Has the construction of digital infrastructure improved the efficiency of environmental governance? — Evidence from 235 cities in China

doi:10.21203/rs.3.rs-4479523/v1

Download PDF

Research Article

Has the construction of digital infrastructure improved the efficiency of environmental governance? — Evidence from 235 cities in China

https://doi.org/10.21203/rs.3.rs-4479523/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Since its reform and opening up, China's economy has undergone rapid development, but thishas also led to serious environmental pollution. Environmental governance has become a major important task in economic development. Can the construction of digital infrastructure promote the application of digital technology and improve the efficiency of environmental governance? This article uses panel data from 235 cities in China to empirically verify the significant positive relationship between digital infrastructure construction and environmental governance efficiency. The level of technological innovation and public environmental attention have a mediating effect, while the upgrading of industrial structure has a positive regulatory effect.The positive effect of digital infrastructure on environmental governance efficiency shows heterogeneity, with economically developed cities being smaller than economically underdeveloped cities and environmentally friendly key cities being smaller than non-environmentally friendly key cities. Therefore, to improve the efficiency of environmental governance, it is necessary to increase investment, build scientific and technological innovation infrastructure and an environmental information network public platform, build a modern digital infrastructure system, and promote scientific and technological innovation and industrial upgrading.

Other Public Policy

Digital infrastructure

Environmental governance efficiency

Mediation effect

Regulatory effect

Environmental pollution not only affects economic development but also endangers human health and ecological security. According to WHO statistics, 70% of diseases and 40% of deaths worldwide are related to environmental pollution (Sun Qiupeng, 2019). ^[^]. China's rapid and extensive economic growth in the past 40 years has caused relatively serious environmental pollution. With the deepening of the concept of ecological civilization, the continuous improvement of environmental protection policies and systems, the government's high attention given to environmental issues, and the implementation of a large number of major environmental protection projects, China's environmental governance has achieved certain results. However, due to the extensive economic development in China, environmental pollution sources are extensive, the degree of pollution is deep, the causes of pollution are complex and variable, and the difficulty of governance is high. Therefore, environmental protection still faces significant challenges. Improving the efficiency of environmental governance has long been a major concern for the government, scholars, and society. The factors that affect the efficiency of environmental governance are complex, related not only to changes in people's environmental protection concepts but also to environmental regulatory policies, economic growth models, government fiscal decentralization, technological innovation, urbanization, and industrial structure (Wu Jianzu and Wang Rongjuan, 2019; Wang Fanfan, 2019; Tang Yiqiu, Shi Jie, 2019; Shi Jianjun, 2019). ^[^,^,^,^] The development of digital technology has provided a new technological tool for improving the efficiency of environmental governance, becoming an important factor affecting the efficiency of environmental governance. Strengthening the infrastructure construction of digital technology applications has become the key to improving the efficiency of environmental governance. The construction of digital infrastructure can provide basic conditions for the rapid acquisition, collection, processing, storage, and retrieval of environmental data and improve the government's environmental monitoring capabilities. This approach can also facilitate the rapid dissemination of environmental information, increase public environmental demands, innovate environmental governance channels, promote the interconnection of environmental data and information, achieve scientific and intelligent decision-making in environmental governance, and enhance real-time response ability and governance efficiency (Li Li, Sun Yifeng, 2016).^[^] Some studies also suggest that the impact of digital infrastructure construction on the efficiency of environmental governance exhibits a U-shaped relationship with lag and heterogeneity (Xiao Renqiao, Wang Ran, Qian Li, 2023, Pang Ruizhi et al., 2021).^[^,^] What is the impact of digital infrastructure construction on the efficiency of environmental governance? What is the impact mechanism? What kind of heterogeneity exists? This topic is worth further research.

By using panel data from 235 cities in China from 2011 to 2020, this article plans to employ fixed effect model to empirically analyze the impact of digital infrastructure construction on environmental governance efficiency and the heterogeneity of these impacts, verify the important role of digital infrastructure construction in improving environmental governance efficiency, and provide inspiration for digital infrastructure construction. The potential marginal contributions of this article are as follows: ① systematically verifying the improvement effect of digital infrastructure construction on environmental governance efficiency, enriching the empirical research on the factors affecting environmental governance efficiency. ② Verifying the mediating and moderating effects of digital infrastructure construction on improving environmental governance efficiency, and improve the causal chain of the impact mechanism. ③ From the perspective of public attention, this paper analyses how the construction of digital infrastructure can enhance the public’s attention on the environment, thereby improving the efficiency of environmental governance.

