Using the panel data of 284 cities from 2011 to 2020 in China, this research statistically tests the direct impact and internal mechanism of digital finance on urban haze governance. The results show that: (1) The development of digital finance can significantly inhibit the concentration of urban haze, and a stronger inhibitory effect in areas where the government pays more emphasis to haze governance and in cities with high level; (2) After mechanism inspection, it is found that digital finance can indirectly promote urban haze governance by influencing green innovation, cooperative innovation, industrial structure upgrading and producer services agglomeration; (3) The results of spatial econometric analysis show that digital finance can suppress the haze concentration in the region, and simultaneously inhibit the neighboring areas through spillover effects; (4) Further inspection shows that the spatial spillover effect of digital finance on haze governance has an obvious spatial attenuation feature, demonstrating that a dense area of spatial spillover is within 310 km. The spillover effect gradually disappears when the threshold is exceeded.
Research Article
Does the Development of Digital Finance Promote the Haze Governance? ——Empirical evidence from 284 cities in China
https://doi.org/10.21203/rs.3.rs-2347968/v1
This work is licensed under a CC BY 4.0 License
Journal Publication
published 01 Mar, 2023
Read the published version in Environmental Science and Pollution Research →
You are reading this latest preprint version
Digital finance
Haze governance
Green innovation
Cooperative innovation
Industrial structure upgrading
Producer services agglomeration
Spatial spillover effect
Since the reform and opening-up, the Chinese economic development is renowned as the miracle in the history of global economy. However, with the rapid development of economy, such growth model in the characterization of traditional extensity is in gradual discrepancy with environment, climate, and so forth. Within these factors, the smog pollution is exclusively transparent, notably for the imperil in public health (Madrigano et al, 2010; Cao and Chen, 2021), transport safety (Wan et al, 2020), and public security (Bondy et al, 2020). However, concerning that the Chinese haze pollution reveals extensive influence range, high occurrence frequency, long duration, and intensive difficulty in prevention and control, the aim at haze governance is hard to be realized if with the exclusive dependence on end-of-pipe treatment. While the rapid development of new generation of statistical technology such as Internet, big data, and artificial intelligence, the new opportunity is provided for the source governance in Chinese haze government.
As one of the typical representatives of the digital economy era, digital finance is a new financial model in which new digital technologies are deeply integrated with the traditional financial industry. It has the unique advantages of breaking the time and space limitations of traditional financial activities, lowering the threshold of financial services, saving transaction costs and alleviating information asymmetry (Li et al, 2020; Wan et al, 2022). Digital finance helps guide social capital from the "two high and one surplus" industries to the environment-friendly and green production industries, making efforts from the source to achieve coordination between economic and social development and ecological environmental protection. In October 2021, The State Council of China issued the Action Plan for Carbon Dioxide Peaking Before 2030, which clearly proposed to promote the integrated development of digital, intelligent and green industries. It is manageable to see that the deep integration of "digital" and "green" has become an inevitable choice to achieve high-quality economic development. In this context, exploring the haze emission reduction effect of digital financial development and its mechanism is of great practical significance in promoting environmental governance and winning the "battle of pollution prevention and control".
At present, many scholars regard digital finance as a "new engine" of economic growth and have conducted profound discussions on related issues, following that: (1) the connotation, development, and measurement of digital finance (Thorsten et al, 2007; Gomber et al, 2017); (2) the impact of digital finance on national macroeconomic development, such as economic growth (Sun and Tang , 2022), green development (Guo et al, 2022; Zhong, 2022), total factor productivity (Gao et al, 2022; Chen et al, 2022) and the quality of economic growth (Liu et al, 2022; Yin et al,2022); (3) how does digital finance affect resident consumption (Yang et al, 2022; Li et al,2020), innovation and entrepreneurship (Rao et al, 2022; Cao et al, 2021; Nambisan et al, 2019), industrial upgrading (Feng et al, 2022) and income disparity(Yao and Ma, 2022), and other social hot spots. Under the "carbon neutrality" and "emission peak" objective, the role of digital finance has also received extensive attention in improving energy efficiency, saving pollution and reducing emissions, and achieving pollution control (Muganyi et al, 2021), but few literatures have directly examined the role of digital financial development on national haze governance and the impact mechanisms in this process. In addition, in the face of increasingly serious environmental problems, the central government has begun to include environmental issues in the performance assessment of local officials, changing the conventional "GDP-only" development model and increasing the emphasis that local governments attach to environmental issues such as haze governance. As a result, it is worth studying that whether the difference in government emphasis will affect the digital finance on haze governance.
In view of this, this research work focuses on whether digital finance can help China promote the governance for city haze pollution, where the influence of digital finance on the governance is investigated through direct effect, indirect mechanism, steric effect, and so forth. This investigation may enrich the existed research as follows: (1) from the perspectives such as reducing repetitive research and development and increasing the agglomeration of service industries, the influence mechanism of digital finance on urban haze governance is examined, and the heterogeneity of different levels of government emphasis and cities is further considered and tested; (2) considering that the haze pollution is spillover and unbounded and based on the empirical test of the spatial mechanism of digital finance on haze governance, the spatial regional boundary of digital finance on urban haze governance is further investigated, which improves the credibility of the research results, and also enriches the empirical methods in this field; (3) using NASA satellite monitoring data, the annual mean of PM2.5 concentration in each sample city is calculated, which more accurately estimates the real situation of urban haze pollution.
At present, digital finance in China has shown vigorous development and has played a crucial role in promoting urban haze governance. This section firstly analyses the direct impact of digital finance on urban haze governance and examines the heterogeneous effects of different degrees of government emphasis and different levels of cities. Secondly, the indirect transmission mechanisms of green innovation, collaborative innovation, advanced industrial structures, and productive services agglomeration respectively are examined. Finally, the spatial effects of digital finance on haze governance were explored and corresponding research hypotheses are put forward.
2.1 Direct impact of digital finance on haze governance
With the rapid development of the Internet, digital finance has increasingly integrated with traditional finance by relying on new digital technologies such as artificial intelligence and big data. Digital finance has significantly reduced the costs and pollution that may arise in the development of finance and related industries by taking traditional business online, improving the efficiency of financial services and promoting green consumption (Lu, 2018; Du et al, 2022). With more inclusive and efficient service capabilities, it effectively promoted the country's green development and haze governance. Specifically, digital finance relies on online business models such as the Internet and online banking to help users handle online loans, financial management and payment and other financial businesses. This new financial model improves the efficiency of traditional financial services and breaks time and space constraints, while also reducing the consumption of resources such as labor, transportation, and paper, and greatly reducing the emission of pollutants such as haze. In addition, many new service platforms derived from digital finance, such as "Ant Forest" and second-hand trading platforms, will implicitly cultivate the concept of green consumption and environmental awareness among residents, reduce resource waste and pollutant emissions by encouraging green consumption patterns and reuse of resources, thereby promoting haze governance.
