Svoboda | Graniru | BBC Russia | Golosameriki | Facebook
Next Article in Journal
Rheological, Mechanical, and Micro-Structural Property Assessment of Eco-Friendly Concrete Reinforced with Waste Areca Nut Husk Fiber
Previous Article in Journal
Technical Feasibility for the Boosting of Positive Energy Districts (PEDs) in Existing Mediterranean Districts: A Methodology and Case Study in Alcorcón, Spain
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 15
Issue 19
10.3390/su151914133
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Environmental Regulation and Spatial Spillover Effect of Green Technology Innovation: An Empirical Study on the Spatial Durbin Model

by
Xin Zhang
* and
Feng Xu
School of Economics and Finance, South China University of Technology, Guangzhou 510641, China
*
Author to whom correspondence should be addressed.
Sustainability 2023, 15(19), 14133; https://doi.org/10.3390/su151914133
Submission received: 15 August 2023 / Revised: 21 September 2023 / Accepted: 22 September 2023 / Published: 24 September 2023
(This article belongs to the Section Economic and Business Aspects of Sustainability)
Browse Figures
Versions Notes

Abstract

:
This study empirically examined the spatial spillover effect of various environmental regulations on green technology innovation using panel data from 284 cities at the prefecture level in China between 2007 and 2019. A geographical–economic spatial weight matrix was constructed, and the spatial Durbin model was employed to identify the specific characteristics of this spillover effect. The findings indicate that the spatial spillover effect of green technology innovation primarily occurs through geographical transmission. However, there is no significant spatial autocorrelation when using the economic distance weight matrix. Various types of environmental regulations influence the spatial spillover effect of green technology innovation in distinct ways. Specifically, market-motivated environmental regulation exhibits a U-shaped relationship with the spatial spillover effect, while command-controlled environmental regulation demonstrates an inverted U-shaped relationship, suggesting a complementary effect. Additional research shows that the upgrading of industrial structure acts as a mediator between command-controlled environmental regulation and the spatial spillover effect of green technology innovation. Government departments should comprehensively coordinate market-motivated environmental regulation and command-controlled environmental regulation, accurately assess the intensity of command-controlled measures, and prevent the migration of green technology innovation elements caused by excessive regulatory measures within enterprises.

1. Introduction

Since the establishment of the People’s Republic of China, there was a progressive evolution in public awareness regarding environmental issues. This evolution gave rise to several distinct stages in the development of environmental regulation: the initial stage, the development stage, the turning stage, and the deepening stage. Figure 1 demonstrates a significant increase in the issuance of environmental policies in China during the advanced stage, which began after 2012. Notably, this period witnessed noteworthy improvements in China’s environmental regulation system. Moreover, the consistent trend in the number of green patent applications in China aligns with the introduction of green policies. These observations underscore the importance of further research and discussion to explore the crucial role played by government environmental regulation in fostering the innovation of green technologies. In recent years, the Chinese government placed significant emphasis on green development. In 2020, it introduced the “double carbon” objective, with the aim of reaching the peak of carbon emissions by 2030 and achieving carbon neutrality by 2060. Green technology innovation serves as a vital driving force in attaining this objective [1]. However, it is important to acknowledge that green technology innovation comes with dual externalities. While enterprises investing substantial amounts in research and development (R&D) innovation costs can generate certain benefits through green technology innovation, the spillover of technological knowledge prevents these enterprises from fully capitalizing on these benefits. Consequently, enterprises often lack the motivation to engage in green technology innovation [2]. In light of this, the government should implement appropriate policy interventions to incentivize enterprises to undertake R&D and apply green technology innovation. Therefore, it is of utmost importance for the government to analyze the characteristics of the spatial spillover effect of environmental regulation on green technological innovation, as this will contribute to the advancement of enterprise-driven green innovation.
Numerous studies examined the influence of environmental regulation on green technology innovation, but these studies produced varying results [3,4,5,6,7,8]. Furthermore, there is a lack of research specifically focusing on the spatial spillover effect of green technology innovation. Green technological innovation differs from conventional technological innovation due to its unique external characteristics [9]. Therefore, the findings from existing research on ordinary technological innovation do not directly explain the observed spatial spillover effect in green technological innovation [10]. By incorporating the concept of the spatial spillover effect, a more comprehensive understanding of the impact of environmental regulation can be achieved [11]. This study aims to analyze the spatial spillover effect of environmental regulation on urban green technology innovation from a spatial perspective. It seeks to reveal the relationship between environmental regulation and urban green technology innovation by simultaneously considering regional industrial structure change and industrial transfer. The study utilizes panel data from 284 Chinese cities at the prefecture level, spanning from 2007 to 2019. By constructing a geographical and economic weight matrix, the study investigates the association between environmental regulation and the spatial spillover effect of green technology innovation. Based on cutting-edge literature and classical theories at home and abroad, this paper summarizes relevant studies on environmental planning and green technology innovation. Domestic and foreign studies mainly focus on the static analysis of environmental regulation on green technology innovation, and rarely analyze the dynamic impact of environmental regulation on the change in industrial structure, which leads to the flow of enterprises as the carrier of green technology innovation. The existing research results ignore the impact of spatial factors on the innovative development strategy of the urban green economy, which is gradually receiving attention from some scholars in the field of spatial economics. This study analyzes the spatial spillover effect of environmental regulation on urban green technology innovation from the perspective of the spatial spillover effect revealing the relationship between the two. This study offers several notable contributions. Firstly, it focuses on a different research object compared to prior studies, which often employed provincial panel data. Instead, this study utilizes panel data from prefecture-level cities, enabling a more precise examination of the spillover effects of green technology innovation. Secondly, this study examines the influence of environmental regulation on the spillover effects of green technology innovation from a spatial perspective. By doing so, it establishes a theoretical foundation for local governments to effectively promote green economic development using various approaches. Thirdly, this study delves into the mechanism by exploring the mediating role of industrial structure. It investigates how environmental regulation affects green technology innovation through the intermediary effect of industrial structure.

