	0.3959	0.4917	0.1644	0.3439	0.4920	0.1556	0.3524	0.4961	0.1389	0.3650
2022	0.4939	0.1663	0.3398	0.4939	0.1197	0.3864	0.4897	0.1644	0.3459	0.4960	0.1044	0.3996 1

¹ G_w: contribution of intra-regional variation. G_nb: contribution of inter-regional variation. G_t: contribution of hyper-variable densities.

illustrates differences sources of green finance dimensions. The intra-regional differences in green credit index, green investment index, green insurance index, green bond index, green support index, green fund index, and green equity index are mainly from inter-regional differences. The average contribution rates are 51.00%, 50.90%, 50.96%, 50.82%, 50.63%, 51.00%, and 50.74%, separately. Moreover, the average contribution rates of inter- regional hypervariable density are higher than those of the hypervariable net worth. This means that the cross-overlapping problem among different regions is an important reason for the uneven development.

5.2. Formation Mechanism of Spatial Differences in Green Finance

5.2.1. QAP Correlation Analysis

Table 7. QAP correlation analysis results.

Region	Index/Correlation	Green Finance	Green Credit	Green Investment	Green Insurance	Green Bond	Green Support	Green Fund	Green Equity
Nationwide	Green Finance	1 ***	0.993 ***	0.991 ***	0.993 ***	0.988 ***	0.972 ***	0.994 ***	0.985 ***
	Green Credit	0.993 ***	1 ***	0.998 ***	0.999 ***	0.994 ***	0.979 ***	0.988 ***	0.978 ***
	Green Investment	0.991 ***	0.998 ***	1 ***	0.997 ***	0.992 ***	0.967 ***	0.984 ***	0.975 ***
	Green Insurance	0.993 ***	0.999 ***	0.997 ***	1 ***	0.993 ***	0.974 ***	0.986 ***	0.977 ***
	Green Bonds	0.988 ***	0.994 ***	0.992 ***	0.993 ***	1 ***	0.959 ***	0.982 ***	0.973 ***
	Green Support	0.972 ***	0.979 ***	0.967 ***	0.974 ***	0.959 ***	1 ***	0.969 ***	0.96 ***
	Green Fund	0.994 ***	0.988 ***	0.984 ***	0.986 ***	0.982 ***	0.969 ***	1 ***	0.991 ***
	Green Equity	0.985 ***	0.978 ***	0.975 ***	0.977 ***	0.973 ***	0.96 ***	0.991 ***	1 ***
The North	Green Finance	1 ***	0.995 ***	0.992 ***	0.994 ***	0.99 ***	0.976 ***	0.995 ***	0.988 ***
	Green Credit	0.995 ***	1 ***	0.998 ***	0.999 ***	0.995 ***	0.982 ***	0.99 ***	0.983 ***
	Green Investment	0.992 ***	0.998 ***	1 ***	0.997 ***	0.993 ***	0.972 ***	0.988 ***	0.98 ***
	Green Insurance	0.994 ***	0.999 ***	0.997 ***	1 ***	0.993 ***	0.978 ***	0.99 ***	0.982 ***
	Green Bonds	0.99 ***	0.995 ***	0.993 ***	0.993 ***	1 ***	0.967 ***	0.985 ***	0.978 ***
	Green Support	0.976 ***	0.982 ***	0.972 ***	0.978 ***	0.967 ***	1 ***	0.974 ***	0.965 ***
	Green Fund	0.995 ***	0.99 ***	0.988 ***	0.99 ***	0.985 ***	0.974 ***	1 ***	0.992 ***
	Green Equity	0.988 ***	0.983 ***	0.98 ***	0.982 ***	0.978 ***	0.965 ***	0.992 ***	1 ***
The South	Green Finance	1 ***	0.991 ***	0.988 ***	0.99 ***	0.985 ***	0.965 ***	0.993 ***	0.98 ***
	Green Credit	0.991 ***	1 ***	0.997 ***	0.999 ***	0.992 ***	0.973 ***	0.983 ***	0.971 ***
	Green Investment	0.988 ***	0.997 ***	1 ***	0.996 ***	0.991 ***	0.958 ***	0.98 ***	0.967 ***
	Green Insurance	0.99 ***	0.999 ***	0.996 ***	1 ***	0.991 ***	0.967 ***	0.982 ***	0.969 ***
	Green Bonds	0.985 ***	0.992 ***	0.991 ***	0.991 ***	1 ***	0.948 ***	0.976 ***	0.965 ***
	Green Support	0.965 ***	0.973 ***	0.958 ***	0.967 ***	0.948 ***	1 ***	0.96 ***	0.95 ***
	Green Fund	0.993 ***	0.983 ***	0.98 ***	0.982 ***	0.976 ***	0.96 ***	1 ***	0.989 ***
	Green Equity	0.98 ***	0.971 ***	0.967 ***	0.969 ***	0.965 ***	0.95 ***	0.989 ***	1 ***

