Does financial agglomeration contribute to the coordination of pollution and carbon reduction? Test analysis based on the threshold model and spatial Durbin model

It is crucial for China to determine whether financial agglomeration can increase the degree of coordination between pollution and carbon reduction to expedite the achievement of full green economic transformation. This study, using the data of 283 cities in China from 2007 to 2021 as samples, innovatively explores the threshold effect and spatial effect of financial agglomeration (FA) on the coordination of pollution and carbon reduction (CPC) by constructing a panel threshold regression model and a spatial Durbin model. It also examines the spatial mechanism and regional heterogeneity of these two factors from a spatial perspective. The findings indicate the following: First, FA has a notable threshold effect on the CPC. The presence of threshold variables, such as government support, industrial structure, and financial development, results in the emergence of two distinct threshold effects. Conversely, when the threshold variable of FA is considered, only a single threshold effect is observed. And the threshold value of FA is 0.2972. Second, throughout the research period, both the degree of FA and the level of the CPC exhibit notable characteristics of positive spatial correlation. Third, the impact of FA on the CPC exhibits a nonlinear “U-shaped”characteristic in spatial terms, with government support and human capital serving as the underlying mechanisms. Fourth, the spatial impact of FA and its threshold effect vary by location. This study deeply explores the influence mechanism of FA on CPC, and puts forward policy suggestions relevant policy suggestions from the perspectives of government support, adjustment of industrial frameworks, and progress in the financial sector, which can provide policy references for enhancing the level of CPC of China.

1 Introduction

At present, China is under pressure to prevent air pollution and achieve the “dual carbon” objective (Chen et al., 2017; Zhu et al., 2023). On the one hand, while experiencing rapid economic growth, China is also facing the problem of excessive consumption of fossil fuels and severe air pollution (Shi et al., 2022). Through government intervention, the Chinese government has eliminated outdated and polluting facilities, reduced coal usage, and improved air quality (Labzovskii et al., 2019). The work on air pollution prevention and control in China has shifted from controlling the total amount of pollutants to regional joint prevention and control. However, the overall situation of air pollution control in China is still grim (Liu et al., 2023). According to the 2022 Communique regarding the condition of China’s ecological environment, 37.2% of the 339 cities at the prefecture level and higher, which amounts to 126 cities, exceeded the established environmental quality standards. Among them, 86 cities, or 25.4% of the total, had fine particulate matter (PM_2.5) levels above the standard level. Fifty-five cities, or 16.2%, had inhalable particulate matter (PM₁₀) levels above the recommended level. Ozone (O₃) levels were higher than the norm in 92 cities (27.1%). Despite this, the root, structural, and trend pressures for carbon reduction have not yet been alleviated (Swart et al., 2004; Ahmad et al., 2021). The traits of air pollutants and CO₂ emissions share the same root, origin, and process (Vandyck et al., 2018). In terms of both prevention and management, they share similarities (Rafaj et al., 2013; Yi et al., 2022). These findings provide scientific and operational support for achieving the coordination of pollution and carbon reduction (CPC). Moreover, driven by globalization and informatization, financial agglomeration (FA) presents a positive development trend (Ye et al., 2025; Zhai and An, 2021). China has developed a “core-periphery” pattern influenced by FA, where the eastern coastal cities serve as the core and the central and western inland cities function as the periphery. The Global Financial Index (GFCI) report released in March 2023 by the British think tank Z/Yen Group states that Shanghai is positioned as the seventh most prominent financial center globally, while Shenzhen and Beijing are ranked 12th and 13th, respectively. It is evident that the overall development trend of FA in China is positive. With the advancement of ecological civilization construction and the “carbon peak and carbon neutrality” initiative in China, the impact of FA on the CPC is becoming increasingly clear. On the one hand, FA can break through regional restrictions through external economies of scale, innovation incentives, resource allocation, and network synergy effects by guiding the cross-regional allocation of financial resources. It can provide necessary financial support, such as funds, talent, and technologies, for industrial technological innovation and industrial structure adjustment in neighboring regions; strengthen the investment of governments and enterprises in the CPC; and continuously overcome technical challenges in governance, thereby promoting the development of a green economy and the process of the CPC. On the other hand, owing to the negative effects of excessive competition and cost constraints, FA may also inhibit the CPC. The CPC is a complex project involving multiple fields, which requires efforts from systematic governance, comprehensive governance, and source governance to gradually achieve. It cannot be achieved without the strong support of financial tools, such as credit, bonds, funds, and insurance. The financial functions of FA can fully realize the positive interaction and connection between financial tools and the CPC and accelerate the process of the CPC. In addition, China’s FA shows diverse and heterogeneous spatial characteristics; thus, the effects of FA on the CPC vary across regions. As a product of the profound evolution of the financial industry, can FA help increase the level of the CPC? What is the mechanism of action and the spatial law between them? How strongly do FA and the CPC affect each other in space? Addressing these issues is crucial.

FA refers to the dynamic process through which financial elements and resources are gathered, coordinated, and integrated in a spatial context as the financial sector develops (Pandit et al., 2001). It includes various financial institutions, such as financial enterprises and financial intermediaries (Kindleberger, 1978). The research on FA has addressed the concept (Kukalis, 2010), formation motivation (Arunpold et al., 2014), and measurement method of FA (Cheng, 2016; Wang et al., 2020; Yuan et al., 2020) and the impact of FA on technological innovation (Sun and Li, 2022; Hu et al., 2023), industrial structure upgrading (Liu et al., 2024), resilience in urban economies (Ye et al., 2025; He et al., 2024), and environmental pollution (Yuan et al., 2022; Fan et al., 2023). Studies have indicated that financial concentration affects economies of scale (Ma and Stern, 2008; Tao et al., 2023) and has structural effects (Wen et al., 2021), resource allocation effects (Buera et al., 2011), and technology spillover effects (Goldin, 1966). It influences the distribution of financial resources across regions and offers essential financial assistance, including capital, skilled personnel, and technology, to surrounding areas to enhance environmental quality. On the other hand, owing to the negative impact of excessive competition and cost constraints, FA can also adversely affect the environment. Research on the CPC primarily emphasizes the core concept, horizontal measurement (Li et al., 2019; Zhang et al., 2022), space-time evolution (Chen et al., 2023; Hou et al., 2024; Meng et al., 2025), influencing factors (Zhang et al., 2023; Ren et al., 2024; Chen et al., 2024; Yin et al., 2024), and realization paths (Wang et al., 2024; Liang et al., 2024). On the basis of previous research and our study focus, we define the CPC as a state of coordinated control of air pollution reduction and carbon emission reduction, as well as a comprehensive green economic transformation, that is formed with the support of policies, structures, and technologies that target the common origin, process, and source of air pollutants and carbon emissions. Scholars have employed global air pollution simulation models, life cycle assessment methods, synergy coordinate systems and other methods to calculate the CPC level. The results reveal that environmental policies, technological progress, industrial structure, and energy structure affect the level of the CPC. Scholars have proposed development paths for the CPC from the perspectives of policy control, industrial structure optimization, technological innovation, multiparty governance, and the digital economy.

A literature review revealed that current research has the following shortcomings. First, few scholars have explored the relationship between FA and CPC. Numerous researchers have examined the impact of FA on environmental pollution and carbon emissions, but they have often adopted a limited perspective. Previous studies have only explored the impact of FA on environmental pollution or the influence of FA on carbon reduction. Second, the mechanisms through which FA impacts pollution levels and carbon reduction remain inadequately delineated. Despite some empirical evidence of how FA affects environmental pollution or carbon emissions, it is still necessary to investigate how FA impacts the CPC. Third, the strategies implemented to coordinate pollution control and carbon reduction have been predominantly focused on a single region, a single industry, or a single policy. From the perspective of cities, testing the impact of FA on the CPC is crucial. Thus, it is essential to develop a detailed and practical approach at the city level.