2. Literature Review and Theoretical Analysis

2.1. Literature Review

The impact of market-motivated and command-controlled environmental regulation on green technology innovation is a prominent topic that garners significant attention from researchers. However, substantial disparities exist among the existing research findings. Recent research presented three distinct perspectives regarding the impact of environmental regulation on green innovation in enterprises. Firstly, it was argued that such regulation can foster green innovation [6,12,13,14]. Secondly, there was a viewpoint suggesting that it may hinder green innovation [3,15,16,17]. Additionally, another perspective suggested a non-linear relationship between the two, characterized by a U-shaped pattern of initial inhibition followed by promotion or an inverted U-shaped pattern of initial promotion followed by inhibition [7,18]. Furthermore, some scholars expressed skepticism and believed that the impact of environmental regulation on green technological innovation in enterprises remains uncertain [19,20,21]. The impact of environmental regulation on green technology innovation is mainly reflected in the following aspects: The first is the crowding out effect, mainly reflected in the command-controlled environmental regulation of enterprise green technological innovation; under the environmental regulation, the enterprise will increase because of the environmental governance investment to achieve environmental regulation of the standard, thus may be out of the enterprise R&D investment, increasing the enterprise the management cost, and this will reduce the enterprise green technology innovation ability [5,8]. The second is the compensation effect, which is mainly reflected in the impact of market-motivated environmental regulation on enterprises’ green technology innovation. If enterprises keep other inputs unchanged, market-motivated environmental regulation will subsidize enterprises’ pollution control behavior in various forms, reducing the cost of enterprises. With an incentive nature, it can promote enterprises to improve technology and thus promote technological innovation of enterprises, which can partially or even completely offset the cost of environmental regulation [22]. The third is the transfer effect, which is mainly reflected in the indirect impact of environmental regulation on enterprises and capital flow caused by regional green technology innovation. Due to the differences of the regional environmental regulation standard impact on the regional industry and capital transfer, polluting industries to reduce environmental costs improves the enterprise’s competitive advantage from strictly an environmental regulation region to a relatively loose environmental regulation and improves green technology innovation ability and transfers them together [23].
Enterprises and capital flow cause indirect influences on regional green technology innovation, environmental regulation has a reverse effect on industrial structure adjustment, and regional industrial structure and environmental regulation have spatial externalities, which affect the layout and adjustment of industrial structure in surrounding regions [24]. Command-controlled environmental regulations can push out those enterprises with serious pollution and low productivity from the market, thus promoting technological innovation in the whole market [25]. At the same time, regions with high market-motivated environmental regulations will also attract enterprises to move to this region [26]. It also leads to the spatial spillover effect of green technology innovation. Research on the spatial spillover of technological innovation primarily investigates its effects by constructing geographical and economic distance matrices. Existing research commonly acknowledges the prominent spillover effect of technological innovation in both geographical and economic dimensions. Nonetheless, studies focusing on the spatial spillover effect of green innovation are relatively scarce, with a lack of corresponding empirical research examinations. Regarding the spatial spillover of general innovation, certain scholars validated the presence of a regional technological innovation spillover effect through geospatial matrix construction [27,28]. The regional technological innovation spillover effect demonstrates a notable trend of distance attenuation [29,30]. Concerning economic spatial spillover effects, China’s innovation resources are mainly concentrated in the economically developed regions of the east, resulting in economic spatial agglomeration phenomena. Relevant studies also revealed the presence of economic spatial autocorrelation in technological innovation. Some scholars further examined the influence of economic spatial distance on technological innovation through the construction of an economic distance matrix. They concluded that technological innovation tends to spill over to regions with closer economic distances, while areas with substantial economic disparities face challenges in generating significant spillover effects due to constraints related to technical talent and industrial foundations [31,32]. There is limited research on the spatial spillover effect of green technology innovation. Existing studies primarily approach green technology innovation from the perspective of regional heterogeneity and suggest that environmental regulation has varying impacts on green technology innovation across different regions [33]. Some scholars analyze this phenomenon through the spatial spillover effect of environmental regulation policies. They argue that increased intensity of environmental policies can stimulate interregional industrial transfer, leading to changes in industrialization processes and the industrial structure of neighboring regions. This promotes economic development and R&D investment in the regions experiencing industrial transfer while enhancing the green innovation capacity of adjacent areas [34].

2.2. Theoretical Analysis

Economic activities in a region are frequently interconnected with its neighboring regions, leading to manifestations of spatial equilibrium, competition, diffusion, and other phenomena. These phenomena partly represent the spatial correlation and interdependence between regions [35]. The spatial equilibrium between regions is influenced by factors such as geographical and natural conditions, scale benefits resulting from agglomeration of enterprises, regional costs of material and information circulation, etc. Spatial competition between regions is primarily characterized by the fact that closer distances correspond to more intense competition among manufacturers [36]. The spatial diffusion effect between regions occurs when the economic center disseminates and shares technical knowledge with neighboring regions [37]. Regional economies exhibit complex structures as they intertwine and interact, forming intricate and diverse networks. These networks generate various effects, including direct impacts, indirect effects, spillover effects, and more [38].
By analyzing spatial diffusion, we can observe the spillover of innovation in terms of geographical space. This allows for the rapid dissemination of new technologies, products, and processes from one region to another, facilitating development across different geographical areas [39]. The primary mechanism of innovation spillover involves generating knowledge spillover through interactions between innovative entities within a region, which includes mutual learning and exchange [40]. Consequently, shorter geographical distances facilitate greater exchange and interaction, thereby promoting innovation spillover in geographical space [41]. The spatial interdependence of economic and social development among neighboring regions increases the likelihood of innovation producing spatial spillover effects [42]. Green technological innovation, as a form of innovation, is similarly subject to widespread spillover effects due to regional exchanges and interactions [43]. Considering externality theory, green technology innovation imposes costs on enterprises without fully capturing all the associated benefits. While enterprise-driven green technology innovation aims to maximize social benefits, it falls short of achieving Pareto optimality due to limited initiative [17]. The large-scale spillover of green technological innovation requires adequate external intervention. Meanwhile, enterprises, as the primary drivers and carriers of green technology innovation, engage in interregional transfers, facilitating spillover. Enterprises, seeking cost reduction, tend to relocate from economically advanced regions to less developed ones [13]. Consequently, the components of green technology innovation accompany enterprise transfers. As a result, spillover of green technological innovation is unlikely to happen between regions with comparable levels of economic development. Based on the aforementioned analysis, we propose that:
Hypothesis 1.
There is a geographical spatial spillover effect of green innovation.
Market-motivated environmental regulations can guide enterprises by providing incentives for investments in environmental protection, technological transformation, and through various means, such as special investments, fiscal subsidies, tax relief, loan incentives, and green procurement, among other preferential policies. These measures aim to offset the costs incurred by enterprises in energy conservation and consumption reduction [44]. In terms of spatial competition, when there is a significant disparity between the intensity of market-motivated environmental regulation incentives and the costs associated with green innovation R&D investments, enterprises may opt to adopt existing technologies, leading to a “siphon effect” that negatively impacts green technology innovation in neighboring areas. If the intensity of environmental regulation incentives is sufficient to offset enterprises’ investments in green innovation R&D, enterprises are encouraged to engage in green technology innovation [45]. This, in turn, enhances their green technology innovation capabilities, resulting in positive spillover effects. Based on the aforementioned analysis, we propose:
Hypothesis 2.
The geographical spillover effect of market-motivated environmental regulation on green innovation exhibits a U-shaped pattern.
Command-controlled environmental regulation enforces rigorous energy conservation and emission reduction measures on enterprises through mandatory measures, such as enhancing environmental assessment in planning, upgrading emission standards, and phasing out outdated production capacity. From a spatial equilibrium standpoint, enterprise costs play a crucial role in achieving spatial balance. If the cost increase resulting from the strength of command-controlled environmental regulation falls within enterprises’ manageable range, they will upgrade their technologies to comply with environmental protection requirements and enhance regional green technology innovation capability. However, if the cost increase surpasses enterprises’ capacity, they may face elimination due to obsolete production capacity, prompting industrial relocation. Consequently, the components of green technology innovation will also relocate along with the enterprises, thereby reducing regional green technology innovation capacity [14]. Based on the aforementioned analysis, we propose:
Hypothesis 3.
The geospatial spillover effect of command-controlled environmental regulation on green innovation exhibits an inverted U-shaped pattern.