Digital infrastructure refers to a new generation of information infrastructure that can reflect the characteristics of the digital economy, including the 5G, internet, data centers, artificial intelligence, the industrial internet and other fields (National Industrial Information Security Development Research Center, 2020).^[^] Current research mainly uses single indicator measurements and comprehensive indicator system evaluations to measure the construction level of digital infrastructure. Single-indicator measurements typically utilize datasets related to investment flows in information transmission, software, and information technology services and evaluate investment in new infrastructure construction using the perpetual inventory method (Pan Yaru and Gu Hengda, 2022). ^[^{] The} comprehensive indicator system is evaluated mainly by the entropy weight method to comprehensively evaluate the overall level of Digital infrastructure by combining four-dimensional indicators such as long-distance optical cable lines, mobile phone switch capacity, industrial robot installation, and the number of internet access ports, as well as three dimensions such as construction degree, use cost, and coverage (Fan et al., 2022; Tang and Yang, 2023).^[^,^] This article adopts a comprehensive indicator system to evaluate the quality of digital infrastructure.

Environmental governance involves formulating and implementing environmental governance decisions (United Nations Development Programme, 2002). ^[^{] The} efficiency of environmental governance refers to the input-output ratio of resources used in the process of environmental governance. Scholars usually construct an indicator system to evaluate the efficiency of environmental governance by examining the output status indicators of environmental governance, such as the proportion of environmental governance input resources (human, material, and financial) and governance output, as well as funding, personnel, technology investment, industrial "three wastes" governance output, and environmental quality changes from multiple perspectives (Young, 2013; Zhang Wuyi et al., 2013; Zhao Zheng and Song Tao, 2013).^[^,^,^] In this article, the efficiency of environmental governance is defined as the input-output ratio in the process of environmental governance to evaluate whether the output of environmental governance matches the input.

The construction of a digital infrastructure provides favorable basic conditions for promoting 5G, the internet, artificial intelligence, the industrial internet and digital technology applications. The application of 5G communication, the internet, artificial intelligence and digital technology will help the government more quickly and accurately understand the environmental pollution situation in various regions, industries and enterprises and realize real-time and full process supervision of pollution sources, thus improving the quality of supervision. Moreover, rapid information transmission has also occurred between governments at all levels and between governments and enterprises, effectively reducing the occurrence of environmental pollution but untimely governance measures, thereby improving the efficiency of environmental governance. Digitization enables the government to have the ability to intelligently govern the environment, increasing the rigor, reasonableness, and effectiveness of environmental governance policies, further improving the efficiency of environmental governance.

Based on this, hypothesis 1 is proposed: the construction of digital infrastructure will promote the improvement of environmental governance efficiency

The application of digital infrastructure helps information to spread more conveniently across time and space, promotes interconnection and sharing among individuals, and solves the supply-demand contradiction in regional innovation system innovation activities (Han Xianfeng et al., 2019). ^[^] The use of digital infrastructure has promoted the elimination of information flow barriers between regions, enabling underdeveloped regions to better integrate into efficient and interconnected digital networks, access diverse information and cutting-edge technological achievements, and cooperate with developed regions for innovation and common progress (Ma Qian et al., 2022).^[^] The establishment of digital platforms promotes the widespread application of information technology, reduces or breaks down technological barriers, helps broaden the channels for information exchange between regions and enterprises, enhances information transparency (Hendriks, 1999),^[^] accelerates the diffusion of technological innovation, and enhances the innovation level of the entire society. The application of technological innovation in environmental governance can improve the rationality and accuracy of environmental governance decision-making and implementation. For example, green technological innovation can significantly suppress the emission of industrial pollutants (licensing and Zhang Yafeng, 2021).^[^] Based on digital infrastructure and technological innovation, the government is able to screen, process, and analyze massive amounts of data, accurately grasp the occurrence patterns of environmental violations, predict the trends and development trends of environmental events in a timely manner, and improve its ability to handle environmental emergencies (Chen Jian, 2023),^[^] thereby improving the efficiency of environmental governance.

Based on the above viewpoint, Hypothesis 2 is proposed: Digital infrastructure promotes the efficiency of environmental governance by enhancing the level of technological innovation.

The construction and use of digital infrastructures such as the internet and big data make it easier for the general public to learn about environmental information. Once environmental problems occur, especially during certain environmental emergencies, they quickly attract widespread public attention, causing great pressure for social and environmental governance and prompting the government to pay more attention to environmental problems, increase investment in environmental governance, improve the industrial structure, and reduce environmental pollution. Moreover, the internet enables the public to obtain more environmental protection information, better supervise the environmental governance behavior of the government and enterprises, and help improve the efficiency of the environmental governance of the government and enterprises. Previous studies have shown that public environmental attention has a positive impact on corporate environmental behavior and local government environmental governance. For example, Wu Libo et al. (2022) ^[^] suggested that public environmental attention can reduce the degree of information asymmetry between the government and enterprises, suppress collusion between local governments and enterprises, promote the government to increase environmental regulation intensity and investment in environmental infrastructure construction, and ultimately encourage enterprises to strengthen environmental governance investment. Moreover, the market demand for green products and investment in environmental protection enterprises should increase, the expansion of environmental protection enterprises should be promoted, and thus, environmental quality should be improved. Public attention plays an important role in the environmental governance game between the government, enterprises, and the public. Through the establishment of environmental protection organizations, petitions, Weibo, and other public channels, environmental protection information is provided to the government, environmental protection interests and demands are expressed, and pollution disputes are directly resolved. Consumption and investment choices affect the profits of polluting enterprises and their environmental protection behavior.