To address the limitations of the literature, we use data from 283 cities in China from 2007 to 2021 as a sample. From a nonlinear and spatial perspective, we empirically examine the threshold effect and spatial spillover effect of FA on the CPC in cities and propose refined and differentiated paths for the CPC. The innovation points of this study are as follows: (1) We fill the gap in research on the impact of FA on the CPC in the context of prominent environmental problems and propose policy suggestions as a reference for encouraging the alignment of efforts to reduce pollution and carbon emissions. (2) We investigate the effects and underlying mechanisms of FA on reducing urban pollution and enhancing synergies in carbon reduction from nonlinear and spatial viewpoints and offer empirical support for further expanding research in this area. (3) We employ different robustness testing methods to enhance the reliability of the research findings and analyze the heterogeneity of the empirical results to uncover the diverse impacts of FA in different regions.

The remainder of this paper is structured as follows: The second section presents a theoretical analysis and proposes research hypotheses. The third section outlines the research methodologies and data utilized. The fourth section examines the empirical findings. The fifth section presents the conclusion along with policy suggestions.

2 Theoretical analysis and the research hypothesis

2.1 Threshold effect of FA on the CPC

The threshold effect of FA on the CPC is reflected mainly in the following aspects. First, there are certain differences in the effects of different levels of government support on the CPC. When there is little government support, both the government’s environmental protection supervision and financial support are relatively weak, such that businesses are not motivated to conduct research and development on green technologies or energy saving and emission reduction projects. As a result, enterprises’ investment in environmental pollution and governance is relatively small, and thus, the promoting effect on the CPC is weak. When the government’s level of support is relatively high, policy subsidies and tax incentives represent positive signals of the government’s promotion of the CPC, which enhances enterprises’ confidence and enthusiasm in technological innovation and industrial transformation and facilitates the realization of the goals of the CPC. Second, the impact of FA on the CPC may have certain thresholds when manifested at different levels of industrial structure. According to the Pety–Clark theorem, with the development of FA, the industrial structure shows a gradually increasing trend in terms of the proportion of the secondary and tertiary industries. When the industrial structure is at a relatively low level, the production relations among industries tend to be simplified and homogenized, and the corresponding research and development level is rather low; thus, the promoting effect on the CPC is not obvious. As the industrial structure level continues to improve, enterprises devote increased research and development efforts to low-carbon technologies and actively promote their application and promotion in the industrial sector. Moreover, the increase in the proportion of the tertiary industry can provide a solid industrial foundation for the green transformation and development of enterprises, thereby reducing their energy consumption and promoting the CPC process. Third, the impact of FA on the CPC may be influenced by the level of financial development. When financial development is at a relatively low level, under the influence of production scale and economic development, financial development not only fails to counteract the negative effects of FA on the CPC but also may increase the pressure on the CPC. When financial development reaches a relatively advanced stage, it strengthens the positive role of FA in the CPC through technological innovation and energy transition, promoting the steady progress of such coordination. In light of the preceding analysis, this hypothesis is proposed in this study.

H1

H1. Government support, industrial structure, and financial development have a threshold effect on the process through which FA influences the CPC, such that the impact of FA on the CPC is nonlinear.

2.2 The spatial effect of FA on the CPC

There are two spatial effects of FA on the CPC: the polarization effect and the trickle-down effect. FA can lead to an imbalance in resource allocation among areas and a decline in the efficiency of resource utilization. In the early stage of the development of FA, the polarization effect causes resource elements, such as funds and talent, to flow rapidly to the FA area from surrounding areas. This results in a shortage of financial funds and the loss of financial talent for enterprises located in the surrounding areas and causes problems such as financing constraints and low production efficiency. This further has adverse effects on the expansion of the economic scale of enterprises, the development of technological innovation, and the transformation of the industrial structure. Thus, it has an inhibitory effect on the CPC in the surrounding areas. When FA develops to a certain extent, influenced by the trickle-down effect, FA can strengthen the cooperative relationships of cross-regional enterprises through information spillover. This can not only reduce the costs of the CPC but also promote the overall advancement of the CPC through means such as resource integration and industrial transfer. On the basis of the foregoing analysis, we propose the following hypothesis.

H2

H2. FA has a “U”-shaped spatial spillover effect on the CPC.

H3

H3. Government support and human capital play a moderating role in the “U”-shaped spatial spillover effect of FA on the CPC.

3 Methods and data

3.1 Model setting

3.1.1 Panel threshold regression model

With reference to Hansen (1999), we established a panel threshold regression model for regression analysis, as depicted in Formula 1:

${\text{CPC}}_{\text{it}}={\mathrm{\alpha }}_0+{\mathrm{\alpha }}_1{\text{FA}}_{\text{it}}\times \mathrm{I}\left({\mathrm{q}}_{\text{it}}\le \mathrm{\theta }\right)+{\mathrm{\alpha }}_2{\text{FA}}_{\text{it}}\times \mathrm{I}\left({\mathrm{q}}_{\text{it}}>\mathrm{\theta }\right)+{\mathrm{\alpha }}_3{\mathrm{X}}_{\text{it}}+{\mathrm{\delta }}_{\mathrm{i}}+{\mathrm{\epsilon }}_{\text{it}}(1)$

where CPC_it denotes the combined efforts aimed at reducing pollution and carbon emissions in city i during year t, and the FA_it variable denotes the level of financial agglomeration in city i in year t. X_it represents the control variable. The symbol I (·) denotes the indicator function, which is equal to 1 if the condition within the parentheses is met and 0 if it is not met. q_it stands for a collection of threshold variables. δ_i denotes the individual effect. ε_it symbolizes the unpredictable factor that causes disturbances in the system.

Since there may be more than one threshold variable, the single-threshold model is expanded to include multiple thresholds in our research. The equation represented by Formula 2 is

${\text{CPC}}_{\text{it}}={\mathrm{\alpha }}_0+{\mathrm{\alpha }}_1{\text{FA}}_{\text{it}}\times \mathrm{I}\left({\mathrm{q}}_{\text{it}}\le {\mathrm{\theta }}_1\right)+{\mathrm{\alpha }}_2{\text{FA}}_{\text{it}}\times \mathrm{I}\left({{\mathrm{\theta }}_1<\mathrm{q}}_{\text{it}}\le {\mathrm{\theta }}_2\right)+···+{\mathrm{\alpha }}_{\mathrm{n}-1}{\text{FA}}_{\text{it}}\times \mathrm{I}\left({{\mathrm{\theta }}_{\mathrm{n}-1}<\mathrm{q}}_{\text{it}}\le {\mathrm{\theta }}_{\mathrm{n}}\right){+{\mathrm{\alpha }}_{\mathrm{n}}{\text{FA}}_{\text{it}}\times \mathrm{I}\left({\mathrm{q}}_{\text{it}}>{\mathrm{\theta }}_{\mathrm{n}}\right)+\mathrm{\alpha }}_{\mathrm{n}+1}{\mathrm{X}}_{\text{it}}+{\mathrm{\delta }}_{\mathrm{i}}+{\mathrm{\epsilon }}_{\text{it}}(2)$

where θ₁, θ₂, ···, and θ_n denote the threshold values associated with various threshold variables.

3.1.2 Spatial measurement model

Furthermore, on the basis of the studies of scholars such as Chen et al. (2022) and Elhorst (2012), the following three spatial econometric models are developed in this study.