2.3. Intermediary Effect of Industrial Structure

As government environmental regulations become increasingly stringent, enterprises are required to adhere to stricter laws and regulations. Consequently, the production costs of enterprises significantly increase, thereby diminishing the competitiveness of local businesses and prompting them to consider options such as cross-regional transfer or industrial transformation. In comparison to industrial transformation, cross-regional transfer offers advantages including shorter completion time, lower costs, and reduced development uncertainty [46]. Hence, when there are changes in the local environmental regulatory framework, an enhancement in the local supervision level prompts highly resource-dependent polluting enterprises to relocate to regions with less stringent regulations. This triggers adjustments in both their own industrial structure and that of neighboring areas [47]. Simultaneously, the elements of green innovation are relocated to these neighboring areas for redistribution. Based on the aforementioned analysis, we propose:
Hypothesis 4.
Command-controlled environmental regulation induces adjustments in regional industrial structure, consequently resulting in the spatial spillover of green innovation.

3. Research Design

3.1. Model Settings

Spatial econometric models encompass various models, such as the spatial autoregressive model (SAR), spatial error model (SEM), and spatial Durbin model (SDM). In spatial econometric model selection, the LM test is commonly utilized for model identification [48]. However, the LM test has certain limitations associated with its characteristics. Simulation analysis reveals that the LM test effectively distinguishes between SAR and SEM models, but its reliability is limited for other tests [49]. Thus, when choosing models, the SDM model can be directly selected for regression analysis [50]. The spatial Durbin model (SDM) combines the characteristics of both the SAR and SEM models while incorporating a spatial lag term for both the explanatory and dependent variables. Introducing these spatial lag variables offers several advantages. Firstly, it helps reduce errors arising from missing variables during the modeling process. Secondly, it enhances the effectiveness of addressing spatial differences [51]. Hence, this study employs the spatial Durbin model, incorporating both spatial and temporal characteristics, for analysis as it addresses the double-fixed nature of the data [51,52].
G T P i t = α + ρ W G T P i t + β 1 E R i t − 1 + β 2 E R i t − 1 2 + β 1 W E R i t − 1 + β 2 W E R i t − 1 2 + δ C O N T R i t + λ i + η t + ε i t
Considering the time lag in patent applications, we incorporated environmental regulation with a one-phase lag (ERit−1) and introduce its quadratic term (ERit−12) to explore the nonlinear association between environmental regulation and innovation in green technology. CONTR represented the set of control variables. Λi denoted the individual fixed effect, ηt represented the time fixed effect, and εit refers to the random disturbance term conforming to εit ~ N(0, σ2). W represented the spatial weight matrix, where W1ij is the geographical distance spatial weight matrix defined as 1/dij. Here, dij represented the distance in terms of longitude and latitude between cities. The economic distance spatial weight matrix, denoted as W2ij, was defined as 1/|Gi – Gj|, where Gi and Gj represented the average per capita GDP of cities I and j during the period from 2007 to 2019.

3.2. Variable Definition

3.2.1. Interpreted Variable

Green technology innovation (GTP): Green technology innovation can be measured in various ways, such as using principal component analysis to assess its efficiency or utilizing green patents to evaluate innovation performance. Comparatively, green patents provide richer information regarding green technology innovation and offer a better reflection of regional capabilities and levels of green innovation. In this study, the number of green patent applications was employed to measure the regional capacity for green technology innovation.

3.2.2. Explanatory Variable

Environmental regulation (ER): This study focuses on government-led initiatives, thus market-motivated environmental regulation and command-controlled environmental regulation were chosen as the primary areas of investigation.
Market-motivated environmental regulation (MER): It is primarily assessed based on the level of government investment in environmental protection and the subsidies provided to businesses for environmental protection purposes. As the study utilized cities at prefecture-level panel data, the availability of environmental pollution control investment data for individual cities beyond 2007 from the China Urban Statistical Yearbook is limited. Thus, considering local data availability, this study adopted the measurement approach introduced by Zhang and Xu (2022) [53], which estimated the intensity of market-motivated environmental regulation using the product of the government’s public financial budget expenditure and the natural logarithm of the solid waste utilization rate.
Command-controlled environmental regulation (CER): It can be measured using various indicators, including industrial wastewater, industrial sulfur dioxide, industrial solid waste emissions, energy consumption per unit of GDP, and analyzing the government work report to gauge the intensity of environmental regulation. Often, the intensity of environmental regulation is assessed by calculating a comprehensive index based on multiple pollutant indicators. In this study, the methodology proposed by Qiang, et al. [54] was adopted.
First, the unit pollutant emissions of each city are linearly standardized.
P i j s = P i j - min P j / max P j - min P j
Among them, Pij is the pollutant emission per unit output value of j pollutant in city I, max(Pj) and min(Pj) are the maximum and minimum values of each indicator in all cities, and Psij is the standardized value of the indicator.
Secondly, the adjustment coefficient is calculated. The proportion of pollutant emissions and the emission intensity of different pollutants in different cities differ greatly. The adjustment coefficient is used to approximate the difference of pollutant characteristics. The calculation formula of the adjustment coefficient is:
W j = P i j − P i j ¯
P i j ¯ is the urban average value of j pollutant emission per unit output value during the sample period.
Finally, the intensity of the command-controlled environmental regulation of each city is calculated.
C E R i = 1 3 ∑ j = 1 3 W j P i j s
CERi is the order-controlled environmental regulation intensity of city i.