Herein, we propose hypothesis 1a: digital finance development can promote urban haze governance.
The Chinese government attaches great importance to the control of haze pollution, which is crucial to public safety and sustainable development, and regards it as a top priority of the current environmental management work. Due to the obvious negative externalities of haze pollution and other problems, the central government has set a clear target for the annual average concentration of PM2.5, and has taken the annual assessment results of haze control as an important basis for evaluating local government officials (Liu et al, 2021; Zhu, 2014; Zhu and Zhang, 2019). In this context, local officials will pay more attention to the prevention and governance of haze at source, actively improve the design and planning of related policies and gradually improve the green standard system. Furthermore, the advantages of digital finance in data analysis, information matching and other fields will be exerted, and the flow of funds will be guided into environmental protection and green industries, which ultimately have a positive impact on the governance of haze.
Herein, we propose hypothesis 1b: in cities with a high level of government emphasis, digital financial development can better promote haze governance.
In addition, digital finance has a path dependence on the traditional financial market, and its development is closely related to the level of urban economic. Considering that the city level is another dimension to measure the economic development of cities, large and medium-sized cities generally have a higher level of economic development and obvious advantages in many fields such as politics, economy, and culture. Such cities are often given priority in development, and their accompanying policy effects may influence the impact of digital finance on haze control.
Herein, we propose hypothesis 1c: digital finance development can better promote haze governance in cities with higher level.
2.2 Indirect impact of digital finance on haze control
2.2.1 Technological innovation
Generally, green innovation is a research and development(R&D) process accompanied by a long cycle, high investment, and high risk. The scientific research budget of enterprises is often difficult to meet the huge funds needed for green innovation in the initial R&D, achievement transformation and later market promotion. Although governments have introduced corresponding policies to support and subsidize green innovation activities, which have improved the enthusiasm of enterprises to participate in green innovation to a certain extent, but the key challenge of financing is still unresolved. As one of the typical representatives of the digital economy era, digital finance is a new financial model in which new digital technology is deeply integrated with the traditional financial industry. Firstly, digital finance can break the time and space limitations of traditional finance, and provide reliable financial support for relevant subjects participating in green innovation activities by integrating idle funds and widening financing channels. At the same time, this process also saves a lot of economic and time costs for lenders, indirectly reducing the cost of green innovation and improving the profitability (Khin and Ho, 2018). Secondly, digital finance can alleviate information asymmetry. Relying on the advantages of new digital technologies in data analysis, information matching and other fields, digital finance can analyze a large amount of behavioral data of related subjects (mainly enterprises) with green innovation over a period of time, leading to more accurate identification and evaluation of their credit and qualifications (Beck et al, 2018), and avoidance of potential risks, which to a certain extent increases the likelihood of success of green innovation activities. Finally, the unique inclusiveness of digital finance can effectively compensate for the disadvantages of traditional financial institutions in neglecting the "tail" of micro and small enterprises and disadvantaged groups, providing micro and small enterprises and relatively less developed regions with the possibility to participate in green innovation activities by lowering the service threshold, etc.
Green innovation is an important way to alleviate the environmental pollution. As relevant innovations are widely used in production and life, they will effectively improve the cleanliness and energy efficiency of enterprises, and promote the generation of solar energy, photovoltaic and other renewable energies, reduce pollutant emissions in the process of production, R&D, and residential consumption, aiming at achieving regional haze governance at source. In addition, green innovation can also reduce the level of urban pollution at the end of environmental governance and promote work of haze governance by reducing haze management costs and improving haze management efficiency.
Herein, we propose hypothesis 2a: digital finance can indirectly promote urban haze governance by enhancing green innovation capability.
Under the influence of digital finance, innovation activities will show more obvious characteristics of openness and diversity. With its unique advantages of breaking through time and space constraints, digital finance will provide real possibilities for joint research and development and collaborative innovation between different regions (cities). Thanks to the instant interaction in the digital economy era, to effectively fill the gaps in regional innovation, regions with a relatively weak innovation base can more conveniently and timely absorb the latest research achievements of developed regions in basic research and other aspects in the process of participating in cooperative innovation (Gomber et al, 2018);while regions with relatively strong innovation capacity can accelerate the marketisation of R&D results through the spillover effect of knowledge. On the whole, beyond accelerating the diffusion of new knowledge and strengthening exchanges and cooperation between regions, digital finance has also significantly reduced the cost of repeated R&D, stimulated the enthusiasm of different regions to participate in joint R&D and collaborative innovation, and ultimately improved the overall innovation capacity of the country (Teece, 2018). In addition, when choosing partners in different regions, it is inevitable to examine the professional capabilities of the target party and its self-adaptability. The rapid development of digital finance can reduce the cost of information collection, speed up the transmission of information, and identify potential risks in the process of cooperative research and development, providing an important technological guarantee for more extensive and in-depth collaborative innovation activities between different cities (Nambisan et al, 2017).
The effectiveness of technological innovation in alleviating environmental pollution has been well documented in existing studies. Compared to the "fighting alone" innovation situation in various regions, trans-regional joint R&D and collaborative innovation will avoid duplication of R&D, save innovation costs, and reduce the risk of failure. Meanwhile, it will also improve the joint control mechanism of trans-regional haze governance by taking advantages of the regional complementarity of economic development, and develop joint R&D in the cutting-edge energy-saving and pollution governance technologies. The joint research and development of related core technologies will ultimately promote the regional haze governance.
Herein, we propose hypothesis 2b: digital financial development can indirectly promote urban haze governance by increasing the intensity of cooperation and innovation between cities.
2.2.2 Industrial structure
From the perspective of industrial structure upgrading, firstly, digital finance can provide many opportunities for enterprises to carry out innovative and entrepreneurial activities by virtue of reducing financing costs and widening financing channels (Shane and Venkataraman, 2000),which in turn will give rise to a number of high-quality new enterprises and increase the added value of related products; Secondly, relying on digital technologies such as artificial intelligence and big data, digital finance will accelerate the creation of new products and new markets, replace traditional industries and backward enterprises, and gradually form industrial clusters with high-tech as the core. In general, most of the above processes will take place in advanced manufacturing and service industries, and will continue to penetrate into other traditional industries. With the increasing use of digital finance in various industries, it will greatly promote the upgrading of the industrial structure from low level to high level (Levine, 2002), that is, the proportion of the tertiary industry will continue to increase, and eventually the Chinese industrial structure will upgrade. With the continuous advancement of China's industrial structure upgrading process, it will significantly reduce the reliance on highly polluting traditional energy sources, and promote urban haze governance by increasing the use of low-carbon energy and green energy, while improving quality and efficiency and optimizing the energy structure.