The spatial autoregressive (SAR) model is shown in Formula 3:

${\text{CPC}}_{\text{it}}={\mathrm{\alpha }}_0+\mathrm{\rho }\mathrm{W}\times {\text{CPC}}_{\text{it}}+{\mathrm{\alpha }}_1{\text{FA}}_{\text{it}}+{\mathrm{\alpha }}_2{\text{FA}}_{\text{it}}^2+{\mathrm{\alpha }}_3{\mathrm{X}}_{\text{it}}+{\mathrm{\delta }}_{\mathrm{i}}+{\mathrm{\lambda }}_{\mathrm{t}}+{\mathrm{\epsilon }}_{\text{it}}(3)$

The spatial error model (SEM) is shown in Formula 4:

${\text{CPC}}_{\text{it}}={\mathrm{\beta }}_0+{\mathrm{\beta }}_1{\text{FA}}_{\text{it}}+{\mathrm{\beta }}_2{\text{FA}}_{\text{it}}^2+{\mathrm{\beta }}_3{\mathrm{X}}_{\text{it}}+{\mathrm{\varphi }}_{\text{it}}+{\mathrm{\delta }}_{\mathrm{i}}+{\mathrm{\lambda }}_{\mathrm{t}}(4)$

${\mathrm{\varphi }}_{\text{it}}=\mu \mathrm{W}{\mathrm{\varphi }}_{\text{it}}+{\mathrm{\epsilon }}_{\text{it}}$

The spatial Durbin model (SDM) is shown in Formula 5:

${\text{CPC}}_{\text{it}}={\mathrm{\gamma }}_0+\mathrm{\rho }\mathrm{W}\times {\text{CPC}}_{\text{it}}+{\mathrm{\gamma }}_1{\text{FA}}_{\text{it}}+{\mathrm{\gamma }}_2{\text{FA}}_{\text{it}}^2+{\mathrm{\gamma }}_3\mathrm{W}\times {\text{FA}}_{\text{it}}+{\mathrm{\gamma }}_4\mathrm{W}\times {\text{FA}}_{\text{it}}^2+{\mathrm{\gamma }}_5{\mathrm{X}}_{\text{it}}+{\mathrm{\gamma }}_6\mathrm{W}\times {\mathrm{X}}_{\text{it}}+{\mathrm{\delta }}_{\mathrm{i}}+{\mathrm{\lambda }}_{\mathrm{t}}+{\mathrm{\epsilon }}_{\text{it}}(5)$

Among them, CPC_it signifies the degree of collaboration in reducing pollution and carbon emissions, and FA_it represents the level of financial agglomeration. X_it represents the control variable. ρ signifies the coefficient for spatial autoregression. W represents the matrix of spatial weights. µ is the spatial error coefficient. δ_i represents the individual fixed effect, and λ_t represents the time fixed effect. ε_it represents the random disturbance in the model.

To carry out a more in-depth analysis of how FA influences the overall CPC in various spatial contexts, three spatial weight matrices were developed following the research framework proposed by Zhao et al. (2022) and Kubara and Kopczewska (2024), with 283 Chinese cities as samples: a matrix of spatial adjacency weights (W₁), a matrix of spatial inverse distance squared (W₂), and a spatial weight matrix of socioeconomic characteristics (W₃).

To investigate the potential moderating effects of government support and human capital on the “U”-shaped spatial relationship between FA and the CPC, we developed the following regression model:

$\begin{array}{ll}\hfill {\text{CPC}}_{\text{it}}=& {\mathrm{\pi }}_0+{\mathrm{\pi }}_1{\text{FA}}_{\text{it}}+{\mathrm{\pi }}_2{\text{FA}}_{\text{it}}^2+{\mathrm{\pi }}_3{\text{FA}}_{\text{it}}{\mathrm{M}}_{\text{it}}+{\mathrm{\pi }}_4{\text{FA}}_{\text{it}}^2{\mathrm{M}}_{\text{it}}+{\mathrm{\pi }}_5{\mathrm{M}}_{\text{it}}\hfill \\ \hfill & +{\mathrm{\pi }}_6{\mathrm{X}}_{\text{it}}+\mathrm{\rho }\mathrm{W}\times {\text{CPC}}_{\text{it}}+{\mathrm{\theta }}_1\mathrm{W}\times {\text{FA}}_{\text{it}}+{\mathrm{\theta }}_2\mathrm{W}\times {\text{FA}}_{\text{it}}^2\hfill \\ \hfill & +{\mathrm{\theta }}_3\mathrm{W}\times {\text{FA}}_{\text{it}}\times {\mathrm{M}}_{\text{it}}+{\mathrm{\theta }}_4\mathrm{W}\times {\text{FA}}_{\text{it}}^2\times {\mathrm{M}}_{\text{it}}+{\mathrm{\theta }}_5{\mathrm{W}\times \mathrm{M}}_{\text{it}}\hfill \\ \hfill & +{\mathrm{\theta }}_6\mathrm{W}{\mathrm{X}}_{\text{it}}+{\mathrm{\delta }}_{\mathrm{i}}+{\mathrm{\lambda }}_{\mathrm{t}}+{\mathrm{\epsilon }}_{\text{it}}\hfill \end{array}(6)$

Within this context, M_it refers to the moderating variables, which encompass government support and human capital. W denotes the spatial adjacency weight repetition matrix W₁. If $\left({\mathrm{\theta }}_3{\text{FA}}_{\text{it}}+{\mathrm{\theta }}_4{\text{FA}}_{\text{it}}^2+{\mathrm{\theta }}_5\right.$) takes on any value in FA_it and exceeds 0, it suggests that the moderating variable has the potential to increase the overall level of the “U”-shaped curve representing the synergy between FA and CPC. If ${\mathrm{\theta }}_1{\mathrm{\theta }}_4$ <${\mathrm{\theta }}_2{\mathrm{\theta }}_3$, it suggests that the moderating variable results in a leftward shift of the curve’s inflection point; conversely, the inflection point shifts to the right. If ${\mathrm{\theta }}_4$ <0, it suggests that the presence of the moderating variable causes the “U”-shaped curve to flatten; in its absence, the curve exhibits a steeper inclination.

3.2 Variable selection and statistical analysis of the data

3.2.1 Explained variables

The coordination of pollution and carbon reduction (CPC). We established two primary evaluation indices to measure the CPC. Metrics for assessing pollution reduction include the yearly average levels of PM_2.5, emissions of industrial smoke and dust, and emissions of sulfur dioxide from industrial sources. Industrial sulfur dioxide is among the main sources of air pollution. The use of industrial sulfur dioxide emissions can fully reflect the direct effect of pollution reduction in the industrial field. Industrial smoke and dust, which are among the main sources of particulate matter pollution, can cause haze pollution. Therefore, the emission volume of industrial smoke and dust can be used to reflect the degree of reduction in air pollution. As fine particulate matter, PM_2.5 poses a serious threat to human health. The annual average concentration of PM_2.5 can be regarded as an important indicator for measuring air pollution. Carbon reduction indicators include carbon dioxide emissions, green coverage in built-up areas, and overall electricity usage across society. The level of the CPC is assessed using the entropy method alongside the coupling coordination model. Carbon dioxide, as the primary greenhouse gas, is closely related to global warming and the greenhouse effect. The amount of carbon dioxide emissions can directly reflect the effect of carbon reduction. The overall electricity usage across society can reflect the level of energy consumption and thus affect carbon emissions. The green coverage area of built-up areas can reflect the carbon absorption capacity of a city. An increase in green area is conducive to increasing carbon sinks and therefore can be used as an indicator to measure carbon reduction. We adopted the methods of relevant scholars and employed the entropy method and the coupling coordination model to calculate the level of the CPC. Moreover, we believe that pollution reduction and carbon emission reduction are equally important; thus, we set the weights of both indicators to 0.5.