3.2.3. Control Variable

To improve the accuracy of the estimation, this study incorporated various control variables. These variables include the level of economic development, measured by per capita GDP; the degree of opening up, measured by the proportion of output value from foreign-invested enterprises in the total industrial output value; the degree of marketization, measured by the ratio of private and individual employees to the total number of employees; infrastructure, measured by the ratio of actual paved road area to the land area of the prefecture-level city; greening rate of built-up areas, measured by the ratio of green area to built-up area; scientific research investment, measured by the ratio of R&D expenditure to GDP; and the degree of financial development, measured by the ratio of financial institutions’ loan balance at the end of the year to GDP.

3.3. Data Source and Descriptive Statistics

This study analyzed two sets of data. The first set is derived from the Statistical Yearbook of Chinese Cities, with exclusions made for regions with severe data deficiencies (Tibet Autonomous Region, Turpan City in Xinjiang Uygur Autonomous Region, Haidong City in Qinghai Province, Bijie City in Guizhou Province, Qinzhou City in Guangxi Zhuang Autonomous Region, Chaohu City in Anhui Province, and Sansha City in Hainan Province). The second set of data is obtained from the Green Patent Research Database (GPRD) of the China Research Data Service Platform (CNRDS) and is matched with local and municipal data. Since 2007, the country actively promoted green development. Additionally, taking into account the impact of the COVID-19 epidemic in 2020, the flow of green innovation factors among regions was constrained. Consequently, panel data for 284 prefecture-level cities spanning the period from 2007 to 2019 were collected. Because spatial econometric model analysis has a low tolerance for missing values, linear interpolation is used to fill in the gaps in the data.
Table 1 presents the descriptive statistics for the main variables. The relatively large standard deviations of each variable indicate noticeable differences between cities and regions. Additionally, the examination of the variable variance inflation factor reveals values below 2.5, with a mean of 1.64, indicating the absence of multicollinearity. Figure 2 illustrates the temporal and spatial distribution of green technological innovation in China. The data from six time intervals between 2007 and 2019 were selected for mapping. The figure reveals a progression of China’s green technological innovation from weak to strong. Throughout this process, the growth of green innovation generally improved, with no apparent polarization phenomenon. The development of green technology innovation capacity showed a comprehensive strengthening trend extending from core cities to regions, resulting in reduced regional disparities in green innovation capacity.

4. Research Results and Analysis

4.1. Spatial Correlation Test

According to the current practice of most scholars [55], this chapter adopts global Moran’s I index and local Moran’s I scatter plot to test the spatial correlation characteristics of green technology innovation in cities under different spatial weight matrices from 2007 to 2019. The expression of Moran’s I is as follows:
Moran ’ s   I = ∑ i = 1 n ∑ j = 1 n w i j ( x i − x ¯ ) ( x j − x ¯ ) s 2 ∑ i = 1 n ∑ j = 1 n w i j
where xi is the observed value and Wij represents the spatial weight matrix, S2 is the sample variance, namely:
S 2 = ∑ i = 1 n ( x i − x ¯ ) 2 n .
The global Moran’s I index is used to test the global spatial autocorrelation. When the index is positive, it means that the urban green technology innovation has a positive spatial correlation. On the contrary, there is a spatial negative correlation. In addition, the closer the absolute value of the index is to 1, the higher the spatial correlation degree of urban green technology innovation is. If the absolute value is equal to 0, then there is no spatial correlation. The study employs the Moran index to examine the global spatial autocorrelation of green technology innovation using various spatial weight matrices from 2007 to 2019. The results are presented in Table 2. Regardless of the distance weight matrix or the economic weight matrix, green technology innovation exhibits significant positive spatial correlation across different years, indicating clear spatial dependence. Therefore, the introduction of a spatial econometric model for analysis is warranted.
The global Moran index can reflect the overall characteristics of spatial correlation of each variable, but it cannot describe the type and degree of spatial correlation of specific regions. Therefore, it is necessary to further test the local spatial correlation of each variable by drawing the method of the Moran scatter plot. The Moran scatter plot divides the plane rectangular coordinate system into four quadrants, each of which represents a different type of spatial distribution. Among them, the first and third quadrants represent spatial positive correlation, which are called “high-high”-type and “low-low”-type spatial autocorrelation, respectively. The second and fourth quadrants represent spatial negative correlations, called “high-low” and “low-high” spatial autocorrelations, respectively. The study analysis on the local spatial autocorrelation of green technology innovation using the Moran scatter chart, as depicted in Figure 3. Whether utilizing the distance weight matrix or the economic weight matrix, urban areas exhibited agglomerations of both “high-high” and “low-low” patterns for green innovation, with the geographic weight matrix displaying a more pronounced agglomeration effect. The upper right quadrant of the graph indicates cities with high levels of green innovation, while the surrounding cities also exhibit high-high clustering, suggesting a potential spatial spillover effect.