Herein, we propose hypothesis 3a: digital finance development can indirectly contribute to urban haze control by promoting the industrial structure upgrading.
From the perspective of producer services agglomeration, the vigorous development of high-tech industries and new industries is one of the most important directions to improve the quality of China's economic development. As an important part of the service industry, producer services need the support of various production factors (technology, talent, and capital, etc.). The rapid development of digital finance speeds up the transregional flow of above factors, while strengthening the horizontal and vertical linkages among related industries. Finally, it will promote the agglomeration of productive service industries (Bruhn and Love, 2014). With the improvement of the agglomeration of producer services, relevant enterprises concentrated in a certain space may cooperate to build complete waste treatment equipment in order to save production costs, avoiding duplication of investment while also achieving energy saving and emission reduction. In addition, industrial agglomeration can better attract related talents to work, and gradually form a high-level and professional research team, which is conducive to promoting integration and complementary advantages among various sectors of the productive services industry (Xu and Lin, 2018), providing a constant impetus to promote the transformation and upgrading of the manufacturing industry, achieving efficient production, energy saving and haze governance.
Herein, we propose hypothesis 3b: digital finance development can indirectly achieve urban haze governance by increasing the agglomeration of productive services.
2.2 The spatial spillover effect of digital finance on haze governance
With the help of digital technology, the greatest advantage of digital finance is that it can break the time and space limitations of traditional economic activities. While speeding up information collection and transmission, it also strengthens the intensity and depth of the spatial connection of financial and economic activities between regions. In the context of China's vigorous development of a green economy and the increased emphasis on pollution governance, we should fully Seize the opportunities brought by digital finance to deal with the new challenges that may be faced in the future national competition. Existing studies have found that digital finance not only has a significant spatial correlation, but can also influence carbon emissions, green development, economic growth, and economic quality in neighboring regions through spillover effects (Wu et al, 2022; Li et al, 2022; Liu et al, 2022). As an important part of the regional economic quality, environmental protection and haze governance will inevitably show stronger spatial correlation under the influence of digital finance.
Herein, we propose hypothesis 4a: while digital finance promotes the urban haze governance, it will also affect the haze governance of neighboring cities through spatial spillover effects.
In addition, although digital finance can promote haze governance of surrounding cities through spatial spillover effects, there are certain spatial boundaries for the spillover effects, and with the increase of spatial distance, the spillover effects brought by digital finance will gradually decline (Helsley and Strange, 1990). Firstly, although digital finance can largely break time and space constraints, financial institutions are still more likely to engage in high-frequency exchanges with local financial service demanders, but the impact on surrounding areas (especially remote areas) may be relatively limited. Secondly, local government protectionism will be another important reason for limiting spatial spillover borders in reality. Although China's market-oriented reforms have achieved significant results, market segmentation and trade barriers between different administrative regions still objectively exist, and local protectionism will lead to a significant attenuation of the cross-regional (especially beyond provincial borders) effects of digital finance.
Herein, we propose hypothesis 4b: with the increase of spatial distance, the spillover effect of digital finance on haze control in neighboring regions will show a gradual decay, and there is a certain spatial boundary.
This research selects panel data of 284 prefecture-level cities in China over the period of 2011-2020 as the research sample (Tibet, Hong Kong, Macao, Taiwan, and some cities with serious data loss are not considered). The data are collected from the “City Statistical Yearbook in China”, “Work Reports and statistical yearbooks of government”, Web of Science website, and NASA satellite monitoring data.
3.1 Explained variable
To comprehensively examine the effect of the digital finance on haze governance, this paper selects the PM2.5 concentration as the indicator of haze pollution. Considering that China has only monitored PM2.5 in key cities and regions since 2012, and detected PM2.5 in all cities nationwide in 2015, the lack of data is relatively serious. Therefore, the annual mean PM2.5 concentration in 284 cities during the sample survey period is comprehensively calculated by using the relevant data from satellite and ground monitoring stations (Wang et al, 2019).
3.2 Explanatory variable
At present, the measurement methods for the development of digital finance are relatively limited. The "China Digital Inclusive Finance Index" issued by the Digital Finance Research Center of Peking University is used as the proxy indicator to measure the development of digital finance in Chinese cities (Guo et al, 2016).
3.3 Mediator variable
3.3.1 Green innovation
The number of green patent applications was chosen to measure the green innovation capacity of cities (Du et al, 2021). The main reasons for choosing the number of patent applications instead of the number of grants are as follows: Firstly, the number of patent applications is more stable and timely, which can more accurately reflect the green innovation capacity of cities in the current year; Secondly, the development of digital finance has provided more possibilities for enterprises to carry out entrepreneurial and innovative activities, which has increased the enthusiasm of various innovative subjects such as industry, academia and research institutions to participate in innovation. Therefore, the number of patent applications can better reflect the vitality of urban innovation, which is more suitable for the research scenario in this research.
3.3.2 Collaborative innovation
The number of papers cooperatively published among diffident cities is used to measure the city's cooperative innovation capability. Compared with joint patents, cooperative publishing of papers tends to show stronger cooperation and complexity. Therefore, this research uses Python software to collate the information of authors in each city during the research period from the Web of Science website, and count the number of papers cooperatively published after respectively excluding the samples of independent authors or authors in the same city.
3.3.3 Industrial structure upgrading
With the continuous development of economy, the proportion of industry and service industry will gradually increase, which generally shows the process of evolution from the primary industry to the secondary industry and finally to the tertiary industry. Therefore, some scholars use the ratio between tertiary industry and secondary industry to measure the upgrading of the industrial structure, but this measurement ignores the trend of the primary industry. In order to remedy this deficiency, this research measures the industrial structure upgrading by calculating the ratio between tertiary industry and the sum of primary industry and secondary industry.
3.3.4 Agglomeration of productive services
Generally, the more employees employed in a particular industry in the region, the higher the concentration of enterprises in that industry in the region. Therefore, this research selects the relative intensity of the number of industry practitioners to reflect the degree of concentration. According to equation (1), we obtain the agglomeration of productive service industries in each city, where Eis is the number of employees in industry (s) of city (i), and Ei represents the number of employees in all producer services in city (Liu et al, 2022; Fan, 2003).
3.4 Control variable
In order to accurately calculate the impact digital finance on haze governance, the following seven control variables are set. (1) Government intervention (Gov), measured by the proportion of fiscal expenditure in GDP. (2) Human capital (Hum), measured by the number of students enrolled in colleges and universities; (3) Financial development scale (Fin), calculated as the ratio between the year-end RMB loans balance of financial institutions and regional GDP; (4) The level of opening (Open), obtained by calculating the ratio of the city's total imports and exports to GDP; (5) Traffic intensity (Tra), measured here by the urban highway passenger volume; (6)The level of economic development (Pgdp), is usually assessed the per capita GDP of the city. The results of the descriptive statistics of each variable are shown in Table 1.