Based on the above discussion, Hypothesis 3 is proposed: Digital infrastructure can promote the improvement of environmental governance efficiency by increasing public environmental attention.

For a long time, unreasonable industrial structure and extensive production have been important causes of natural environmental pollution and damage. The application and industrial ecology brought about by the "platforms" formed by the construction of digital infrastructure have a multiplier effect on "one industry leading a hundred industries". Network connection technologies such as 5G networks and the Internet of Things can drive the development of the industrial chain, trigger collective and disruptive technological breakthroughs, and induce the emergence of new technologies, products, and formats (Guo Chaoxian et al., 2020),^[^] forming a new green and reasonable industrial structure that can significantly promote industrial structure upgrading (He Yumei and Zhao Xinhao, 2021).^[^] From the perspective of environmental governance, a reasonable industrial structure means that while improving production efficiency, management level, and product value, it can also improve environmental conditions and have a positive impact on environmental governance.

Therefore, hypothesis 4 is proposed: In the process of promoting environmental governance efficiency through digital infrastructure, industrial structure upgrading has a positive regulatory effect.

3.1 Quantitative Model

To study the impact of digital infrastructure on the efficiency of environmental governance, the following model is constructed:

$${Y}_{it}={a}_{0}+{a}_{1}{X}_{it}+\sum _{j=1}^{6}{\theta }_{j}{Z}_{jit}+{\mu }_{i}+{\delta }_{t}+{\epsilon }_{it} （1）$$

where i represents an individual city, t represents year, ${Y}_{it}$is the dependent variable, representing the environmental governance efficiency of city i in year t, ${a}_{0}$ is a constant term, ${X}_{it}$ is the core explanatory variable coefficient, representing the level of digital infrastructure in city i in year t, ${\theta }_{j}$ is the coefficient vector of the control variable Z, ${\mu }_{i} \text{a}\text{n}\text{d} {\delta }_{t}$ represent the fixed effects of cities and years, respectively, and ${\epsilon }_{it}$ is a random interference term.

3.2 Variable Definition

3.2.1 Dependent variable

The efficiency of environmental governance (Eeg) is calculated using the DEA-SBM model with non-radial distance function based on the input-output indicators of environmental governance. The specific input-output indicator system of environmental governance is shown in Table 1. The data on the energy conservation and environmental protection expenditures of local governments in Table 1 are derived from the annual financial budget and final accounts reports on the official website of each city government or financial department, which are adjusted using the price index of fixed assets investment. Because the National Bureau of Statistics no longer released the index after 2019, adjustments were made using the Industrial Producer Purchase Price Index after 2020. Three types of unexpected outputs are represented by the comprehensive score calculated using the entropy method.

Table 1

Environmental Governance Efficiency Evaluation Index System
Indicator Name		Measurement method	unit
Input	Capital investment	Local government expenditure on energy conservation and environmental protection	Ten thousand yuan
Input	Labor input	Number of employees in the water conservancy environment and public facility management industry	ten thousand people
Output	Expected output	Comprehensive utilization rate of general industrial solid waste	%
		Harmless treatment rate of household waste	%
		Centralized treatment rate of sewage treatment plants	%
		Green area	hectare
	Unexpected output	Industrial wastewater discharge	10000 tons
		Industrial sulfur dioxide emissions	ton
		Industrial smoke and dust emissions	ton

3.2.2 Core explanatory variables

The level of digital infrastructure construction (Dinf) in each city is characterized by constructing a comprehensive evaluation index system using principal component analysis and the entropy weight method to calculate the comprehensive score. The specific evaluation indicators are shown in Table 2. Since it is difficult to obtain optical cable line length data and internet broadband access port data at the prefecture-level city level, this paper refers to the practices of Wang Qin et al.

Table 2

Evaluation Index System for Digital Infrastructure
Variable Name	Calculation method	unit
Per capita telecommunications business volume	Total telecommunications business volume/year-end total population	Yuan/person
Information professionals	Information transmission, computer services, and software industry practitioners	ten thousand people
Internet penetration	Number of internet users/total population at the end of the year	Household/Hundred People
Mobile phone penetration rate	Number of mobile phone users/total population at the end of the year	Department/hundred people
Fiber optic cable line density	Fiber optic cable line length/administrative area	Kilometers per square kilometer
Number of internet-broadband access ports	Number of internet broadband access ports	Ten thousand

3.2.3 Mechanism variables

(1) Mediating variables: Technological innovation level (Tech) and public environmental attention (Pea). The number of annual invention patent authorizations in each city was selected to measure the level of technological innovation. Following the approach of Wu Libo et al. (2022),^[^] the overall environmental pollution index of each city's Baidu search was used to measure the public's environmental attention in the region. Because the number of invention patent authorizations in some cities and the total Baidu search environmental pollution index are 0, logarithmic processing is performed by adding 1.