3.2.2 Explaining variables

Financial agglomeration (FA). We employ the financial geographic density method to evaluate the level of FA in urban areas throughout China. The level of FA is represented by the ratio of the year-end balance of deposits to loans relative to the built-up area.

3.2.3 Threshold variable

The threshold variables include government support (GS), industrial structure (IS), financial development (FD), and financial agglomeration (FA). Among them, the level of GS is indicated by the proportion of public budget expenditure to the gross domestic product (GDP), and a greater ratio indicates a higher degree of GS. The degree of IS is evaluated by comparing the added value created by the tertiary sector to the added value of the secondary sector. A higher ratio indicates an elevated level of IS. FD is quantified by the proportion of the loan balance maintained by financial institutions relative to GDP at the conclusion of the fiscal year. A higher ratio signifies a greater level of FD. In addition, FA is both an explanatory variable and a threshold variable in this paper.

3.2.4 Control variables

The control variables include the urbanization rate (UR), commercial scale (CS), energy consumption intensity (EI), and economic development level (GDP). The urbanization rate is quantified as the proportion of the population living in urban areas compared to the entire population. The level of CS is quantified by the numerical representation of the quantity of businesses that exceed this scale. The degree of EI is measured as the proportion of total energy utilized, represented in tons of standard coal, to the total regional gross domestic product, indicated in ten thousand yuan. The logarithmic transformation of per capita GDP serves as a proxy for the economic development level.

3.3 Data sources

The data for all the variables in this study primarily come from the China Environmental Statistical Yearbook, the China Urban Statistical Yearbook, the China Carbon Emission Database, and the EPS database. To address the issue of missing data, both interpolation techniques and value averaging methods are employed. Furthermore, to remove the influence of the price element, the data related to price are deflated by the GDP deflator.

Descriptive statistics were first obtained for the selected variables. The findings revealed that demonstrates a notable disparity between the lowest and highest values across all the variables. The variance inflation factor (VIF) test was employed to assess the interrelationships among all the variables. The findings revealed that the VIF values of all the variables were less than 10, indicating that no multicollinearity existed among the variables; thus, further empirical analysis can be undertaken.

4 Empirical results and analysis

4.1 Threshold effect analysis

4.1.1 Threshold effect test

Before applying the panel threshold regression model, we used the bootstrap self-sampling technique (Hansen, 2000) to conduct 1000 iterations to explore the presence of a threshold effect within the sample data. Table 1 presents the results of the threshold effect test with government support, industrial structure, financial development, and financial agglomeration as threshold variables.

Table 1

Table 1. Threshold effect test results.

According to Table 1, the threshold model concerning government support, industrial structure, and financial development as threshold variables successfully met the criteria for the double-threshold effect test. In contrast, the threshold model related to financial agglomeration as a threshold variable achieved success only in the single-threshold effect test. The above threshold test is explained by the likelihood ratio function graph. As shown in Figures 1–4, the LR values of the likelihood ratio statistics corresponding to the thresholds of government support, industrial structure, financial development, and financial agglomeration all approach 0 and are significantly less than 7.35, which supports the validity of the threshold estimation.

Figure 1

Figure 1. Results of the threshold effect of GS testing.

Figure 2

Figure 2. Results of the threshold effect of IS testing.

Figure 3

Figure 3. Results of the threshold effect of FD testing.

Figure 4

Figure 4. Results of the threshold effect of FA testing.

4.1.2 Examination of the outcomes of threshold regression analysis

The results of the panel threshold regression are presented in Table 2. In general, the effects of FA on the CPC present significant double threshold effects with regard to government support, industrial structure, and financial development and a single threshold effect with regard to financial agglomeration. The data in Table 3 clearly show that (1) when the GS is ≤ 0.1031, the influence coefficient of FA on CPC is −0.0079, which is statistically significant at the 1% level. These findings indicate that FA had a detrimental effect on the CPC during this period. The root of the issue can be linked to the following factors: A low level of government support can result in inadequate policy direction and a lack of oversight that lead to the misallocation of resources toward traditional polluting industries. Additionally, the absence of sufficient financial incentives from the government can hinder companies’ efforts in researching and developing green technologies, as well as their ability to support the CPC. When 0.1031<GS ≤ 0.2317, the FA coefficient is 0.0046 and meets the significance threshold at the 1% level, suggesting that government support can mitigate the adverse effects of FA on the CPC to some degree. When GS > 0.2317, the FA coefficient is 0.0134, and FA has a stronger promoting effect on the CPC. This may be because an increase in government support leads to the implementation of fiscal measures such as subsidies and tax incentives, which can motivate businesses to innovate and improve their production technologies. Moreover, when the government establishes standardized policies and regulatory frameworks for environmental protection, it compels companies to conserve energy and reduce emissions, thus improving the overall effectiveness of the CPC in the area. (2) When IS ≤ 0.7269, the coefficient of FA concerning CPC is negative, and the significance test at the 1% level shows that FA adversely affects the CPC. One potential explanation is that in regions where the industrial structure is underdeveloped, there is a heavy reliance on traditional manufacturing and energy-intensive sectors. During this phase, financial investments tend to gravitate toward industries that cause high pollution and emissions, leading to path dependency that can entrench high-carbon industries. This situation can negatively affect the alignment of efforts aimed at reducing pollution and cutting carbon emissions. When 0.7269<IS ≤ 0.8863, the estimated coefficient for FA remains negative; however, it does not achieve significance at the 10% level. This finding indicates that the adverse effect of FA on the CPC is not obvious at this time. When IS > 0.8863, the estimated coefficient of FA is 0.0089 and is significant at the level of 1%, indicating that when the industrial structure is developed enough, the improvement in FA can help improve the level of carbon reduction coordination. This can be attributed to the ongoing enhancement of the industrial structure, which leads to a greater allocation of financial resources to low-carbon and knowledge-intensive sectors. This shift encourages companies to invest in the research and development of clean technologies, creating a multiplier effect in technological innovation. Such developments are favorable for achieving the dual objectives of the CPC. (3) When FD ≤ 0.5638, the FA coefficient is −0.0348, suggesting that FA does not support the enhancement of the CPC. One potential explanation is that although financial development facilitates an increase in production scale and stimulates economic growth, it simultaneously results in issues such as air pollution and carbon emissions. This, in turn, negatively influences the CPC, exacerbating the detrimental effects of financial concentration on the CPC. When 0.5638<FD ≤ 0.8361, the coefficient for FA is −0.0072 and meets the significance criterion at the 1% level, which suggests that enhancing the level of financial development can mitigate the adverse effects of FA on the CPC. When FD > 0.8361, the adverse impact of FA on the CPC transforms into a beneficial effect. The reason for this could be that advanced financial development expands the avenues for investment and financing available to businesses. It directs financial resources toward research and development in green technologies, compelling companies to shift toward sustainable and low-carbon industrial practices and thus contributing positively to the CPC. (4) When FA≤0.2972, the FA coefficient is −0.0102 and significant at the 1% level, suggesting that enhancements in FA adversely affect the CPC. When FA>0.2972, the coefficient of FA is notably positive, suggesting that enhancing the level of FA at that time can greatly increase the CPC level.

Table 2

Table 2. Threshold effect regression results.

Table 3

Table 3. Results of the robustness validation on a threshold effect.

4.1.3 Robustness test of the threshold effect

Additionally, we implement the following methods to conduct robustness tests. (1) Tail reduction treatment. In this paper, all the variables were reduced by 1% at the top and bottom of the distribution, and the threshold effect regression was reperformed. (2) Excluding the data for provincial capitals and cities above the provincial city level. We removed the data from provincial capital cities and municipalities that are directly governed by the central government and then re-ran the regression analysis. (3) Deleting control variables. We eliminated the control variable related to GDP and performed the regression analysis again. As indicated by the results in Table 3, the estimated coefficients and significance levels of FA concerning the aforementioned threshold variables align closely with the baseline results, thereby confirming the robustness of the main regression outcomes.