Table 1 Descriptive statistics of variables
|
Variable type |
Symbol |
Average |
Standard deviation |
Minimum value |
Maximum value |
|
Explained variable |
PM2.5 |
42.250 |
15.112 |
11.614 |
108.526 |
|
Explanatory variable |
Diff |
1.747 |
0.682 |
0.170 |
3.345 |
|
|
Bre |
1.654 |
0.672 |
0.019 |
3.265 |
|
|
Dep |
1.718 |
0.699 |
0.043 |
3.497 |
|
|
Dig |
2.113 |
0.829 |
0.027 |
5.812 |
|
Mediator variable |
Ginn |
0.007 |
0.017 |
0.000 |
0.271 |
|
|
Cinn |
0.049 |
0.191 |
0.000 |
4.780 |
|
|
Str |
0.787 |
0.423 |
0.113 |
5.200 |
|
|
Agg |
1.029 |
0.273 |
0.261 |
1.860 |
|
Control variable |
Gov |
0.204 |
0.103 |
0.040 |
0.920 |
|
|
Hum |
10.531 |
1.456 |
1.100 |
14.080 |
|
|
Fin |
1.021 |
0.625 |
0.120 |
9.620 |
|
|
Open |
0.176 |
0.285 |
0.000 |
2.491 |
|
|
Tra |
8.286 |
1.084 |
2.197 |
12.566 |
|
|
Pgdp |
10.721 |
0.574 |
8.773 |
13.056 |
3.5 Regression design
3.5.1 Panel regression model
3.5.2 Mediation effect model
In order to further explore the indirect transmission mechanism in the process of digital finance promoting urban haze governance, the mediating effect model shown in equation (3) is set. Green innovation(Ginn), cooperative innovation(Cinn), industrial structure upgrading(Str) and agglomeration of productive services (Agg) are the four mediating variables set in this research, and the selection of control variables is consistent with the baseline model. At present, a large number of literatures has adopted the "three-step approach" in the field of psychology to test the mediating model. However, due to the differences in discipline ideas between economics and psychology, it is easy to lead to the "misuse" of this method. Therefore, this paper uses empirical tests to examine the impact of digital finance on intermediary variables, and the relevant impact of intermediary variables on China's urban haze governance is explained through corresponding economic theories and objective evidence in reality (Kenny et al, 1998; Hayes, 2017; Jiang, 2022).
3.5.3 Spatial metrological model
Generally, there is a degree of spatial correlation among different regions (cities), and the use of traditional panel models may lead to research errors. Based on the measurement of spatial correlation, this paper constructs the spatial Durbin model as shown in Formula (4) to test the possible spatial spillover effects of digital finance when it affects haze governance (the test results of model selection are shown later). Among them, W includes the adjacent matrix and inverse distance weight matrix, 
indicates the impact of digital finance on the haze governance of neighboring cities, and the setting of other variables is consistent with the baseline model. On this basis, the spatial spillover boundary of digital finance on haze governance is further measured by adjusting the settings of the spatial matrix.
4.1 Direct impact of digital finance on urban haze governance
4.1.1 baseline regression
According to the benchmark model set previously, this section uses ordinary least squares (OLS) to explore the direct impact of the total digital finance index and the three dimensional sub-indices (breadth, depth and digitalization of digital finance) on urban haze governance. Firstly, the results of model (2) show that the coefficient of the total digital finance index on PM2.5 concentration is -7.471 at the 1% significance level, indicating that the development of digital finance can significantly inhibit urban haze and improve air quality, effectively testing hypothesis 1a. From the regression results of model (3) and model (5), the breadth and digitalization of digital finance items are -7.615 and -1.843 respectively and pass the significance test, indicating that every 1% increase in the breadth and digitalization, the average confidents of PM2.5 decreased by 7.615 and 1.843 μg/m3, but the depth of digital finance has not generated a significant impact. Specifically, the improvement of the degree and digitalization can make more convenient use of digital technologies such as the Internet and artificial intelligence, which reduces transaction costs and consumption of resources such as labor, transportation, paper, and other resources by breaking the boundaries of time and space. The continuous improvement of digital financial breadth has provided more inclusive financial services and more diversified financing channels for the "tail" groups, effectively expanding the breadth of financial services. As more groups and regions enjoy the convenience of digital finance, it improves the efficiency and reduces financing costs, while also avoiding the waste of resources and pollutant emissions, thus curbing urban haze significantly.
Table 2 The result of baseline regression
|
Variable |
model (1) |
model (2) |
model (3) |
model (4) |
model (5) |
|
Diff |
-8.543*** (-5.68) |
-7.471*** (-4.66) |
|
|
|
|
Bre |
|
|
-7.615*** (-4.57) |
|
|
|
Dep |
|
|
|
0.276 (0.27) |
|
|
Dig |
|
|
|
|
-1.843*** (-6.21) |
|
Gov |
|
-2.220 (-0.77) |
-1.406 (-0.49) |
0.245 (0.09) |
-0.413 (-0.15) |
|
Hum |
|
-0.249 (-0.70) |
-0.132 (-0.37) |
-0.122 (-0.34) |
-0.244 (-0.68) |
|
FIN |
|
-0.247 (-0.87) |
-0.199 (-0.70) |
-0.263 (-0.92) |
-0.289* (-1.02) |
|
Open |
|
-2.263** (-2.48) |
-1.919** (-2.12) |
-1.702* (-1.86) |
-2.041** (-2.26) |
|
Tra |
|
0.363* (1.85) |
0.374* (1.91) |
0.394** (2.00) |
0.436** (2.23) |
|
Pgdp |
|
-2.725*** (-4.20) |
-2.488*** (-3.77) |
-3.230*** (-5.02) |
-3.116*** (-4.89) |
|
Cons |
70.468*** (35.19) |
102.802*** (11.23) |
98.967*** (10.87) |
97.809*** (10.60) |
99.468*** (10.97) |
|
City |
YES |
YES |
YES |
YES |
YES |
|
Year |
YES |
YES |
YES |
YES |
YES |
|
N |
2840 |
2840 |
2840 |
2840 |
2840 |
|
Adj-R2 |
0.914 |
0.925 |
0.925 |
0.924 |
0.925 |
Notes: t statistics in (), ***, **and * are significance levels at 1%, 5% and 10%.