(2) Adjusting variables: Industrial structure upgrading (Insu). Referring to the approach of Zhou Changlin et al. (2007),^[^] the level of regional industrial structure upgrading is calculated. The specific formula is as follows:

$$Insu=\sum _{i=1}^{n}{k}_{i}\sqrt{\frac{{p}_{i}}{{L}_{i}}} ,i=\text{1,2},3\dots n$$

where ${k}_{i}$ is the proportion of the output value of industry i to the total output value, $Pi/Li$ represents the labor productivity of industry i, $\sqrt{\frac{{p}_{i}}{{L}_{i}}}$ is the industry-level coefficient of industry i, and $Insu$ indicates the level of industrial structure upgrading.

3.2.4 Control variables

(1) Economic development level (Edl). Using the per capita GDP of each city to measure the level of regional economic development, this study explores the role of the economic development level in the impact of digital infrastructure on environmental governance efficiency.

(2) Population density (Popd). The population per square kilometer in each region was used as population density.

(3) Openness to the outside world (Open). Employing the actual amount of foreign investment used by each city each year to measure the level of regional openness to the outside world.

(4) Fiscal decentralization (Fisd). Measure local fiscal decentralization by the ratio of revenue to expenditure within the general budget of local government finance.

(5) Government technology expenditure (Gtex). The amount of government financial technology expenditure is used to measure regional government technology expenditure.

(6) Traditional infrastructure (Tinfra). The traditional infrastructure construction level in a region is measured using the length of urban roads and bridge roads.

3.3 Data sources

The data used in this article come from the "China Urban Statistical Yearbook", "China Urban Rural Construction Statistical Yearbook", "China Environmental Statistical Yearbook", "China Science and Technology Statistical Yearbook", and various provincial statistical yearbooks. The GDP data are adjusted according to the GDP reduction index provided by the National Bureau of Statistics, and non-ratio data for the control variables are logarithmically processed. Some missing data are filled in using interpolation. Because the energy conservation and environmental protection expenditure data collected by the author cover only 261 cities, after deleting some cities with significant missing data, this article retained data from 235 cities for empirical analysis.

Descriptive statistical calculations were performed for each variable, and the results are shown in Table 3.

Table 3

Descriptive Statistical Results
Variable	Sample size	mean value	Standard error	minimum value	Maximum value
Eeg	2350	1.019	0.070	0.685	1.855
Dinf	2350	0.601	0.042	0.567	0.967
Tech	2350	5.308	1.767	0	11.055
Pea	2350	2.886	0.878	0	4.951
Insu	2350	5.382	0.548	3.476	7.171
Edl	2350	10.428	0.647	9.026	12.97
Open	2350	11.78	1.85	2.586	16.479
Fisd	2350	0.477	0.221	0.07	1.541
Popd	2350	5.852	0.875	2.303	7.882
Gtex	2350	10.303	1.339	6.268	15.014
Tinfra	2350	6.562	0.944	4.174	9.558

Table 3 shows that the standard errors of the dependent variable (environmental governance efficiency) and the core explanatory variable (digital infrastructure) are relatively small, at only 0.070 and 0.042, respectively. This means that the sample statistics are very close to the values of the population parameters, and the sample is representative of the population. The average efficiency of environmental governance is 1.019, with a minimum value of 0.685 and a maximum value of 1.855. The minimum value of digital infrastructure is 0.567, the maximum value is 0.967, and the average is 0.601.

4.1 Benchmark regression results

Table 4 reports the basic results of using OLS estimation for Eq. (1). The regression results in columns (1), (2), and (3) of Table 4 show the fixed city year without control variables, the fixed city individual with control variables, and the fixed city year with control variables. The fitting coefficients of digital infrastructure are all significant at the 1% level, and the coefficients are all greater than 0, indicating that digital infrastructure promotes the improvement of environmental governance efficiency. The regression results in column (3) show that every 1% increase in the level of urban digital infrastructure construction will improve the efficiency of environmental governance by 0.297%. Assumption 1 is proven.