4.2. Basic Regression

Hypothesis 1 of this paper states that there is a geographical spatial spillover effect of green innovation. Refer to Table 3 for the estimated outcomes of the spatial Durbin model. With respect to the geographic weight matrix, the spatial autoregressive coefficient ρ for models in columns (1) and (2) is significantly positive at a 1% level, indicating a positive spatial autocorrelation of green technology innovation. Conversely, under the economic weight matrix, the spatial term coefficient ρ in models of columns (3) and (4) is statistically insignificant, suggesting the absence of spatial autocorrelation for green technology innovation in terms of economic weight matrix. Therefore, there is no spillover of green technology innovation from cities with high economic development to cities with similar economic development levels. This outcome supports Hypothesis 1. The inconsistency with the previous Moran‘s I test could stem from the overlap between the geographic weight matrix and the economic weight matrix in the distribution of green technological innovation. The introduction of additional variable controls can yield more accurate and realistic results.
Under the geographical distance weight matrix, the coefficients of market-motivated environmental regulation (MERt−1) and command-controlled environmental regulation (CERt−1), along with their quadratic terms (MERt−12, CERt−12) in columns (1) and (2), are not significant. This indicates that the direct impact of market-motivated environmental regulation on the city itself is insignificant. However, the coefficient of the spatial lag term of market-motivated environmental regulation (W × MERt−1) is significantly negative at a 1% level, while its quadratic terms (W × MERt−12) are significantly positive at a 5% level. These results demonstrate a U-shaped cumulative proximity effect (feedback effect) of market-motivated environmental regulation. Specifically, the impact of urban market-motivated environmental regulation on surrounding cities, in turn, affects their own green technology innovation in a U-shaped manner. The spatial autoregressive coefficient of command-controlled environmental regulation (W × CERt−1) is significantly positive at a 1% level, whereas its quadratic term (W × CERt−12) is significantly negative at a 5% level, indicating an inverted U-shaped cumulative proximity effect. There is a significant cumulative proximity effect under the geographical distance weight matrix, suggesting the presence of spatial spillover effects. These findings can be attributed to the proximity between regions within the geographical feature matrix, whereby environmental regulation induces changes in the market environment, leading to the flow of green innovation elements and even the transfer of productive enterprises as carriers of green innovation. This aspect will be further analyzed below.

4.3. Spatial Effect Decomposition

Hypothesis 2 of this paper states that the geographical spillover effect of market-motivated environmental regulation on green innovation exhibits a U-shaped pattern. Table 4 presents the direct effect, indirect effect, and total effect of market-motivated environmental regulation. The direct effect represents the combined impact of the main effect and feedback effect of market-motivated environmental regulation on the city where it is implemented. The indirect effect refers to the influence of neighboring cities’ market-motivated environmental regulation on the green innovation of the studied city. In column (1), the coefficient for the intensity of market-motivated regulation and its quadratic term direct effect are not statistically significant, indicating that the direct impact of market-motivated regulation intensity on the city’s green technology innovation is not significant. In column (2), the coefficient for the indirect effect of market-motivated regulation (MERt−1) is significantly negative at a 5% level, while the coefficient for its quadratic term (MERt−12) indirect effect is significantly positive at a 5% level. These findings suggest a U-shaped relationship between the market-motivated regulation intensity of neighboring cities and its impact on the studied city. Therefore, Hypothesis 2 is confirmed, stating that increasing the intensity of market-motivated regulation in neighboring cities will initially reduce the green technological innovation of the studied cities but will eventually improve it. This may be attributed to the “siphon effect” caused by the intensified market-motivated regulation, which attracts green innovation elements to concentrate in neighboring cities. Once the market-motivated regulation intensity reaches a certain threshold in neighboring cities, a positive “spillover effect” occurs, enhancing the local green technology innovation capacity.
Hypothesis 3 of this paper states that the geospatial spillover effect of command-controlled environmental regulation on green innovation exhibits an inverted U-shaped pattern. As depicted in Table 5, the direct effect coefficient of command-controlled environmental regulation (CERt−1) in column (1) is positively but insignificantly related to green technology innovation in the city. However, its quadratic term (CERt−12) exhibits a significant negative relationship at the 10% significance level, indicating an inverted U-shaped pattern. When the intensity of order control regulation increases, enterprises in the city tend to reduce costs and seek green technology innovation as a means to mitigate the impact of command-controlled regulation. Nevertheless, green technology innovation necessitates long-term and uncertain investments, leading to enterprises’ reluctance to bear additional costs. Consequently, they may introduce green innovation elements from neighboring cities, resulting in a “siphon effect”. Beyond a certain threshold, higher levels of command-controlled regulation can lead to the gradual relocation of enterprises to regions with less stringent environmental regulations. As key drivers and enablers of green technology innovation, enterprises also transfer their green technology innovation elements, ultimately diminishing the capacity for green technology innovation. Furthermore, in column (2), the indirect effect coefficient of command-controlled environmental regulation (CERt−1) is significantly positive at the 5% level, while the indirect effect coefficient of its quadratic term (CERt−12) is significantly negative at the 1% level. This finding confirms Hypothesis 3 and demonstrates an inverted U-shaped relationship between the command-controlled environmental regulation in neighboring cities and local cities. The industry exhibits agglomeration effects, attracting related enterprises along the industry chain and supporting businesses to concentrate in both cities and surrounding areas. The spillover effect of green innovation induced by command-controlled regulations in neighboring cities propagates to the industry chain enterprises located in focal cities, thus exhibiting similar trends in green innovation. These results highlight that enhancing the intensity of command-controlled regulation elicits an inverted U-shaped direct and indirect effect on green technology innovation. Importantly, command-controlled environmental regulations can prompt regulated enterprises to relocate, leading to the outflow of green innovation elements and subsequent transformations in the urban industrial structure.