4.1.2 Endogeneity test
In order to avoid errors in the research results caused by endogenous problems, the historical data of the number of fixed phones in 1984 was selected as the instrumental variable. Firstly, considering that the emergence and development of digital finance rely on new digital technologies such as the Internet and big data, the development and transformation of modern digital technologies cannot be separated from the basic conditions created by the original Internet technology; Secondly, with the advancement of new digital communication and 5G technology, the influence of traditional communication methods such as landline telephones on current economic and innovation activities has gradually diminished, meeting the requirement of exclusivity as an instrumental variable. Thirdly, considering that sectional data of a single year cannot be directly put into the panel model for regression, the number of Internet users in the previous year is used to construct an interaction term with the historical data of landline telephones in 1984 as the instrumental variable by referring to the method of Nunn and Qian (2014) (Nunn and Qian, 2014). According to the results in Table 3, the digital finance still reveals a significant inhibitory effect on urban haze concentration after considering the potential endogeneity, which is consistent with the fitting results of baseline regression. In addition, the Wald F-statistic is 16.702 (greater than 10), which rejects the original hypothesis of a weak instrumental variable, indicating that the instrumental variable set in this research is reasonably valid.
Table 3 The result of endogeneity test
|
Variable |
The first stage |
The second stage |
|
Diff |
PM2.5 |
|
|
Diff |
- |
39.766** (1.97) |
|
Tele |
-0.019*** (-4.01) |
- |
|
Control variable |
YES |
YES |
|
City |
YES |
YES |
|
Year |
YES |
YES |
|
Cragg-Donald Wald F statistics |
16.702*** |
- |
Notes: t statistics in (), ***, **and * are significance levels at 1%, 5% and 10%.
4.1.3 Heterogeneity analysis
The previous findings suggest that digital finance can suppress haze and advance urban haze governance generally. Considering that China is a vast country and there are significant differences in the degree of emphasis on haze governance in different regions. So, it is wondered whether the inhibition of digital finance on urban haze concentration will exhibit some differences depending on the emphasis paid by local governments to this issue. In order to test this conjecture, this section focuses on the generation, prevention, treatment of haze, and captures the keywords in the work reports of each government and counts the word frequency1. If the word frequency is higher, it indicates that the local government pays more attention to haze governance, where the cities are divided into higher, middle, and lower groups according to the level of governmental emphasis. The results in Table 4 show that cities with every 1% increase in digital finance where the government pays more attention to haze governance will drive down the haze concentration by 10.978μg/m3, showing a stronger inhibitory effect compared to the baseline regression. In cities with middle emphasis, the development of digital finance can also reduce the concentration of haze, but the inhibition effect is lower than that of cities with high emphasis. In cities with lower emphasis, digital finance has not yet had a significant impact on urban haze concentration. The possible reason for this result is that if we want to achieve regional environmental governance and ecological optimization, we cannot do without emerging products in the Internet era such as digital finance, as well as the emphasis of local government on this issue. Therefore, digital finance will have a stronger inhibitory effect on local haze concentration in regions where the government pays more attention to haze governance, which effectively verifies the hypothesis 1b.
Secondly, the research samples are divided into 35 large and medium-sized cities and other cities according to the city level, and the regression results in Table 4 were obtained. The inhibiting effect of the digital finance on haze concentration is evidently heterogeneous, with the coefficient in large and medium-sized cities being higher than that in the other cities, illustrating that digital finance lays a stronger inhibition effect over the haze concentration of large and medium-sized cities. The possible reason is that large and medium-sized cities are mostly municipalities under the direct central government or provincial capitals, with obvious advantages in many fields such as politics, economy, and culture. Because of more attention to high-quality and sustainable development in that large and medium-sized cities, digital finance can bring about a stronger pollution control effects, verifying the hypothesis 1c.
Table 4 The result of heterogeneity analysis
|
Variable |
Level of government emphasis |
Level of city |
|||
|
higher |
middle |
lower |
35 large and medium-sized cities |
other cities |
|
|
Diff |
-10.978*** (-3.27) |
-4.140* (-1.84) |
-2.600 (-0.95) |
-11.795*** (-2.01) |
-6.391*** (-4.60) |
|
Control variable |
YES |
YES |
YES |
YES |
YES |
|
City |
YES |
YES |
YES |
YES |
YES |
|
Year |
YES |
YES |
YES |
YES |
YES |
|
N |
950 |
950 |
940 |
350 |
2490 |
|
Adj-R2 |
0.927 |
0.934 |
0.913 |
0.905 |
0.928 |
Notes: t statistics in (), ***, **and * are significance levels at 1%, 5% and 10%.
4.2 Indirect effects of digital finance on urban haze governance
As mentioned in the previous section, digital finance can not only directly promote urban haze governance, but also indirectly reduce urban haze concentration by influencing technological innovation (green innovation, cooperative innovation) and industrial structure (industrial structure upgrading, agglomeration of productive services). In order to verify the above transmission path, this section constructs the mediation effect model to test the indirect mechanism of digital finance in promoting urban haze governance.
The contents in Table 5 respectively show the regression results of the above four indirect action paths. Firstly, the coefficient of Diff item in model (6) is 0.248 and significant at the level of 1%, indicating that the development of digital finance can enhance the green innovation capacity of cities. With the latest achievements in green innovation (such as solar energy, photovoltaic and other renewable energy) being more widely used in all aspects of production and life, effectively achieving regional haze governance from the source. The regression coefficient of digital finance on the intensity of cooperation and innovation between cities in model (7) is significantly positive, which fully demonstrates that digital finance provides the possibility of Collaborative innovation by its unique advantage of breaking time and space constraints. The improvement of the intensity of cooperative innovation can also indirectly advance urban haze governance through avoiding duplication of R&D, joint R&D of core technologies, and establishing the joint haze governance organization, which Separately verifies the hypothesis 2a and hypothesis 2b.
Secondly, it can be seen from the models (8) - (9) that every increase of each unit in the development level of digital finance will promote the upgrading of the industrial structure and the agglomeration of producer services by 0.121 and 0.155 units respectively. With the advancement of China's industrial structure upgrading and the increase of the agglomeration of productive services, it will significantly reduce the reliance on traditional energy sources such as coal and oil. At the same time, a high-level research team will be gradually formed to provide continuous impetus for promoting the transformation and upgrading of traditional manufacturing industries and achieving urban haze governance, respectively verifying hypotheses 3a and 3b.
Table 5 The result of internal mechanism
|
Variable |
Model (6) |
Model (7) |
Model (8) |
Model (9) |
|
Ginn |
Cinn |
Str |
Agg |
|
|
Diff |
0.248*** （9.59） |
0.237*** （7.01） |
0.121*** （2.67） |
0.155*** （4.13） |
|
Control variable |
Yes |
Yes |
Yes |
Yes |
|
City |
Yes |
Yes |
Yes |
Yes |
|
Year |
Yes |
Yes |
Yes |
Yes |
|
Obs |
2840 |
2840 |
2840 |
2840 |
|
Adj-R2 |
0.836 |
0.790 |
0.923 |
0.873 |
Notes: t statistics in (), ***, **and * are significance levels at 1%, 5% and 10%.