Table 4

Benchmark Regression Results
	The efficiency of Environmental governance (Eeg)
	(1)	(2)	(3)
Dinf	0.269***	0.631***	0.297***
	(0.104)	(0.113)	(0.105)
Edl		-0.044***	-0.041***
		(0.015)	(0.015)
Open		-0.006***	-0.005**
		(0.002)	(0.002)
Fisd		-0.031	0.022
		(0.023)	(0.027)
Popd		-0.016	-0.054
		(0.037)	(0.039)
Gtex		0.009**	0.007*
		(0.004)	(0.004)
Tinfra		0.015**	0.004
		(0.006)	(0.008)
Constant term	0.857***	1.096***	1.524***
	(0.062)	(0.281)	(0.298)
Urban fixed effects	yes	yes	yes
Year fixed effects	yes	no	yes
Obs	2350	2350	2350
${R}^{2}$	0.244	0.204	0.251
Note: Numbers in parentheses report robust standard errors. , and **** represent significance levels of 10%, 5%, and 1%, respectively.

4.2 Robustness testing

4.2.1. Endogeneity test for instrumental variable 2SLS estimation

The benchmark regression results show that the construction of digital infrastructure in various cities can promote the improvement of urban environmental governance efficiency. However, considering the potential endogeneity issues between the two, endogeneity testing is necessary. Following the approach of Zhang Xun et al. (2020),^[^] this article uses the spherical distance from each city to Hangzhou, multiplied by the total postal service volume of each city, as an instrumental variable to conduct a two-stage OLS regression. The results are shown in Table 5. The results show that the coefficient of digital infrastructure is still significantly positive, and the KP-LM statistic is significant at the 1% level. The Cragg Donald Wald F statistic is 128.638, which is greater than the critical value of 16.38. This finding passes the weak instrumental variable identification test, which means that after solving the endogeneity problem, digital infrastructure still has a positive impact on environmental governance efficiency.

Table 5

Endogeneity test results
	(1)	(2)
	Digital infrastructure	Eeg
Dinf		1.494***
		(0.416)
Instrumental variable	0.006***
	(0.001)
Control variable	yes	yes
Urban fixed effects	yes	yes
Year fixed effect	yes	yes
Kleibergen Paap rk LM statistical		48.872
Kleibergen Paap rk LM statistical		[0.000]
Cragg Donald Wald F statistical		128.638
Cragg Donald Wald F statistical		{16.38]
Obs		2350
${R}^{2}$		0.024
Note: Numbers in parentheses, square brackets and curly braces represent robust standard errors, the P value, and the critical value at the 10% level of the weak identification test, relatively.

4.2.2. Estimation of Replacement Variables and Robustness Testing of Changing Sample Individuals

To further test the robustness of the basic regression results, alternative measurement methods were used to calculate the explanatory variable and the dependent variable OLS regression was performed on Eq. (1), and the results are shown in columns (1) and (2) of Table 6. The dependent variable was replaced by the regression results calculated using the Shan Haojie (2008) ^[^] capital stock calculation method, and the core explanatory variable was calculated using the entropy method in column (2). Then, the method of changing sample individuals is adopted to test the robustness of the empirical regression results. The third column of Table 6 shows the regression results after removing the sample data from four municipalities directly under the central government. According to Table 6, after the calculation methods for the explanatory variables and core explanatory variables are changed, the regression results, as well as those for the sample individuals, are robust. The fitting regression coefficients of the core explanatory variables are significantly positive, further indicating that digital infrastructure has a positive effect on the efficiency of environmental governance. The benchmark regression results of this article are robust, once again proving the validity of hypothesis 1.

Table 6

Robustness test results
	(1)	(2)	(3)
Dinf	0.187**	0.190***	0.408***
	(0.088)	(0.070)	(0.145)
Edl	-0.027*	-0.040***	-0.042***
	(0.014)	(0.015)	(0.015)
Open	-0.005***	-0.005**	-0.005**
	(0.002)	(0.002)	(0.002)
Fisd	0.035	0.021	0.023
	(0.025)	(0.027)	(0.027)
Popd	-0.040	-0.061	-0.051
	(0.039)	(0.039)	(0.039)
Gtex	0.003	0.007*	0.007*
	(0.004)	(0.004)	(0.004)
Tinfra	0.007	0.004	0.004
	(0.007)	(0.008)	(0.008)
Constant term	1.382***	1.724***	1.452***
	(0.293)	(0.291)	(0.306)
Urban fixed effects	Yes	Yes	Yes
Year fixed effect	Yes	Yes	Yes
Obs	2350	2350	2310
${R}^{2}$	0.248	0.250	0.251
Note: () reports robust standard errors. , and**** represent significance levels of 10%, 5%, and 1%, respectively.

4.3 Heterogeneity analysis

The positive effect of digital infrastructure construction on environmental governance efficiency was previously demonstrated, and the differences in positive effects between the two are further analyzed from the perspectives of economic development and government attention.