4.2 Spatial effect analysis

4.2.1 Spatial correlation analysis

To assess whether a spatial relationship exists between FA and CPC, we employed the global Moran index method (Moran, 1950) to test the spatial correlation between regional FA levels and CPC levels under the spatial adjacency weight matrix (W₁). The findings revealed that the Moran index values for FA and CPC in China from 2007 to 2021 were consistently above 0. Furthermore, these values successfully passed the 1% significance test, demonstrating a notable positive spatial correlation and interdependence between FA and the CPC in the country.

To delve deeper into the local spatial agglomeration features of the relationship between financial concentration and the CPC, we replaced the W₁ matrix with a block matrix and introduced it into the local spatial autocorrelation test. On this basis, taking 2007, 2012, 2017, and 2021 as time nodes, local Moran scatter plots for China’s FA and CPC were created. The findings illustrated in Figures 5, 6 indicate that FA and the CPC present a spatial distribution that primarily consists of “high agglomeration” and “low agglomeration” zones, with a notable positive correlation between the two. This finding indicates that areas marked by significant FA (or CPC) are usually located next to other regions that exhibit comparably high levels, whereas zones with low FA (or CPC) tend to be encircled by other regions of low levels.

Figure 5

Figure 5. Moran scatter diagram for FA.

Figure 6

Figure 6. Moran scatter diagram for CPC.

4.2.2 Selection test of the spatial measurement model

The findings from the spatial correlation analysis indicate that FA and the CPC are significantly spatially correlated. Before the spatial effects of FA on the CPC are analyzed, an appropriate spatial econometric model must be chosen. With reference to Elhorst (2010), we employed the LM test, LR test, Hausman test, and Wald test to select appropriate spatial measurement models. We conducted model selection tests using three different weight matrices. The findings demonstrate that a spatial Durbin model with bidirectional fixed effects is appropriate for examining how FA spatially affects the CPC.

4.2.3 Examination of the outcomes of spatial regression

Grounded in the spatial adjacency weight matrix (W₁), spatial inverse distance square matrix (W₂), and spatial weight matrix of socioeconomic characteristics (W₃), the spatial Durbin model, which incorporates bidirectional fixed effects, was employed to empirically investigate the spatial impacts of FA on the CPC across various spatial weight matrices. The regression analysis results are presented in Table 4.

Table 4

Table 4. Spatial Durbin model datum regression results.

The autoregressive coefficients ρ of the CPC, calculated under the three different matrices, are 0.4923, 0.6867, and 0.2812 (Table 4). All these coefficients are statistically significant at the 1% level. This suggests that the CPC could generate a beneficial spatial spillover effect in neighboring areas. With respect to the control variables, the data in Table 4 demonstrate that (1) all the coefficients for the urbanization rate (UR) across the three matrices are positive and pass the 1% significance level test. This suggests that enhancing the urbanization rate can notably increase the level of the CPC within the spatial context. (2) The coefficient of industrial scale (CS) shows a notably positive effect only when the spatial weight matrix of socioeconomic characteristics is considered. This suggests that, within this specific spatial weight matrix, the industrial scale can greatly increase the level of the CPC. The reasons for this can be summarized as follows: On the one hand, the increase in industrial scale in adjacent regions not only promotes technological innovation among local enterprises but also generates an industrial clustering effect. This can facilitate collaboration, exchange, and sharing of technological experiences between neighboring and local areas, which supports the alignment of the CPC. On the other hand, the environmental degradation resulting from the growth of industrial scale in surrounding areas imposes environmental pressure on the region, prompting it to implement proactive environmental protection measures and increase the level of the CPC. (3) The energy consumption intensity (EI) coefficient is notably negative when the matrices of W₁ and W₂ are used. This finding indicates that an increase in energy consumption intensity adversely affects the regional CPC. This could be attributed to the rise in traditional energy use, which has led to greater environmental pollution issues and thereby diminished the combined benefits of the CPC. (4) The coefficient representing the level of economic development (GDP) is notably positive across all three spatial weight matrices, suggesting that increases in the GDP of adjacent regions can greatly increase the level of the CPC. This may be because demonstration effects and knowledge spillovers from regions with advanced economic development, which possess greater talent and technologies, increase the level of the CPC in a focal area.

The spatial regression results indicate that the spatial lag coefficient associated with the CPC is not statistically equal to zero. The spatial regression coefficient associated with FA does not provide an unbiased representation of the extent to which FA impacts the CPC in relation to both the region itself and adjacent areas. To strengthen the investigation of the explanatory variable’s marginal impact on the dependent variables, we use the partial differential method to analyze how FA spatially affects the CPC, utilizing three distinct spatial weight matrices. The data in Table 5 indicate that across the three matrices, the quadratic coefficients for the direct effect, indirect effect, and total effect of FA are all positive and successfully meet the significance level test of at least 5%. This finding indicates that FA in the neighboring region and the local region has a significant “U”-shaped nonlinear influence on the CPC.

Table 5

Table 5. Decomposition results of spatial effects.

4.2.4 Robustness test of spatial effects

To assess the robustness of the aforementioned research findings, we employed the following methods to perform robustness testing. (1) Adjusting the sample period. The sample period is adjusted from 2007 to 2021 to 2007–2019, and the spatial Durbin model is reintroduced for empirical analysis. (2) Replacing the matrix. Once again, we develop an economic distance spatial weight matrix (W₄) and utilized the spatial Durbin model for regression analysis. (3) Replacing the measurement model. The spatial impact of FA on the CPC is re-evaluated through the development of a dynamic spatial Durbin model that incorporates bidirectional fixed effects. Table 6 presents the findings from the robust regression analysis. These results attest to the reliability of the spatial regression outcomes.

Table 6

Table 6. Robustness test results of spatial effects.

4.2.5 Mechanism of action test

Table 7 display the decomposition of spatial effects regarding the impact of FA on the CPC, with government support and human capital serving as moderating factors. The findings indicate that both government support and human capital significantly influence the U-shaped spatial relationship between FA and the CPC. They contribute to an increase in the overall level of the U-shaped curve, resulting in a leftward shift of the inflection point and a flattening of the curve shape. One potential explanation is that the government facilitates financial assistance and innovative financial tools for the CPC by utilizing methods such as regional collaborative innovation, efficient cross-regional resource distribution, and the collaboration of financial platforms across regions. This approach allows the beneficial effects of FA on the CPC to be realized more swiftly. Areas with a significant amount of human capital enjoy benefits in innovation, particularly in information and technology. The transfer of knowledge and the spread of technology facilitate the research and implementation of green technologies in nearby regions, which helps enhance the overall level of the CPC. Additionally, a strong human capital base aids in lowering the risks and costs associated with investment and financing in the region, thus accelerating the realization of economies of scale from financial clustering and contributing positively to the advancement of the CPC.

Table 7

Table 7. The decomposition results of spatial effects under GS and HC regulation.

4.3 Analysis of regional heterogeneity

4.3.1 Analysis of the threshold effect across different regions

To assess whether the threshold effect exhibits regional differences, we categorized 283 sample cities in China into four distinct regions on the basis of the classification method established by the National Bureau of Statistics of China in 2011. The outcomes of the threshold regression for multiple regions are presented in Table 8. According to the data in Table 8, there is no single threshold for FA in the western region. This may be due to the relatively underdeveloped economy, insufficient policy guidance, and immature financial market in the western region, which prevent FA from exerting a threshold effect. Government support in the central region likewise fails to exert a threshold effect. In the eastern region, FA produces a dual threshold effect. Many of the valuation coefficients for FA in Northeast China do not meet the significance criteria, possibly because Northeast China depends heavily on its industrial sector, which makes environmental management challenging. Additionally, the slow progress of the green finance market and a lack of technological advancement hinder the impact of financial concentration on CPC, which leads to the absence of a noticeable threshold effect.