4.4. An Analysis of the Intermediary Effect of Industrial Structure

Most of the pollution-type enterprises primarily belong to the secondary industry, which comprises productive industrial enterprises. The relocation of these enterprises results in alterations in the local industrial structure. Therefore, this study examines the urban industrial structure to analyze the transfer of pollution-type enterprises within cities. To test the hypothesis regarding the influence of environmental regulations on the transfer of polluting enterprises, this paper adopts the intermediary effect test method described by Wen et al. (2004) [56]. By employing a geographical distance spatial matrix, we construct the following spatial Durbin model to explore the impact of environmental regulation on the transfer of polluting industries. The measurement of the industrial structure (Struc) entails calculating the ratio of tertiary industry value-added to secondary industry value-added. The remaining model settings are as follows.
G T P i t = α + ρ W G T P i t + β 1 E R i t − 1 + β 2 E R i t − 1 2 + β 3 W E R i t − 1 + β 4 W E R i t − 1 2 + δ C O N T R i t + λ i + η t + ε i t
S t r u c i t = α + ρ W S t r u c i t + β 1 E R i t − 1 + β 2 E R i t − 1 2 + β 1 W E R i t − 1 + β 2 W E R i t − 1 2 + δ C O N T R i t + λ i + η t + ε i t
G T P i t = α + ρ W G T P i t + β 1 E R i t − 1 + β 2 E R i t − 1 2 + β 3 W E R i t − 1 + β 4 W E R i t − 1 2 + β 5 S t r u c i t − 1 + δ C O N T R i t + λ i + η t + ε i t
Hypothesis 4 of this paper states that command-controlled environmental regulation induces adjustments in regional industrial structure, consequently resulting in the spatial spillover of green innovation. The regression results presented in Table 6 represent the direct effects of explanatory variables on the explained variables, encompassing both main effects and feedback effects. Since the direct effects of market incentive environmental regulations in the regression results above are not significant (Table 4), we focus solely on examining the intermediary effects of command-controlled environmental regulation. In column (1), a significantly inverted U-shaped relationship is observed between command-controlled environmental regulation and green technology innovation. In column (2), a significantly U-shaped relationship emerges between command-controlled environmental regulation and industrial structure. However, in column (3), the impact of command-controlled environmental regulation is found to be insignificant, whereas industrial structure (industrial structure t−1) significantly diminishes green technology innovation at a 1% level of significance. These findings indicate a mediating effect of the industrial structure. Since the command-controlled environmental regulation in column (3) is not significant, it can be concluded that the industrial structure fully mediates the relationship. Hypothesis 4 is thus validated. Command-controlled environmental regulation facilitates the reconfiguration of regional industrial structure, consequently reducing green technological innovation. This process leads to the transfer of low-end industries and a spillover effect on green technological innovation. Although the intermediary effect test yields complete mediation, considering the regional heterogeneity in the impact of environmental regulation on green technology innovation, the overall regression exhibits an insignificant pattern. Therefore, it can be inferred that the spillover effect of environmental regulation on green technology innovation is primarily manifested through changes in the industrial structure resulting from the transfer of low-end industries.

5. Robustness Test

5.1. Replacement Space Weight Matrix

The research further verifies the regression results by replacing the spatial weight matrix, and the replacement geographical distance weight matrix is an adjacency matrix (W3ij). When city i and city j are adjacent, W3ij = 1; if it is not adjacent, W3ij = 0. The degree of urban economic development is also closely related to the number of local talents. Considering that the spatial spillover of ordinary innovation will spill over to cities with similar economies and talents, in order to distinguish the different spillover characteristics of green technology innovation and ordinary innovation, a talent weight matrix (W4ij) is constructed. W4ij =1/|Hi − Hj|, Hi, and Hj represents the average higher education rate of city i and city j from 2007 to 2019, respectively. It can be seen from Table 7 that the spatial autoregression coefficient is ρ. The regression results are consistent with the significance of the benchmark regression results in Table 3, which verifies the robustness and reliability of the benchmark model estimation results.

5.2. Add the Time Lag Term of The Explained Variable

To mitigate the issue of endogeneity, the explanatory variables of the model include a time lag term for green technological innovation. The regression results are presented in Table 8, and they demonstrate consistency with the benchmark regression results, thereby further confirming the robustness of the model.
The study conducted regression tests using the spatial autoregression model (SAR) and spatial error model (SEM). The regression results for the main variables are largely consistent with the previous findings, thereby confirming the robustness of the research results.

6. Research Conclusions and Policy Recommendations

This study analyzes the spatial spillover effect of environmental regulation on urban green technology innovation from the perspective of spatial spillover effect and examines the mediating role of industrial structure. By using the panel data of 284 Chinese prefecture-level cities from 2007 to 2019, the spillover effect of green technology innovation is tested. It is found that the spatial spillover of green technology innovation is mainly manifested through geographical proximity and does not extend to cities with a similar economic development level, such as ordinary technology innovation. This difference stems from the uniqueness of green innovation and the transfer of green technology innovation elements from economically developed areas to economically underdeveloped areas. The direct impact of market-motivated environmental regulation on urban green innovation is not significant. However, its main manifestation is the U-shaped space spillover effect. Market incentives in neighboring cities create a negative “siphon effect” that attracts elements of green innovation from surrounding areas. When the market-motivated environmental regulation reaches a sufficient level, the enterprise’s green technology innovation motivation will increase and the enterprise’s green innovation ability will be enhanced. This process will also have positive “spillover effects” on surrounding cities. Command-controlled environmental regulation has both an inverted U-shaped direct effect and a spatial spillover effect on urban green technology innovation. This regulation encourages companies to actively innovate in green technologies and upgrade their technologies to meet environmental standards. If the cost increase caused by the intensity of command-controlled environmental regulation is beyond the ability of the enterprise, the enterprise may choose to transfer industry. Therefore, the elements of green technology innovation are transferred together with these enterprises, leading to the upgrading of the urban industrial structure. Industrial structure upgrading is the mediating variable of the relationship between mandatory environmental regulation and green technology innovation. The spillover effect of green technology innovation caused by command-controlled environmental regulation is mainly the transfer of low-end industries promoted by such regulation, which then leads to the transfer of green innovation elements.
Based on the above research, the empirical conclusions of this paper have the following important policy implications for the rational planning and implementation of environmental regulations in local cities and the improvement of urban green innovation ability. Firstly, it is crucial to increase investment in green technology research and development (R&D) as well as green technology demonstration. Given the external nature of green technology innovation, government promotion becomes particularly significant. Investing in green innovation R&D is an effective approach to facilitate independent green technology innovation by enterprises. Considering the characteristics of local industries, the government should provide support for green innovative technologies that promote energy conservation, emission reduction, and industrial upgrading. The government can effectively leverage the geographical spillover effect of green technological innovation to drive the advancement and transformation of relevant enterprises’ green technologies. Secondly, it is essential to establish comprehensive frameworks for both market-motivated environmental regulation and command-controlled environmental regulation. To prevent the “crowding out effect” on enterprises resulting from the cost increase caused by command-controlled environmental regulation, offsetting measures through incentives provided by market-motivated environmental regulation should be implemented. Introducing indirect approaches such as green finance and utilizing financial funds as leverage can stimulate larger-scale financial capital, thereby promoting green credit, guarantees, bonds, and funds to support green technology innovation and facilitate the development and application of green technology. Thirdly, it is crucial to appropriately manage the intensity of command-controlled environmental regulation to foster industrial upgrading and transformation. Implementation of science-based policies and measures is necessary to ensure effective policy implementation, prevent industry transfer and pollution resulting from inadequate policy implementation, reduce the hollowing of regional economies, sustain small and medium-sized enterprises’ vitality, provide increased support to enterprises, encourage their active involvement in green innovation, enhance their capabilities in green technology innovation, and achieve sustainable development at the regional level. Fourthly, it is essential to develop environmental protection policies that align with local industries and innovative development while considering the environmental regulations of surrounding cities. Measures should be taken to prevent the occurrence of policy “depressions” resulting from inappropriate environmental policy settings, which can lead to the formation of “pollution shelters” in the region and make it a destination for the transfer of high-energy consumption and high-pollution capacity. Additionally, avoiding the transfer of green innovation elements alongside industries due to excessive or unreasonable environmental regulations is crucial for maintaining the sustainable development of the regional economy.