Table 7

Heterogeneity analysis results
	The efficiency of environmental governance (Eeg)
	(1)	(2)	(3)	(4)
	Environmental protection key cities	Non environmentally friendly key cities	Economically developed cities	Economically underdeveloped cities
Dinf	0.182**	1.351**	0.175**	1.656***
	(0.086)	(0.561)	(0.086)	(0.607)
Edl	-0.022	-0.059***	-0.037**	-0.037
	(0.017)	(0.023)	(0.017)	(0.023)
Open	-0.005*	-0.004	-0.008**	-0.003
	(0.003)	(0.003)	(0.003)	(0.002)
Fisd	0.054	-0.040	0.019	-0.005
	(0.035)	(0.044)	(0.025)	(0.054)
Popd	-0.070*	-0.027	-0.043	-0.023
	(0.042)	(0.070)	(0.040)	(0.077)
Gtex	0.012	0.007	0.013**	0.008
	(0.008)	(0.005)	(0.006)	(0.005)
Tinfra	-0.000	0.008	0.007	0.004
	(0.012)	(0.010)	(0.009)	(0.011)
Constant term	1.475***	0.927*	1.473***	0.487
	(0.378)	(0.529)	(0.381)	(0.589)
Urban fixed effects	Yes	Yes	Yes	Yes
Year fixed effect	Yes	Yes	Yes	Yes
Obs	1020	1330	1130	1220
${R}^{2}$	0.189	0.293	0.257	0.262
Note: The division of key and non-key cities for environmental protection is based on the "Eleventh Five Year Plan for National Environmental Protection" issued by the State Council in 2007. According to the 2020 Commercial Charm City Ranking, first tier cities, new first tier cities, second tier cities, and third tier cities are classified as economically developed cities, while fourth tier cities and fifth tier cities are classified as economically underdeveloped cities. Robust standard errors are reported within (). , and **** represent significance levels of 10%, 5%, and 1%, respectively.

Table 7 reports the regression estimation results for the impact of digital infrastructure construction on environmental governance efficiency in key environmental protection cities and non-key environmental protection cities, as well as in economically developed and underdeveloped cities. According to the results in columns (1) and (2), for both key environmental protection cities and non-key environmental protection cities, the regression coefficient of digital infrastructure is significantly positive at the 5% level, but the coefficient of key environmental protection cities is smaller than that of non-key environmental protection cities. Similarly, the results in columns (3) and (4) show that the regression coefficients for the fitting of digital infrastructure in economically developed and underdeveloped cities are significantly positive, but the coefficients for economically developed cities are smaller than those for economically underdeveloped cities.

4.4. Mechanistic verification

To further analyze the ways in which the construction of digital infrastructure promotes the improvement of environmental governance efficiency, this article conducts a causal inference test between the core explanatory variable X and the mediating variable M. The test results are shown in columns (1) and (2) of Table 8.

Table 8

Mechanism verification results
	The efficiency of environmental governance (Eeg)
	(1)	(2)	(3)
	Technological innovation level	Public environmental attention	Environmental governance efficiency
Dinf	1.142**	1.316***	0.646***
	(0.556)	(0.427)	(0.233)
Insu			0.020*
			(0.011)
Dinf_*Insu			0.353**
Dinf_*Insu			(0.160)
Edl	0.045	0.103*	-0.037**
	(0.081)	(0.058)	(0.015)
Open	0.040***	0.001	-0.005**
	(0.012)	(0.008)	(0.002)
Fisd	-0.047	-0.065	0.019
	(0.123)	(0.097)	(0.027)
Popd	-0.238	0.658**	-0.042
	(0.314)	(0.263)	(0.040)
Gtex	0.225***	0.048***	0.008**
	(0.024)	(0.016)	(0.004)
Tinfra	-0.111**	0.114***	0.003
	(0.045)	(0.031)	(0.008)
Constant term	3.508	-4.052**	1.595***
	(2.164)	(1.754)	(0.293)
Urban fixed effects	Yes	Yes	Yes
Year fixed effect	Yes	Yes	Yes
Obs	2350	2350	2350
${R}^{2}$	0.966	0.929	0.255

Table 8 shows that the fitting regression coefficients of the intermediate variable, technology innovation level and public environmental attention, as well as digital infrastructure, are significantly positive. This means that the construction of digital infrastructure has improved the level of technology innovation and public environmental attention. As mentioned earlier, the positive externalities of technological innovation can increase enterprise productivity, reduce environmental pollutant emissions, and thus improve environmental governance efficiency. Increasing public environmental attention can effectively supervise the environmental governance work of the government and enterprises, influence the selection of environmental protection strategies by the government and enterprises, and thus improve environmental governance efficiency. Therefore, Hypotheses 2 and 3 are supported.