*** p < 0.01.

reports the QAP(Quadratic Assignment Procedure) correlation analysis results of green finance differences in China and the north and south regions and differences in various dimensions. Green finance inconsistency leads to different correlations between the differences in green finance in different regions and the differences of fractal dimensions. The difference of green finance in China between the northern and southern regions was positively correlated with the difference in all dimensions, and the correlation coefficients passed the statistical test at a certain significance level. Specifically, the differences in green finance in China and the north and south regions were strongly correlated with the differences in green credit, green insurance, green bonds, green support, green funds, and green equity. The correlation coefficient was more than 0.9. Relatively speaking, the differences in green finance were relatively less correlated with the differences in green support and green equity. Green support is measured by the proportion of fiscal expenditure on environmental protection, explaining that there is little difference in fiscal expense regarding environment protection. It is measured by total carbon trading, energy trading, and emission trading/equity market transactions. The difference is less correlated with that of green finance, indicating that the differences in carbon trading, energy trading, and emission trading volume are small, and the contribution to the difference in green finance is not high. From the above analysis, conclusions can be drawn that generally significant positive correlations exist between the green financial composite index and the explanatory variables. The mean value of collinearity VIF was 54.3. The VIF values of green credit, green insurance and green bonds were greater than 10, with obvious collinearity. Therefore, it is necessary to conduct the QAP regression analysis to reveal the effect of the difference of various dimensions on the differences in the green finance composite index.

5.2.2. QAP Regression Results

Table 8. Structural drivers of the formation of green finance differences across the country and between the north and south.

	Region	Nationwide		The North		The South
Variable		Standardized Coefficient	p	Standardized Coefficient	p	Standardized Coefficient	p
Intercept		0.0000	0.0000	0.000	0.000	0.0000	0.000
Green Credit		0.3508	0.4740	0.3209	0.525	0.3923	0.478
Green Investment		−0.0363	0.1560	−0.0962	0.228	0.0005	0.568
Green Insurance		0.1156	0.0600	0.2495	0.004	−0.0036	0.402
Green Bonds		0.0501	0.0190	0.0380	0.164	0.0689	0.026
Green Support		−0.0141	0.5180	−0.0161	0.011	−0.0079	0.487
Green Fund		0.5362	0.0000	0.4519	0.000	0.5931	0.000
Green Equity		−0.0023	0.5030	0.0518	0.108	−0.0430	0.142

displays the national sample regression results of green finance differences in China. The differences in the seven structural factors have impacts on the differences in green finance, both positively and negatively. For the whole country, the difference in green funds had the largest effect intensity, and the standardized regression coefficient reached 0.536. The second was the driving effect of green credit difference, and the standardized regression coefficient was 0.350. Then, the effect intensity had a green insurance difference of 0.1156 and a green bond difference of 0.050. In addition, the differences in green investment, green support, and green equity had a negative correlation on the differences in green finance. Therefore, the expansion of green investment difference, green support difference, and green equity difference will narrow the green finance difference. Different regions can greatly improve the level of green finance by adopting policies according to local conditions and classifications. In northern China, the difference in green funds is the largest, and the standardized regression coefficient reaches 0.452. This is followed by green credit difference and green insurance difference, with standardized regression coefficients, respectively, of 0.321 and 0.250. The strength of the green bond difference is 0.038. The regression coefficient of the green interest difference is 0.0518. The regression coefficient of green investment and green support is negative. In the southern region, the difference in green funds is the largest, and the standardized regression coefficient is 0.593. The second is the driving effect of green credit difference, and the regression coefficient is 0.3923. The third is the green bond difference and the green investment difference. The effect directions of the green insurance difference, green support difference, and green rights difference are negative.

The standardized regression coefficient also reflects how the effect of the difference in green bonds and green funds on the differences in green finance are stronger from north to south, while the effect of the differences in green credit and green support are weaker from north to south.

The differences in the standardization coefficient show that the driving mechanisms of green finance are different in the north and the south. In the north, the degree of marketization is relatively weak, and green finance has certain characteristics of “political achievements”. It relies mainly on executive orders to promote green finance. This is good for green credit and green support, which are relatively easy to administrate. There is a high degree of marketization in the south, and green finance has gained some incentives in the private sector, benefiting market-driven products such as green bonds and green funds. Examining the internal mechanisms of green difference between the north and south, we can see the endogenous driving force of economic development in north and south China. This has important critical significance for understanding the non-equilibrium of the north–south development of green finance in China’s economic reality. At the same time, it can also be seen that the north’s mandatory promotion of a certain green financial product has not been coordinated by the private sector in other products, and the power of administrative intervention needs to be re-examined.