Table 8

Table 8. Regional heterogeneity results of threshold effect.

4.3.2 Heterogeneity analysis of spatial effects

The spatial impact of FA on the CPC was examined in greater detail, with a focus on regional differences. Table 9 displays the results of the spatial effects decomposition analysis. (1) From this table, it is evident that FA in the eastern and northeastern areas exerts a notable spatial U-shaped nonlinear effect on the CPC. Possible reasons are as follows: The average regional differences in FA in the eastern and northeastern regions are significantly greater than those in other regions. As the level of FA increases, the trickle-down effect comes into play, causing resources such as funds, talent, and technologies to spread to surrounding areas and thereby promoting the overall advancement of the CPC. (2) The direct effect of FA on the CPC in western China has a U-shaped nonlinear feature, whereas the indirect effect has an inverted U-shaped nonlinear feature. This might be because neighboring areas not only bring talent and technical expertise to this region, enhancing the level of the CPC, but also cause problems such as population congestion and environmental pollution, which have a negative impact on the CPC. (3) FA in central China does not have an obvious spatial nonlinear influence on the CPC. This may be because the central region depends heavily on industries that are polluting and emit high levels of emissions to achieve economic growth, and increased financial assistance is necessary to facilitate industrial transformation and foster innovation in green technology. The relationship between FA and the CPC does not show obvious spatial nonlinear characteristics.

Table 9

Table 9. Spatial effect regional heterogeneity regression results.

5 Conclusion and suggestions for policy action

Utilizing panel data from 283 Chinese cities over a 15-year period (2007–2021), this study empirically explored the threshold and spatial impacts of FA on the CPC by using a panel threshold regression model and spatial Durbin model under two-way fixed effects. The findings indicate the following: (1) FA has a threshold effect on the CPC. Government support, industrial structure, and financial development exhibit a dual threshold effect, and FA has a single threshold effect. There is regional heterogeneity in the threshold effect of FA on the CPC. (2) From 2007 to 2021, both the FA level and the CPC level showed significant positive spatial correlation characteristics. Both show the characteristics of “high-high agglomeration” and “low-low agglomeration” in terms of geographical distribution. (3) A notable U-shaped spatial relationship exists between FA and the CPC, in which government assistance and human capital serve as moderating factors. Both factors have the potential to increase the overall level of the U-shaped curve, shift the curve’s inflection point to the left, and result in a flatter configuration. (4) Considering the variations across different regions, under the three matrices, a U-shaped nonlinear correlation exists between FA and the CPC in the eastern and northeastern regions. However, the western and central regions do not show an obvious U-shaped nonlinear relationship.

On the basis of the above research, policy recommendations are developed mainly from the following three aspects: (1) Considering the threshold variables that affect the CPC through FA, government support, industrial structure, and financial development levels should be enhanced, and the positive promoting effect of FA on the CPC should be strengthened. (2) Considering the spatial spillover effect of FA on CPC, regional financial cooperation should be strengthened, and the level of FA should be improved to maximize the positive spatial spillover effect of FA on the CPC. (3) Considering the regional heterogeneity results of threshold effects and spatial effects, corresponding policies should be proposed to enhance the targeted nature of the policies.

First, government support and guidance should be increased. The findings indicate that when government support surpasses the initial threshold value of 0.1031, FA has a notable positive effect on the CPC. Under the National Implementation Plan aimed at enhancing the efficiency of the CPC and in light of their actual development conditions, local governments should propose specific collaborative promotion plans for the CPC in key areas such as industry, agriculture, transportation, and ecological construction. The government should promote green innovative businesses by providing tax incentives, green subsidies, and various other methods to support their research and development of low-carbon environmental protection technologies.

Second, the transformation and upgrading of industrial structures should be promoted. The findings indicate that once the industrial structure surpasses the second threshold of 0.8863, FA can notably increase the level of the CPC. The government should promote the adoption of the internet, blockchain, and various digital technologies among businesses to modernize and enhance the traditional manufacturing sector. This initiative aims to reduce excessive consumption and inefficient production capabilities and ultimately rejuvenate the traditional manufacturing industry. Simultaneously, the government should anchor the green transformation and development of the industrial structure; actively improve the policy support system related to green production, green consumption, and green technology; and strengthen policy synergy through carbon trading and environmental regulations to promote green and low-carbon transformation.

Third, high-quality financial development should be promoted. The results show that when the level of financial development is greater than the second threshold value of 0.8361, FA has a notably beneficial effect on the CPC. The government can attract nonbank financial entities, including insurance firms, trust companies, and financial leasing organizations, to enter the market and expand the financial scale through tax incentives, fiscal subsidies, and incentives. Concurrently, accelerating the enhancement of the financial regulatory framework is essential. We need to create a robust legal system for financial regulation that focuses on minimizing financial risk and safeguarding financial security. Additionally, we should increase the efficiency of financial regulation and nurture the stable growth of the financial industry.

Fourth, financial coordination and cooperation among regions should be strengthened. The results of this study indicate that FA exerts a spatial spillover influence on the CPC. The government should create and enhance a mechanism for financial cooperation across regions; build a regional financial cooperation platform; promote the exchange of financial information among local governments, financial supervision and management departments, and financial enterprises; and carry out cooperation in financial business cooperation and financial risk prevention. This will increase the level of FA and amplify its beneficial impact on the CPC.

Finally, differentiated regional policies should be implemented. Government support, industrial structure, and financial development in the western region have a dual threshold effect on the effect of FA on the CPC. Therefore, efforts should be made to strengthen government support and financial development to promote industrial structure transformation in the western region. Moreover, the central region, which is a concentrated area of high-pollution and high-emission heavy industries, should accelerate the adjustment of the industrial structure and maintain an appropriate level of financial development and financial agglomeration, avoiding adverse pressure on the CPC. Additionally, regional and interregional financial cooperation should be strengthened in the eastern and northeastern regions to increase regional financial agglomeration and maximize the spatial spillover effect of FA on the CPC.

Compared with existing research, in previous studies, few studies have explored the relationship between FA and CPC. Moreover, previous studies have mainly focused on examining the impact of FA on the environment from the perspective of either pollution reduction or carbon emission reduction alone, lacking an exploration of the effects of FA on CPC from a dual perspective. This study innovatively incorporates FA and CPC into a unified research framework, opening up a new perspective on the research of FA and CPC. Based on panel data from 283 cities in China, this study empirically examines the threshold effect and spatial spillover effect of FA on CPC in cities from a nonlinear and spatial perspective. Furthermore, this study conducts a theoretical analysis of the mechanism by which financial agglomeration affects the synergy of pollution reduction and carbon emission reduction, and proposes corresponding hypotheses regarding the mechanism. It also uses empirical tools to support the arguments, providing a theoretical explanation and empirical basis for studying the relationship between financial agglomeration and the synergy of pollution reduction and carbon emission reduction. Based on the above, this study proposes refined and differentiated paths for CPC in cities, providing practical references and decision-making basis for China to promote CPC in a coordinated manner and achieve a win-win situation for environmental protection and economic development.

Moreover, this study has two limitations that require further exploration. On the one hand, being limited to the study of cities above the prefecture level, we selected panel data from 283 cities in China, covering the period from 2007 to 2021, as our sample for analysis. With the gradual advancement of the CPC, the period can be further expanded and updated in the future. On the other hand, the spatial impact of FA on the CPC might vary as geographical distance increases. In the future, a thorough investigation can be conducted on the spillover distance of the spatial effects of FA on the CPC.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Author contributions

SC: Conceptualization, Data curation, Methodology, Software, Validation, Visualization, Writing – original draft, Writing – review and editing. SZ: Writing – review and editing. YL: Writing – review and editing. YB: Funding acquisition, Project administration, Resources, Supervision, Writing – review and editing.