The regression results of the moderating variable for industrial structure upgrading are reported in column (3) of Table 8. In the regression estimation, to eliminate the possible multicollinearity issue between the core explanatory variable and the moderating variable, the two variables were centralized. The regression results in column (3) of Table 8 show that the coefficient of digital infrastructure is significantly positive at the 1% level, and the coefficient of the interaction term between digital infrastructure and industrial structure upgrading is significantly positive at the 5% level. This indicates that industrial structure upgrading has a positive moderating effect on the impact of digital infrastructure on environmental governance efficiency. Hypothesis 4 is supported.

5.1 Research conclusion

This article uses data from 235 cities in China to analyze the impact of digital infrastructure on the efficiency of environmental governance and further explores its mechanism from three aspects: technological innovation level, public environmental attention, and industrial structure upgrading. Heterogeneity analysis was conducted based on the characteristics of environmental protection importance and economic development level. The research conclusions are as follows:

(1) The construction of digital infrastructure has a positive impact on the efficiency of environmental governance. The construction of digital infrastructure promotes the upgrading of industrial structure by improving the level of technological innovation and public environmental attention, thereby having a positive promoting effect on the efficiency of environmental governance.

(2) The results of heterogeneity analysis show that whether a city is a key environmental protection city or a non-key environmental protection city or an economically developed city or an economically underdeveloped city, the impact of digital infrastructure on environmental governance efficiency is significantly positive. However, the positive effects of non-environmentally friendly key cities and economically underdeveloped cities are greater than those of environmentally friendly key cities and economically developed cities.

5.2 Policy recommendations

To promote the efficiency of environmental governance in China, comprehensively enhance the efficiency of environmental governance, assist in the green transformation of industries, and achieve high-quality economic development, we propose the following:

(1) Increase investment and build a modern digital infrastructure system. The government and the market should increase investment, accelerate the pace of digital infrastructure construction, promote connectivity between urban and rural areas and regions, and eliminate the digital divide between regions and between urban and rural areas. The government and the market should also play full roles, promote the spatially balanced allocation of digital infrastructure, and build a modern digital infrastructure system.

(2) Strengthen the construction of technological innovation infrastructure and promote technological innovation and industrial upgrading. The government should increase support for scientific and technological innovation, increase investment in scientific and technological innovation elements, build innovation infrastructure, and provide a solid foundation for scientific and technological innovation. Digital infrastructure should be fully utilized, the application of digital technology should be expended, the industrial structure should be upgraded and optimized, outdated and highly polluting production capacity should be eliminated, industrial energy consumption should be reduced green transformation and upgrading should be achieved, and environmental pollution should be reduced.

(3) An environmental information network public platform should be established to enhance the efficiency of government environmental governance. To improve the efficiency of environmental governance, it is necessary to ensure that the government obtains timely and accurate environmental information of public concern. Therefore, the government should establish an environmental information network public platform, publish environmental standards, address environmental issues of public concern, understand the current situation of the environment, optimize decision-making, formulate better environmental governance strategies, and establish positive interactions and co governance situations among the government, enterprises, and the public to improve governance efficiency.