4.3 Spatial spillover effect of digital finance on urban haze concentration
In the Internet era, the cross-regional flow of knowledge, capital and talent and other development elements is increasingly frequent, and the spatial relevance among different cities is gradually enhanced. In the process of promoting local haze governance, digital finance may also suppress the haze concentration of neighboring cities through spillover effects, so a spatial econometric model is selected to carry out further tests in this section.
4.3.1 Spatial weight matrix and model selection
To ensure the accuracy of the regression results, the adjacency matrix and the inverse distance geographic weight matrix are selected to calculate the spatial Moran index of digital finance and haze concentration. From the measured results in Table 6, it can be found that the Moran values of digital finance and haze concentration passed the significance test during the examination period under the two weight matrices, and the spatial correlation showed an upward trend in general. In addition, the results of the LR test show (the test results are shown in Table 7) that the spatial durbin model (SDM) cannot be degraded to a spatial autoregressive or spatial error model, so the SDM model is selected to empirically study the spatial spillover effect of digital finance on urban haze governance.
Table 6 The result of spatial Moran index
|
Year |
adjacency weight matrix |
inverse distance geographic weight matrix |
||||||
|
digital finance |
haze concentration |
digital finance |
haze concentration |
|||||
|
Moran’s I |
Z |
Moran’s I |
Z |
Moran’s I |
Z |
Moran’s I |
Z |
|
|
2011 |
0.479*** |
11.899 |
0.801*** |
19.871 |
0.316*** |
14.450 |
0.501*** |
22.832 |
|
2012 |
0.481*** |
11.974 |
0.804*** |
19.940 |
0.341*** |
15.619 |
0.497*** |
22.660 |
|
2013 |
0.468*** |
11.646 |
0.798*** |
19.805 |
0.332*** |
15.175 |
0.499*** |
22.784 |
|
2014 |
0.417*** |
10.377 |
0.762*** |
18.940 |
0.279*** |
12.814 |
0.480*** |
21.926 |
|
2015 |
0.461*** |
11.470 |
0.822*** |
20.402 |
0.320*** |
14.665 |
0.534*** |
24.354 |
|
2016 |
0.435*** |
10.833 |
0.810*** |
20.091 |
0.316*** |
14.471 |
0.521*** |
23.766 |
|
2017 |
0.480*** |
11.934 |
0.790*** |
19.594 |
0.347*** |
15.877 |
0.497*** |
22.670 |
|
2018 |
0.546*** |
13.557 |
0.808*** |
20.039 |
0.403*** |
18.395 |
0.515*** |
23.460 |
|
2018 |
0.554*** |
13.758 |
0.820*** |
20.324 |
0.418*** |
19.090 |
0.529*** |
24.127 |
|
2018 |
0.571*** |
14.196 |
0.802*** |
19.889 |
0.436*** |
19.911 |
0.518*** |
23.605 |
Notes: ***, **and * are significance levels at 1%, 5% and 10%.
4.3.1 Regression result analysis
If we want to accomplish the target of national haze governance, the joint efforts and participation of each city are indispensable. Considering that digital finance can promote the regional haze governance, it will inevitably affect the haze concentration of neighboring cities. Therefore, based on geographical adjacency matrix and inverse distance weight matrix, this section constructs SDM models to test the possible spatial spillover effects in the process of digital finance influencing haze governance. In consideration of the existence of spatial effect, the more comprehensive perspective was selected to analyses the regression result by referring to the partial differential method of Lesage (Lesage and Fischer, 2008). From the regression results in Table 7, it shows that every 1% increase in the development of digital finance will advance the reduction of haze concentration in this city by 3.749 and 4.296 μg/m3 under the two weight matrices respectively. At the same time, it will also advance the haze concentration in neighboring cities to decrease by 2.198 and 3.244 μg/m3 through the spatial spillover effect, verifying the hypothesis 4.
Table 7 The result of SDM model
|
Variable |
adjacency weight matrix |
inverse distance geographic weight matrix |
|||||
|
Direct effect |
Indirect effect |
Total effect |
Direct effect |
Indirect effect |
Total effect |
||
|
Diff |
-3.749*** (-5.06) |
-2.198* (-1.84) |
-5.947*** (-5.61) |
-4.296*** (-3.77) |
-3.244* (-1.90) |
-7.540*** (-6.09) |
|
|
Control |
Yes |
Yes |
Yes |
Yes |
Yes |
Yes |
|
|
Obs |
2840 |
2840 |
2840 |
2840 |
2840 |
2840 |
|
|
Likelihood |
-5993.520 |
-6688.589 |
|||||
|
LR-lag |
17.82*** |
121.78*** |
|||||
|
LR-error |
19.81*** |
131.53*** |
|||||
Notes: Z statistics in (), ***, **and * are significance levels at 1%, 5% and 10%.
4.4 The spatial spillover boundary of digital finance on urban haze governance
The results of the previous study suggest that while digital finance can promote local haze governance, it can also suppress haze concentrations in neighboring cities through spatial spillover effects. In fact, as the spatial distance between the two regions (cities) continues to expand, the spatial correlation between them will gradually weaken until it disappears. Therefore, there should be a certain regional boundary for the inhibition of digital finance on the haze concentration of neighboring cities. In order to deeply test how far the space spillover scope of digital finance is in the process of promoting haze governance, this section searches for the boundary of spatial spillover by setting different geographical distance boundary (d). Assuming that the direct distance between city i and city j is dij, and if dij is inside the range of geographical distance boundary (d), the unit will be deleted from the spatial matrix, and the specific matrix setting is shown in Formula (5).
The reason why the spatial units inside the boundary (d) are removed instead of those outside the boundary is that the main purpose of setting the spatial boundary is to investigate whether the spillover effect decreases or even disappears along with the increasing distance between spatial units. Firstly, the initial threshold was set to 100KM, and the spatial threshold was gradually enlarged according to 10KM as a unit; Secondly, empirical regressions were conducted sequentially based on the newly generated matrix, and the spatial spillover effect coefficient and t statistic of each regression of digital finance on urban haze concentration are recorded respectively. The specific calculation results are shown in Table 8. It can be found that, with the increase of distance, the spatial spillover effect of digital finance on haze concentration shows obvious spatial attenuation characteristics, which can be roughly divided into two intervals as follows. The first interval is within 310 KM, in which digital finance can produce significant spatial spillover effect on haze governance, and this distance is roughly the range of a province, indicating that the development of digital finance has a significant spatial spillover effect on the haze governance of other cities in the same province. The second interval is beyond 310KM. When the boundary exceeds 310KM, the spillover effect of digital finance on haze governance is no longer significant, effectively verifying hypothesis 4b.