Author Contributions

Conceptualization, X.Z.; Methodology, X.Z.; Software, X.Z.; Validation, X.Z.; Formal analysis, X.Z.; Data curation, X.Z.; Writing – original draft, X.Z.; Writing – review & editing, F.X.; Supervision, F.X.; Project administration, F.X.; Funding acquisition, F.X. All authors have read and agreed to the published version of the manuscript.

Funding

This paper is supported by the Natural Science Foundation of Guangdong Province, China (2022A1515011046).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Du, K.; Li, J. Towards a green world: How do green technology innovations affect total-factor carbon productivity. Energy Policy 2019, 131, 240–250. [Google Scholar] [CrossRef]
  2. Ley, M.; Stucki, T.; Woerter, M. The Impact of Energy Prices on Green Innovation. Energy J. 2016, 37, 41–75. [Google Scholar] [CrossRef]
  3. Cohen, M.C.; Lobel, R.; Perakis, G. The Impact of Demand Uncertainty on Consumer Subsidies for Green Technology Adoption. Manag. Sci. 2016, 5, 1235–1258. [Google Scholar] [CrossRef]
  4. Philipp, B. The Allocation and Effectiveness of China’s R&D Subsidies—Evidence from Listed Firms. Res. Policy 2016, 9, 1774–1789. [Google Scholar] [CrossRef]
  5. Christos, D.; Geoff, P. The Effectiveness of R&D Subsidies: A Meta-regression Analysis of the Evaluation Literature. Res. Policy 2016, 4, 797–815. [Google Scholar] [CrossRef]
  6. Hattori, K. Optimal Combination of Innovation and Environmental Policies under Technology Licensing. Econ. Model. 2017, 8, 601–609. [Google Scholar] [CrossRef]
  7. Cai, W.G.; Li, G.P. The Drivers of Eco-innovation and Its Impact on Performance: Evidence from China. J. Clean. Prod. 2018, 3, 110–118. [Google Scholar] [CrossRef]
  8. Rexhäuser, S.; Rammer, C. Environmental Innovations and Firm Profitability: Unmasking the Porter Hypothesis. Environ. Resour. Econ. 2014, 1, 145–167. [Google Scholar] [CrossRef]
  9. Caniels, M.C. Knowledge Spillovers and Economic Growth: Regional Growth Differentials across Europe; Edward Elgar: Cheltenham, UK; Northampton, MA, USA, 2000. [Google Scholar]
  10. Jinji, N.; Zhang, X.; Haruna, S. Trade patterns and international technology spillovers: Evidence from patent citations. Rev. World Econ. 2015, 151, 635–658. [Google Scholar] [CrossRef]
  11. Shao, X.; Liu, S.; Ran, R.; Liu, Y. Environmental regulation, market demand, and green innovation: Spatial perspective evidence from China. Environ. Sci. Pollut. Res. 2022, 29, 63859–63885. [Google Scholar] [CrossRef]
  12. Ren, Y.M. Industrial Investment Funds, Government R&D Subsidies, and Technological Innovation: Evidence from Chinese Companies. Front. Psychol. 2022, 13, 890208. [Google Scholar] [CrossRef] [PubMed]
  13. Zhang, Y.; Chen, H.; He, Z. Environmental Regulation, R&D Investment, and Green Technology Innovation in China: Based on the Pvar Model. PLoS ONE 2022, 17, e0275498. [Google Scholar] [CrossRef]
  14. Fang, Y.; Shao, Z. Whether Green Finance Can Effectively Moderate the Green Technology Innovation Effect of Heterogeneous Environmental Regulation. Int. J. Environ. Res. Public Health 2022, 19, 3646. [Google Scholar] [CrossRef] [PubMed]
  15. Niu, H.; Zhao, X.; Luo, Z.; Gong, Y.; Zhang, X. Green Credit and Enterprise Green Operation: Based on the Perspective of Enterprise Green Transformation. Front. Psychol. 2022, 13, 1041798. [Google Scholar] [CrossRef]
  16. Wang, P.; Dong, C.; Chen, N.; Qi, M.; Yang, S.; Nnenna, A.B.; Li, W. Environmental Regulation, Government Subsidies, and Green Technology Innovation—A Provincial Panel Data Analysis from China. Int. J. Environ. Res. Public Health 2021, 18, 11991. [Google Scholar] [CrossRef]
  17. Shen, L.; Fan, R.; Wang, Y.; Yu, Z.; Tang, R. Impacts of Environmental Regulation on the Green Transformation and Upgrading of Manufacturing Enterprises. Int. J. Environ. Res. Public Health 2020, 17, 7680. [Google Scholar] [CrossRef] [PubMed]
  18. Hu, Y.; Sun, S.; Dai, Y. Environmental Regulation, Green Innovation, and International Competitiveness of Manufacturing Enterprises in China: From the Perspective of Heterogeneous Regulatory Tools. PLoS ONE 2021, 16, e0249169. [Google Scholar] [CrossRef]
  19. Zhang, H.; Wang, Y.; Li, R.; Si, H.; Liu, W. Can Green Finance Promote Urban Green Development? Evidence from Green Finance Reform and Innovation Pilot Zone in China. Environ. Sci. Pollut. Res. 2023, 30, 12041–12058. [Google Scholar] [CrossRef]
  20. Zhu, Y.; Sun, Z.; Zhang, S.; Wang, X. Economic Policy Uncertainty, Environmental Regulation, and Green Innovation—An Empirical Study Based on Chinese High-Tech Enterprises. Int. J. Environ. Res. Public Health 2021, 18, 9503. [Google Scholar] [CrossRef]
  21. Lee, C.Y. The Differential Effects of Public R&D Support on Firm R&D: Theory and Evidence from Multi-country Data. Technovation 2011, 31, 256–269. [Google Scholar] [CrossRef]