Sun Qiupeng. The role of high-quality economic development in promoting environmental protection and ecological civilization construction [J]. Contemporary Economic Management, 2019,41 (11): 9-14.
Wu Jianzu, Wang Rongjuan. Has environmental protection talks improved the efficiency of local government environmental governance—— Empirical Analysis Based on Double Difference Method [J]. Journal of Public Administration, 2019,16 (01): 54-65+171-172.
Wang Fanfan. The Impact of Economic Growth Goals on the Efficiency of Local Environmental Governance: An Empirical Analysis Based on Panel Data of Cities at or above the Prefectural Level in China [J]. Urban Issues, 2022 (09): 76-86.
Tang Yiqiu, Shi Jie. Fiscal decentralization, local government catch-up behavior, and environmental governance efficiency: threshold effect and transmission mechanism analysis based on data from 87 cities [J]. Journal of Guizhou University of Finance and Economics, 2019 (05): 25-34.
Shi Jianjun. Heterogeneous threshold effect of technological innovation on environmental governance efficiency in the central and western regions [J]. Regional Research and Development, 2019,38 (01): 128-131.
Li Li, Sun Yifeng. The Logic, Political Game and Implementation Mechanism of Collaborative Governance of Ecological Environment [J]. Journal of Xiangtan University (Philosophy and Social Sciences Edition), 2016,40 (03): 24-29.
Xiao Renqiao, Wang Ran, Qian Li. The Nonlinear Impact of Digitalization on Corporate Carbon Performance: The Mediating Role of Green Technology Innovation [J]. Science and Technology Progress and Countermeasures, 2023,40 (05): 96-106.
Pang Ruizhi, Zhang Shuai, Wang Qunyong. Can digitalization improve environmental governance performance—— Empirical evidence from inter provincial panel data [J]. Journal of Xi'an Jiaotong University (Social Sciences Edition), 2021, 41 (05): 1-10.
National Industrial Information Security Development Research Center. Decoding Digital Infrastructure, Empowering Digital Economy Development [EB/OL]. 2020, 05-18.
Pan Yaru, Gu Hengda. The Impact of New Infrastructure Investment on the Transformation and Upgrading of the Service Industry [J]. Reform, 2022, (07): 94-105.
Fan L, Zhang Y, Jin M, et al. Does New Digital Infrastructure Promote the Transformation of the Energy Structure? The Perspective of China's Energy Industry Chain [J] Energies, 2022, 15 (23): 8784.
Tang K, Yang G. Does digital infrastructure cut carbon emissions in Chinese cities? [J] Sustainable Production and Consumption, 2023, 35: 431-443.
United Nations Development Programme, United Nations Development Programme, World Bank, and World Resources Research. World Resources Report [M] China Environmental Science Press, 2002.
Zhang Wuyi, Chen Tianming, Wang Tiedan. Research on the Efficiency of Regional Environmental Governance in China Based on DEA and Malmquist Index [J] East China Economic Management, 2013,27 (02): 172-176.
Zhao Zheng, Song Tao. Efficiency and influencing factors of regional environmental governance in China. Nanjing Social Sciences, 2013 (03): 18-25.
Young O. R. On environmental governance: sustainability, efficiency, and equity [M] Paradigm Publishers, 2013.
Han Xianfeng, Song Wenfei, Li Boxin. Can the Internet become a new driving force for improving China's regional innovation efficiency? [J] China Industrial Economy, 2019 (07): 119-136.
Ma Qian, Zhang Hongbing, Liao Meng. Digital infrastructure construction, knowledge flow, and high-quality urban development: empirical evidence of quasi natural experiments and spatial spillovers [J]. Industrial Economy Research, 2022 (06): 114-128.
Hendriks P. Why Share Knowledge? The Influence of ICT on the Motivation for Knowledge Sharing [J] Knowledge and Process Management, 1999, 6 (2): 91-100.
License, Zhang Yafeng. Can green technology innovation bring green mountains and clear waters—Research from the Perspective of Green Patents [J]. China Population, Resources and Environment, 2021, 31 (05): 141-151.
Chen Jian. Path optimization of digital technology empowering modernization of environmental governance [J]. Journal of Harbin Institute of Technology (Social Science Edition), 2023,25 (02): 80-90.
Wu Libo, Yang Meimin, Sun Keqi. The impact of public environmental attention on corporate and government environmental governance [J]. China Population, Resources and Environment, 2022, 32 (02): 1-14.
Guo Chaoxian, Wang Jiaqi, Liu Haorong A Study on the Path of Empowering China's High Quality Economic Development with "New Infrastructure" [J]. Journal of Beijing Institute of Technology (Social Sciences Edition), 2020, 20 (06): 13-21.
He Yumei, Zhao Xinhao. Can New Digital Infrastructure Promote Industrial Structure Upgrading: Empirical Evidence from 272 Prefectural Cities in China [J]. Science and Technology Progress and Countermeasures, 2021,38 (17): 79-86.
Wang Qin, Li Jing, Ding Keke, Lei Li. Digital infrastructure, factor allocation efficiency, and urban-rural income gap [J]. Statistics and Decision Making, 2023,39 (09): 29-34.
Zhou Changlin, Wei Jianliang. Measurement Model and Empirical Analysis of Industrial Structure Level: Taking Shanghai, Shenzhen, and Ningbo as Examples [J]. Shanghai Economic Research, 2007 (06): 15-21.
Zhang Xun, Yang Tong, Wang Chen et al. Development of Digital Finance and Consumer Growth: Theory and Practice in China [J]. Management World, 2020,36 (11): 48-63.
Shan Haojie. Re estimation of China's Capital Stock K: 1952-2006 [J]. Quantitative Economic and Technological Research, 2008,25(10):17-31.

The authors declare no competing interests.

Download PDF

Version 1

posted

You are reading this latest preprint version

Has the construction of digital infrastructure improved the efficiency of environmental governance? — Evidence from 235 cities in China

Status:

Version 1

Abstract

1. Introduction

2. Theoretical analysis and research hypotheses

3. Models and Data Sources

3.1 Quantitative Model

3.2 Variable Definition

3.2.1 Dependent variable

3.2.2 Core explanatory variables

3.2.3 Mechanism variables

3.2.4 Control variables

3.3 Data sources

4. Empirical results and analysis

4.1 Benchmark regression results

4.2 Robustness testing

4.2.1. Endogeneity test for instrumental variable 2SLS estimation

4.2.2. Estimation of Replacement Variables and Robustness Testing of Changing Sample Individuals

4.3 Heterogeneity analysis

4.4. Mechanistic verification

5. Research conclusions and policy recommendations

5.1 Research conclusion

5.2 Policy recommendations

References

Additional Declarations

Status:

Version 1