Funding

The author(s) declared that financial support was received for this work and/or its publication. The research was supported by the National Social Science Project (No. 16BGL199).

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fenvs.2025.1704922/full#supplementary-material

References

Ahmad, T., Zhang, D., Huang, C., Zhang, H., Dai, N., Song, Y., et al. (2021). Artificial intelligence in sustainable energy industry: Status quo, challenges and opportunities. J. Clean. Prod. 289, 125834. doi:10.1016/j.jclepro.2021.125834

CrossRef Full Text | Google Scholar

Arunpold, S., Tripathi, N. K., Rajesh Chowdhary, V., and Raju, D. K. (2014). Comparison of GPS-TEC measurements with IRI-2007 and IRI-2012 modeled TEC at an equatorial latitude station, Bangkok, Thailand. J. Atmos. Solar-Terrestrial Phys. 117, 88–94. doi:10.1016/j.jastp.2014.06.001

CrossRef Full Text | Google Scholar

Buera, F. J., Kaboski, J. P., and Shin, Y. (2011). Finance and development: a tale of two sectors. Am. Econ. Rev. 101, 1964–2002. doi:10.1257/aer.101.5.1964

CrossRef Full Text | Google Scholar

Chen, Q., Cai, B., Dhakal, S., Pei, S., Liu, C., Shi, X., et al. (2017). CO2 emission data for Chinese cities. Resour. Conservation Recycl. 126, 198–208. doi:10.1016/j.resconrec.2017.07.011

CrossRef Full Text | Google Scholar

Chen, Y., Shao, S., Fan, M., Tian, Z., and Yang, L. (2022). One man's loss is another's gain: does clean energy development reduce CO2 emissions in China? Evidence based on the spatial durbin model. Energy Econ. 107, 105852. doi:10.1016/j.eneco.2022.105852

CrossRef Full Text | Google Scholar

Chen, S., Tan, Z., Wang, J., Zhang, L., He, X., and Mu, S. (2023). Spatial and temporal evolution of synergizing the reduction of pollution and carbon emissions and examination on comprehensive pilot effects–evidence from the national eco-industrial demonstration parks in China. Environ. Impact Assess. Rev. 101, 107147. doi:10.1016/j.eiar.2023.107147

CrossRef Full Text | Google Scholar

Chen, Z., Shi, Y., and Ding, R. (2024). Assessing the utility of low-emission pilot policies for facilitating pollution reduction and carbon mitigation: an empirical investigation using multi-temporal double difference analysis. J. Environ. Manag. 371, 123196. doi:10.1016/j.jenvman.2024.123196

PubMed Abstract | CrossRef Full Text | Google Scholar

Cheng, Z. (2016). The spatial correlation and interaction between manufacturing agglomeration and environmental pollution. Ecol. Indic. 61, 1024–1032. doi:10.1016/j.ecolind.2015.10.060

CrossRef Full Text | Google Scholar

Elhorst, J. P. (2010). Applied spatial econometrics: raising the bar. Spat. Economic Analysis 5, 9–28. doi:10.1080/17421770903541772

CrossRef Full Text | Google Scholar

Elhorst, J. P. (2012). Dynamic spatial panels: models, methods, and inferences. J. Geogr. Syst. 14, 5–28. doi:10.1007/s10109-011-0158-4

CrossRef Full Text | Google Scholar

Fan, W., Wang, F., Liu, S., Chen, T., Bai, X., and Zhang, Y. (2023). How does financial and manufacturing co-agglomeration affect environmental pollution? Evidence from China. J. Environ. Manag. 325, 116544. doi:10.1016/j.jenvman.2022.116544

PubMed Abstract | CrossRef Full Text | Google Scholar

Goldin, K. D. (1966). Economic growth and the individual. J. Finance 21, 550–551. doi:10.2307/2977834

CrossRef Full Text | Google Scholar

Hansen, B. E. (1999). Threshold effects in non-dynamic panels: estimation, testing, and inference. J. Econ. 93, 345–368. doi:10.1016/S0304-4076(99)00025-1

CrossRef Full Text | Google Scholar

Hansen, B. E. (2000). Sample splitting and threshold estimation. Econometrica. 68, 575–603. doi:10.1111/1468-0262.00124

CrossRef Full Text | Google Scholar

He, B., Tang, M., and Zhang, C. (2024). Financial technology, regional financial agglomeration, and urban economic resilience. Finance Res. Lett. 68, 105990. doi:10.1016/j.frl.2024.105990

CrossRef Full Text | Google Scholar

Hou, M., Cui, X., Xie, Y., Lu, W., and Xi, Z. (2024). Synergistic emission reduction effect of pollution and carbon in China's agricultural sector: regional differences, dominant factors, and their spatial-temporal heterogeneity. Environ. Impact Assess. Rev. 106, 107543. doi:10.1016/j.eiar.2024.107543

CrossRef Full Text | Google Scholar

Hu, Y., Wang, C., Zhang, X., Wan, H., and Jiang, D. (2023). Financial agglomeration and regional green innovation efficiency from the perspective of spatial spillover. J. Innovation and Knowl. 8, 100434. doi:10.1016/j.jik.2023.100434

CrossRef Full Text | Google Scholar

Kindleberger, C. P. (1978). Economic response: comparative studies in trade, finance, and growth. Harvard University Press. doi:10.4159/harvard.9780674418875

CrossRef Full Text | Google Scholar

Kubara, M., and Kopczewska, K. (2024). Akaike information criterion in choosing the optimal k-nearest neighbours of the spatial weight matrix. Spat. Econ. Anal. 19, 73–91. doi:10.1080/17421772.2023.2176539

CrossRef Full Text | Google Scholar

Kukalis, S. (2010). Agglomeration economies and firm performance: the case of industry clusters. J. Manag. 36, 453–481. doi:10.1177/0149206308329964

CrossRef Full Text | Google Scholar

Labzovskii, L. D., Mak, H. W. L., Takele Kenea, S., Rhee, J. S., Lashkari, A., Li, S., et al. (2019). What can we learn about effectiveness of carbon reduction policies from interannual variability of fossil fuel CO₂ emissions in east Asia? Environ. Sci. Policy 96, 132–140. doi:10.1016/j.envsci.2019.03.011

CrossRef Full Text | Google Scholar

Li, N., Chen, W., Rafaj, P., Kiesewetter, G., Schöpp, W., Wang, H., et al. (2019). Air quality improvement Co-benefits of low-carbon pathways toward well below the 2 °C climate target in China. Environ. Sci. Technol. 53 (10), 5576–5584. doi:10.1021/acs.est.8b06948

PubMed Abstract | CrossRef Full Text | Google Scholar

Liang, C., Chen, X., and Di, Q. (2024). Path to pollution and carbon reduction synergy from the perspective of the digital economy: fresh evidence from 292 prefecture-level cities in China. Environ. Res. 252, 119050. doi:10.1016/j.envres.2024.119050

PubMed Abstract | CrossRef Full Text | Google Scholar

Liu, L., Zhang, L., Li, B., Wang, Y., and Wang, M. (2024). Can financial agglomeration curb carbon emissions reduction from agricultural sector in China? Analyzing the role of industrial structure and digital finance. J. Clean. Prod. 440, 140862. doi:10.1016/j.jclepro.2024.140862

CrossRef Full Text | Google Scholar

Liu, Z., Tang, Y., Wilson, J., Tao, X., Lv, B., Wang, Z., et al. (2023). Influence of government attention on environmental quality: an analysis of 30 provinces in China. Environ. Impact Assess. Rev. 100, 107084. doi:10.1016/j.eiar.2023.107084