Table 8 The attenuation boundary of digital financial
|
Space distance |
Spillover effect |
Space distance |
Spillover effect |
Space distance |
Spillover effect |
|
100 |
-26.969* (1.66) |
200 |
-22.177** (-2.37) |
300 |
-14.720** (-2.03) |
|
110 |
-20.692 (-1.46) |
210 |
-23.998** (-2.51) |
310 |
-11.624* (-1.8) |
|
120 |
1.514 (0.89) |
220 |
-22.870** (-2.45) |
320 |
-8.522 (-1.50) |
|
130 |
-18.239 （1.60） |
230 |
-20.619** (-2.31) |
330 |
-7.407 (-1.37) |
|
140 |
-16.737* (1.64) |
240 |
-20.542** (-2.25) |
340 |
-5.530 (-1.02) |
|
150 |
-19.140* (1.86) |
250 |
-21.804** (-2.42) |
350 |
-4.661 (-0.90) |
|
160 |
-21.782** (-2.13) |
260 |
-24.382** (-2.50) |
360 |
-1.826 (-0.40) |
|
170 |
-21.293** (-2.04) |
270 |
-22.197** (-2.45) |
370 |
-3.584 (-0.77) |
|
180 |
-22.144** (-2.14) |
280 |
-20.634** (-2.41) |
380 |
-3.444 (-0.72) |
|
190 |
-21.513** (-2.20) |
290 |
-22.219** (-2.47) |
390 |
-2.373 (-0.51) |
Notes: Z statistics in (), ***, **and * are significance levels at 1%, 5% and 10%.
1 The key words are as follows: PM2.5, PM10, haze, air pollution, environmental protection, emission reduction, dust reduction, particulate matter, air quality, green development, clean energy, pollution control, pollution prevention and control, collaborative development, coordinated governance, collaborative pollution control, new energy.
As a new product of the deep integration of digital technology and traditional finance, digital finance is an important engine for achieving green development and is of great significance in promoting regional haze governance. This research selects urban panel data in China, and firstly uses the fixed effect model to examine the direct impact of digital finance on urban haze governance. It is found that in areas where the government pays more emphasis and the city level is higher, digital finance will have a stronger inhibitory effect on haze concentrations. Secondly, the indirect transmission mechanisms of green innovation, cooperative innovation, industrial structure upgrading and producer services agglomeration in the process of digital finance to promote haze governance are explored. Finally, the SDM model is applied to empirically examine the possible spatial spillover effects of digital finance in promoting haze governance, and it is found that the spillover effects of digital finance have obvious spatial attenuation characteristics.
Based on the above empirical findings, the conclusions of this research have the following implications for policies related to haze governance:
Firstly, the government should strengthen the “flexible” regulation and guidance of digital finance, while paying attention to the heterogeneous characteristics of digital finance driving haze governance. In the regional layout of digital finance, we should unleash the potential of digital finance in the allocation of environmental protection funds, and build a "government-market" dual-track operation mechanism to better play the advantage of digital finance to curb the concentration of haze.
Secondly, compared with the traditional "end of pipe treatment", haze governance should focus more on "source control". Green innovation and collaborative innovation should be vigorously promoted. Policies should be formulated to enhance the coverage, depth, and digitalization of digital finance in the field of green innovation and cooperative innovation, provide diversified financial service models and guide financial resources to flow into green enterprises. At the same time, environmental regulation tools should be used more rationally to guide enterprises to focus on key core technologies, develop energy-saving and emission-reducing technologies, create a new low-carbon and environment-friendly energy system, and ultimately promote urban haze governance.
Thirdly, we will accelerate the construction of digital finance supporting facilities such as base stations for communication and big data centers, and unswervingly promote the stable development of digital finance. Leveraging the catalytic function of digital finance in changing industrial structures, we will promote the transformation and upgrading of traditional manufacturing industries, cultivate productive service industry clusters, and the deep integration of manufacturing and productive services, realizing the low-carbon and clean production.
Fourthly, local governments should strengthen exchanges and interactions, and introduce bilateral or multilateral cooperation agreements. Taking advantage of the unique advantages of digital finance to break the time and space constraints, and relying on the space spillover effect to release its spatial contribution to urban haze governance, exerting the inhibition role of digital finance on haze concentration in a wider space.
Ethics approval and consent to participate
Not applicable
Consent for publication
Not applicable
Author contributions
Conceptualization: [Wang Ruqi]; Methodology: [Wang Ruqi]; Formal analysis and investigation: [Wang Ruqi]; Writing - original draft preparation: [Wang Ruqi]; Writing - review and editing: [Wang Ruqi]; Supervision: [Wang Ruqi i].
Funding
This work was supported by National Social Science Fund (No .18BJY105), Postgraduate Research & Practice Innovation Program of Jiangsu Province (KYCX22_1386).
Declaration of Competing Interest
None.
Data Availability Statement
The data is available from the corresponding author upon reasonable request.
- Beck T, Pamuk H, Ramrattan R, et al. Payment instruments, finance and development[J]. Journal of Development Economics, 2018, 133: 162-186.
- Bondy M, Roth S, Sager L. Crime is in the air: the contemporaneous relationship between air pollution and crime[J]. LSE Research Online Documents on Economics, 2020, 7(3):555-585.
- Bruhn M, Love I. The real impact of improved access to finance: Evidence from Mexico[J]. The Journal of Finance, 2014, 69(3): 1347-1376.
- Cao B, Chen Y, Mcintyre R S. Comprehensive Review of the Current Literature on Impact of Ambient Air Pollution and Sleep Quality[J]. Sleep Medicine, 2021, 79:211-219.
- Cao S, Nie L, Sun H, et al. Digital finance, green technological innovation and energy-environmental performance: Evidence from China's regional economies[J]. Journal of Cleaner Production, 2021, 327: 129458.
- Chen Y, Yang S, Li Q. How Does the Development of Digital Financial Inclusion Affect the Total Factor Productivity of Listed Companies? Evidence from China[J]. Finance Research Letters, 2022: 102956.
- Du G, Yu M, Sun C, et al. Green innovation effect of emission trading policy on pilot areas and neighboring areas: An analysis based on the spatial econometric model[J]. Energy Policy, 2021, 156: 112431.
- Du M, Hou Y, Zhou Q, et al. Going green in China: How does digital finance affect environmental pollution? Mechanism discussion and empirical test[J]. 2022.
- Fan C C, Scott A J. Industrial agglomeration and development: a survey of spatial economic issues in East Asia and a statistical analysis of Chinese regions[J]. Economic geography, 2003, 79(3): 295-319.
- Feng S, Zhang R, Li G. Environmental decentralization, digital finance and green technology innovation[J]. Structural Change and Economic Dynamics, 2022, 61: 70-83.
- Gao Q, Cheng C, Sun G, et al. The impact of digital inclusive finance on agricultural green total factor productivity: Evidence from China[J]. Climate Change and Agricultural System Response, 2022.
- Gomber P, Kauffman R J, Parker C, et al. On the fintech revolution: Interpreting the forces of innovation, disruption, and transformation in financial services[J]. Journal of management information systems, 2018, 35(1): 220-265.