6. Conclusions

A green finance index system including green credit, green investment, green insurance, green support, green bonds, green funds, and green equity as sub-dimensions has been established. The green finance index was measured using the entropy method. This paper analyzed the spatial differences, structural sources, and formation mechanisms of green finance in order to provide references for promoting the rational and orderly improvement of green finance. The main conclusions are as follows.

First, although the overall level of green finance was not high, it showed a slow growth trend in fluctuations. From a multi-dimensional perspective, all seven indicators were on the rise within the observation period. The level of green finance in southern provinces was higher than that in northern ones. The average green finance index in southern areas was 0.298, while it was 0.251 in the north.

Second, the green financial index Gini coefficient had a relatively stable trend. The Gini coefficient was between 0.212 and 0.218, with a small fluctuation and a small downward trend in the fluctuation. The spatial difference regarding green finance in northern provinces proved an expanding tendency, while the spatial difference in the southern ones showed a narrowing trend. Indices concerning green investment, green support, green fund, and green equity were the mainsprings for the differences between the northern and southern provinces. The differences in indices relating to green credit, green insurance, and green bonds between north and south regions had downward trends. The differences in green investment index, green support index, green fund index, and green equity index between the north and south provinces showed an uptrend. The differences between the northern and southern provinces were the primary reason for the uneven distribution of green finance, having an average contribution of 51.44%. Among the inter-regional differences between north and south, the contribution of the inter-regional hypervariable density difference far exceeded the contribution of the inter-regional net hypervariable value. Therefore, the cross-overlapping problem between different regions was the main reason for the uneven distribution of green finance across the country.

Third, the difference in green funds was the main structural source of the difference in green finance in China and the north and south provinces. The effect of the green bond difference and green fund difference on green finance differences increased from north to south. The differences in green credit and green support decreased from north to south. These were reflected in the standardized regression coefficients.

There were significant differences in green finance across the country and regions. To improve China’s green finance and promote coordinated development among regions, it is urgent to start from all dimensions of green finance. Therefore, the following suggestions are proposed.

First, in view of the fact that there is still a lot of room for improving the level of green finance in most cities, we should unswervingly make up for shortcomings and continue to improve the green finance level. Most cities have low levels of green bonds and support. Emphasis should be placed on expanding the number of green bonds issued and the scale of fiscal environmental protection expenditures to meet local environmental protection needs.

Secondly, there are large intra-regional and inter-regional differences in green finance across the country and in the north and south. Measures ought to be taken to promote the balanced development of green finance and enhance the green finance level. On the one hand, it is necessary to accelerate the improvement of the level of green finance in the northern region and narrow the gap between it and the southern region. The focus is on strengthening the level of green investment, green bonds, and green support in the northern region. It is also urgent to increase investment in environmental pollution control and fiscal expenditure on environmental protection, while encouraging the issuance of green bonds. On the other hand, it is indispensable to strengthen the “joint” innovation model and narrow the diversities in green finance among provinces in the southern region. Cooperation should be actively promoted between provinces with higher levels of green finance and lower cities to achieve complementary advantages and promote the leading role.

Different dimensions have different contributions and driving forces to the differences in urban green finance, and there is regional heterogeneity. It is necessary to correctly grasp the structural sources and formation mechanisms of green finance differences in different regions. In light of local conditions, the coordinated development of green finance can be quickly and effectively achieved.

At the national level, differences in green funds and green credit are the most prominent problems. Efforts must be made to solve the differences in green finance caused by the above differences. At the regional level, the southern provinces can reduce the differences in green finance by increasing the levels of green funds and green credit. The northern provinces can focus on reducing the green insurance disparity, so as to reduce the degree of green finance disparity.

Emphasis should be placed on collaboration between the private sector and the public sector. Although the mandatory introduction of certain products can shorten some differences, the development of green finance is not sustainable without the participation of private products. In terms of green support, the North should gradually integrate with the private sector and strive to achieve systematic growth in products.

The sustainability of green finance is a hot topic at present. From the analysis of the differences between the north and the south, we can see the differences in the internal driving forces of green development. Limited to the main research topics, this study does not detail research on a regional scale, administrative intervention, and private sector fit. There is no corresponding analysis on the performance of green finance, which will be discussed in future extended research.