CrossRef Full Text | Google Scholar

Ma, C., and Stern, D. I. (2008). China's changing energy intensity trend: a decomposition analysis. Energy Econ. 30, 1037–1053. doi:10.1016/j.eneco.2007.05.005

CrossRef Full Text | Google Scholar

Meng, Q., Chen, X., Wang, H., Shen, W., Duan, P., and Liu, X. (2025). Spatiotemporal evolution and driving factors of the synergistic effects of pollution control and carbon reduction in China. Ecol. Indic. 170, 113103. doi:10.1016/j.ecolind.2025.113103

CrossRef Full Text | Google Scholar

Moran, P. A. (1950). Notes on continuous stochastic phenomena. Biometrika 37, 17–23. doi:10.2307/2332142

PubMed Abstract | CrossRef Full Text | Google Scholar

Pandit, N. R., Cook, G., and Swann, P. (2001). The dynamics of industrial clustering in British financial services. Serv. Industries J. 21, 33–61. doi:10.1080/714005045

CrossRef Full Text | Google Scholar

Rafaj, P., Schöpp, W., Russ, P., Heyes, C., and Amann, M. (2013). Co-benefits of post-2012 global climate mitigation policies. Mitig. Adapt. Strategies Glob. Change 18, 801–824. doi:10.1007/s11027-012-9390-6

CrossRef Full Text | Google Scholar

Ren, Z., Zheng, J., Jiao, L., He, M., Hou, X., Coffman, D., et al. (2024). The synergic impacts of air pollution control policies on pollutants and carbon emissions. J. Environ. Manag. 370, 122730. doi:10.1016/j.jenvman.2024.122730

PubMed Abstract | CrossRef Full Text | Google Scholar

Shi, Q., Zheng, B., Zheng, Y., Tong, D., Liu, Y., Ma, H., et al. (2022). Co-benefits of CO₂ emission reduction from China’s clean air actions between 2013-2020. Nat. Commun. 13, 5061. doi:10.1038/s41467-022-32656-8

PubMed Abstract | CrossRef Full Text | Google Scholar

Sun, S., and Li, X. (2022). Productive services agglomeration, financial development and regional innovation efficiency in the yangtze river econom ic zone. Finance Res. Lett. 50, 103221. doi:10.1016/j.frl.2022.103221

CrossRef Full Text | Google Scholar

Swart, R., Amann, M., Raes, F., and Tuinstra, W. (2004). A good climate for clean air: linkages between climate change and air pollution. An editorial essay. Clim. Change 66, 263–269. doi:10.1023/B:CLIM.0000044677.41293.39

CrossRef Full Text | Google Scholar

Tao, M., Silva, E., Sheng, M. S., Wen, L., and Qi, L. (2023). How financial clustering influences China's green development: mechanism investigation and empirical discussion. J. Environ. Manag. 347, 119081. doi:10.1016/j.jenvman.2023.119081

PubMed Abstract | CrossRef Full Text | Google Scholar

Vandyck, T., Keramidas, K., Kitous, A., Spadaro, J. V., Van Dingenen, R., Holland, M., et al. (2018). Air quality co-benefits for human health and agriculture counterbalance costs to meet paris agreement pledges. Nat. Commun. 9, 4939. doi:10.1038/s41467-018-06885-9

PubMed Abstract | CrossRef Full Text | Google Scholar

Wang, N., Zhu, Y., and Yang, T. (2020). The impact of transportation infrastructure and industrial agglomeration on energy efficiency: evidence from China’s industrial sectors. J. Clean. Prod. 244, 118708. doi:10.1016/j.jclepro.2019.118708

CrossRef Full Text | Google Scholar

Wang, N., Wang, S., Li, Y., Zhao, Y., Chun, T., and Zhang, R. (2024). Extracting critical paths for synergistic control of carbon emissions and air pollution: case of Henan Province. J. Environ. Sci. 155, 235–249. doi:10.1016/j.jes.2024.05.011

PubMed Abstract | CrossRef Full Text | Google Scholar

Wen, H., Lee, C. C., and Zhou, F. (2021). Green credit policy, credit allocation efficiency and upgrade of energy-intensive enterprises. Energy Econ. 94, 105099. doi:10.1016/j.jenvman.2021.114136

CrossRef Full Text | Google Scholar

Ye, Z., Li, J., and Chen, J. (2025). New evidence for the impact of financial agglomeration on urbanization from a spatial econometrics analysis. J. Clean. Prod. 200, 65–73. doi:10.1016/j.iref.2024.103764

CrossRef Full Text | Google Scholar

Ye, Z., Li, J., and Chen, J. (2025). The promotion mechanism of financial agglomeration and human capital on urban economic resilience: based on the moderating effect of industrial structure. Int. Rev. Econ. and Finance 97, 103764. doi:10.1016/j.iref.2024.103764

CrossRef Full Text | Google Scholar

Yi, H., Zhao, L., Qian, Y., Zhou, L., and Yang, P. (2022). How to achieve synergy between carbon dioxide mitigation and air pollution control? Evidence from China. Sustain. Cities Soc. 78, 103609. doi:10.1016/j.scs.2021.103609

CrossRef Full Text | Google Scholar

Yin, Q., Huang, Y., Ding, C., and Jing, X. (2024). Towards sustainable development: can green digital finance become an accelerator for reducing pollution and carbon emissions in China? Sustain. Cities Soc. 114, 105722. doi:10.1016/j.scs.2024.105722

CrossRef Full Text | Google Scholar

Yuan, H., Feng, Y., Lee, J., Liu, H., and Li, R. (2020). The spatial threshold effect and its regional boundary of financial agglomeration on green development: a case study in China. J. Clean. Prod. 244, 118670. doi:10.1016/j.jclepro.2019.118670

CrossRef Full Text | Google Scholar

Yuan, H., Zhang, T., Hu, K., Feng, Y., Feng, C., and Jia, P. (2022). Influences and transmission mechanisms of financial agglomeration on environmental pollution. J. Environ. Manag. 303, 114136. doi:10.1016/j.jenvman.2021.14136

PubMed Abstract | CrossRef Full Text | Google Scholar

Zhai, X., and An, Y. (2021). The relationship between technological innovation and green transformation efficiency in China: an empirical analysis using spatial panel data. Technol. Soc. 64, 101498. doi:10.1016/j.techsoc.2020.101498

CrossRef Full Text | Google Scholar

Zhang, Q., Cai, B., Wang, M., Wang, J., Xing, Y., Dong, G., et al. (2022). City level CO₂ and local air pollutants co-control performance evaluation: a case study of 113 key environmental protection cities in China. Adv. Clim. Change Res. 13 (1), 118–130. doi:10.1016/j.accre.2021.10.002

CrossRef Full Text | Google Scholar

Zhang, B., Wang, N., Yan, Z., and Sun, C. (2023). Does a mandatory cleaner production audit have a synergistic effect on reducing pollution and carbon emissions? Energy Policy 182, 113766. doi:10.1016/j.enpol.2023.113766

CrossRef Full Text | Google Scholar

Zhao, J., Jiang, Q., Dong, X., Dong, K., and Jiang, H. (2022). How does industrial structure adjustment reduce CO2 emissions? Spatial and mediation effects analysis for China. Energy Econ. 105, 105704. doi:10.1016/j.eneco.2021.105704

CrossRef Full Text | Google Scholar

Zhu, J., Wu, S., and Xu, J. (2023). Synergy between pollution control and carbon reduction: china's evidence. Energy Econ. 119, 106541. doi:10.1016/j.eneco.2023.106541

CrossRef Full Text | Google Scholar