- Gomber, Peter and Koch, Jascha-Alexander and Siering, Michael. Digital Finance and Fintech: Current Research and Future Research Directions[J] . Journal of Business Economics, 2017, 87(5), 537-580.
- Guo F, Kong T, Wang J Y, et al. The index system and compilation of Chinese digital inclusive finance[R]. Working paper of Institute of Digital Finance, Peking University, 2016.
- Guo H, Gu F, Peng Y, et al. Does Digital Inclusive Finance Effectively Promote Agricultural Green Development?—A Case Study of China[J]. International Journal of Environmental Research and Public Health, 2022, 19(12): 6982.
- Hayes A F. Introduction to mediation, moderation, and conditional process analysis: A regression-based approach[M]. Guilford publications, 2017.
- Helsley R W, Strange W C. Matching and agglomeration economies in a system of cities[J]. Regional Science and urban economics, 1990, 20(2): 189-212.
- Jiang T. Mediating effects and moderating effects in causal inference[J]. China Ind. Econ, 2022, 5: 100-120.
- Kenny D, Kashy D, Bolger N. Data analysis[M]//The handbook of social psychology: Vols. 1 and 2,. New York: McGraw-Hill, 1998: 233-265.
- Khin S, Ho T C F. Digital technology, digital capability and organizational performance: A mediating role of digital innovation[J]. International Journal of Innovation Science, 2018.
- Lesage J P, Fischer M M. Spatial Growth Regressions: Model Specification, Estimation and Interpretation [J].Spatial Economic Analysis,2008,3(3): 275-304.
- Levine R. Bank-based or market-based financial systems: which is better?[J]. Journal of financial intermediation, 2002, 11(4): 398-428.
- Li G, Zhang R, Feng S, et al. Digital finance and sustainable development: Evidence from environmental inequality in China[J]. Business Strategy and the Environment, 2022.
- Li J, Wu Y, Xiao J J. The impact of digital finance on household consumption: Evidence from China[J]. Economic Modelling, 2020, 86: 317-326.
- Li J, Wu Y, Xiao J J. The impact of digital finance on household consumption: Evidence from China[J]. Economic Modelling, 2020, 86:317-326.
- Liu T, Hu M, Elahi E, et al. Does digital finance affect the quality of economic growth? Analysis based on Chinese city data[J]. Frontiers in Environmental Science, 2022: 1330.
- Liu T, Hu M, Elahi E, et al. Does digital finance affect the quality of economic growth? Analysis based on Chinese city data[J]. Frontiers in Environmental Science, 2022: 1330.
- Liu X, Zhang X, Sun W. Does the agglomeration of urban producer services promote carbon efficiency of manufacturing industry?[J]. Land Use Policy, 2022, 120: 106264.
- Liu Y, Wu J, Yi H, et al. Under what conditions do governments collaborate? A qualitative comparative analysis of air pollution control in China[J]. Public Management Review, 2021, 23(11): 1664-1682.
- Lu L. Promoting SME finance in the context of the fintech revolution: A case study of the UK's practice and regulation[J]. Banking and Finance Law Review, 2018: 317-343.
- Madrigano J, Baccarelli A, Wright R O , et al. Air pollution, obesity, genes and cellular adhesion molecules[J]. Occupational & Environmental Medicine, 2010, 67(5):312-317.
- Muganyi T, Yan L, Sun H. Green finance, fintech and environmental protection: Evidence from China[J]. Environmental Science and Ecotechnology, 2021, 7: 100107.
- Nambisan S, Lyytinen K, Majchrzak A, et al. Digital Innovation Management: Reinventing innovation management research in a digital world[J]. MIS quarterly, 2017, 41(1).
- Nambisan S, Wright M, Feldman M. The digital transformation of innovation and entrepreneurship: Progress, challenges and key themes[J]. Research Policy, 2019, 48(8): 103773.
- Nunn N, Qian N. US Food Aid and Civil Conflict[J]. American Economic Review, 2014, 104(6):1630-66.
- Rao S, Pan Y, He J, et al. Digital finance and corporate green innovation: quantity or quality?[J]. Environmental Science and Pollution Research, 2022: 1-20.
- Shane S, Venkataraman S. The promise of entrepreneurship as a field of research[J]. Academy of management review, 2000, 25(1): 217-226.
- Sun Y, Tang X. The impact of digital inclusive finance on sustainable economic growth in China[J]. Finance Research Letters, 2022, 50: 103234.
- Teece D J. Profiting from innovation in the digital economy: Enabling technologies, standards, and licensing models in the wireless world[J]. Research policy, 2018, 47(8): 1367-1387.
- Thorsten Beck, Asli Demirguc-Kunt, Maria Soledad Martinez Peria. Reaching out: Access to and use of banking services across countries[J]. Journal of Financial Economics, 2007,85(1):234-266.
- Wan J, Pu Z, Tavera C. The impact of digital finance on pollutants emission: evidence from chinese cities[J]. Post-Print, 2022.
- Wan Y, Li Y, Liu C, et al. Is traffic accident related to air pollution? [J]. Atmospheric Pollution Research, 2020, 11(5) :1028-1033.
- Wang S, Zeng J, Liu X. Examining the multiple impacts of technological progress on CO2 emissions in China: a panel quantile regression approach[J]. Renewable and Sustainable Energy Reviews, 2019, 103: 140-150.
- Wu M, Guo J, Tian H, et al. Can Digital Finance Promote Peak Carbon Dioxide Emissions? Evidence from China[J]. International Journal of Environmental Research and Public Health, 2022, 19(21): 14276.
- Xu B, Lin B. What cause large regional differences in PM2. 5 pollutions in China? Evidence from quantile regression model[J]. Journal of cleaner production, 2018, 174: 447-461.
- Yang B, Ma F, Deng W, et al. Digital inclusive finance and rural household subsistence consumption in China[J]. Economic Analysis and Policy, 2022, 76: 627-642.
- Yao L, Ma X. Has digital finance widened the income gap?[J]. Plos one, 2022, 17(2): e0263915.
- Yin X, Qi L, Zhou J. The impact of heterogeneous environmental regulation on high-quality economic development in China: based on the moderating effect of digital finance[J]. Environmental Science and Pollution Research:1-14.
- Zhong K. Does the digital finance revolution validate the Environmental Kuznets Curve? Empirical findings from China[J]. Plos one, 2022, 17(1): e0257498.
- Zhu X, Zhang Y. Diffusion of marketization innovation with administrative centralization in a multilevel system: Evidence from China[J]. Journal of Public Administration Research and Theory, 2019, 29(1): 133-150.
- Zhu X. Mandate versus championship: Vertical government intervention and diffusion of innovation in public services in authoritarian China[J]. Public Management Review, 2014, 16(1): 117-139.