  22. Braun, E.; Wield, D. Regulation as a Means for the Social Control of Technology. Technol. Anal. Strateg. Manag. 1994, 6, 259–272. [Google Scholar] [CrossRef]
  23. Laursen, K.; Salter, A. Open for innovation: The role of openness in explaining innovation performance among U.K. manufacturing firms. Strateg. Manag. J. 2006, 27, 131–150. [Google Scholar] [CrossRef]
  24. Porter, M.E. America’s green strategy. Sci. Am. 1991, 264, 168–179. [Google Scholar] [CrossRef]
  25. Li, X.Y. FDI, Environmental Regulation and Industrial Structure Optimization: An Empirical Evidence Based on Spatial Econometrics Model. Mod. Econ. Sci. 2018, 40, 104–128. Available online: http://www.cqvip.com/qk/96310x/20182/7000463281.html (accessed on 1 March 2022).
  26. Song, M.L.; Wang, S.H. Analysis of Environmental Regulation, Technological Progression and Economic Growth from the Perspective of Statistical Tests. Econ. Res. J. 2013, 48, 122–134. Available online: http://www.cqvip.com/qk/95645x/20133/45125188.html (accessed on 5 March 2022).
  27. Li, X.A. Environmental Regulation, Government Subsidies and Regional Green Technology Innovation. Econ. Surv. 2021, 38, 14–23. [Google Scholar] [CrossRef]
  28. Lim, U. The Spatial Distribution of Innovative Activity in U.S. Metropolitan Areas: Evidence from Patent Data. J. Reg. Anal. Policy 2003, 33, 97–126. Available online: https://jrap.scholasticahq.com/article/9388.pdf (accessed on 15 March 2022).
  29. Zhang, P.; Chen, P.; Xiao, F.; Sun, Y.; Ma, S.; Zhao, Z. The Impact of Information Infrastructure on Air Pollution: Empirical Evidence from China. Int. J. Environ. Res. Public Health 2022, 19, 14351. [Google Scholar] [CrossRef]
  30. Keller, W. Geographic Localization of International Technology Diffusion. Am. Econ. Rev. 2002, 92, 120–142. [Google Scholar] [CrossRef]
  31. Huang, S.; Bai, Y.; Tan, Q. How Does the Concentration of Determinants Affect Industrial Innovation Performance?—An Empirical Analysis of 23 Chinese Industrial Sectors. PLoS ONE 2017, 12, e0169473. [Google Scholar] [CrossRef]
  32. Zhang, P.; Yu, H.; Shen, M.; Guo, W. Evaluation of Tourism Development Efficiency and Spatial Spillover Effect Based on Ebm Model: The Case of Hainan Island, China. Int. J. Environ. Res. Public Health 2022, 19, 3755. [Google Scholar] [CrossRef]
  33. Yang, G.; Gong, G.; Luo, Y.; Yang, Y.; Gui, Q. Spatiotemporal Characteristics and Influencing Factors of Tourism–Urbanization–Technology–Ecological Environment on the Yunnan–Guizhou–Sichuan Region: An Uncoordinated Coupling Perspective. Int. J. Environ. Res. Public Health 2022, 19, 8885. [Google Scholar] [CrossRef]
  34. Franco, C.; Marin, G. The Effect of Within-Sector, Upstream and Downstream Environmental Taxes on Innovation and Productivity. Environ. Resour. Econ. 2015, 66, 261–291. [Google Scholar] [CrossRef]
  35. Yao, M.; Duan, J.; Wang, Q. Spatial and Temporal Evolution Analysis of Industrial Green Technology Innovation Efficiency in The Yangtze River Economic Belt. Int. J. Environ. Res. Public Health 2022, 19, 6361. [Google Scholar] [CrossRef] [PubMed]
  36. Qin, C.L.; Liu, Y.X.; Li, C. Spatial Spillovers and the Convergence of Regional Economic Growth: A Case Study Based on the Yangtze River Delta. Soc. Sci. China 2013, 34, 159–173. [Google Scholar] [CrossRef]
  37. Beckmann, M.J.; Puu, T. Spatial Economics: Density, Potential, and Flow; North Holland Publishing Co.: Amsterdam, The Netherlands, 1985; Available online: https://pure.iiasa.ac.at/2597 (accessed on 15 March 2022).
  38. Qiu, J.; Liu, W.; Ning, N. Evolution of Regional Innovation with Spatial Knowledge Spillovers: Convergence or Divergence? Netw. Spat. Econ. 2020, 20, 179–208. [Google Scholar] [CrossRef]
  39. Martin, R.; Sunley, P. Slow Convergence? The New Endogenous Growth Theory and Regional Development. Econ. Geogr. 1998, 74, 201–207. [Google Scholar] [CrossRef]
  40. Aldieri, L.; Brahmi, M.; Bruno, B.; Vinci, C.P. Circular Economy Business Models: The Complementarities with Sharing Economy and Eco-Innovations Investments. Sustainability 2021, 13, 12438. [Google Scholar] [CrossRef]
  41. Hagerstrand, T. Innovation Diffusion as a Spatial Process; University of Chicago Press: Chicago, IL, USA, 1968. [Google Scholar] [CrossRef]
  42. Beckmann, M.J. The Analysis of Spatial Diffusion Processes. Papers Reg. Sci. Assoc. 1970, 25, 108–117. [Google Scholar] [CrossRef]
  43. Wang, X.; Wang, S.; Zhang, Y. The Impact of Environmental Regulation and Carbon Emissions on Green Technology Innovation from the Perspective of Spatial Interaction: Empirical Evidence from Urban Agglomeration in China. Sustainability 2022, 14, 5381. [Google Scholar] [CrossRef]
  44. Wu, Q.; Li, Y.; Wu, Y.; Li, F.; Zhong, S. The Spatial Spillover Effect of Environmental Regulation on the Total Factor Productivity of Pharmaceutical Manufacturing Industry in China. Sci. Rep. 2022, 12, 11642. [Google Scholar] [CrossRef]
  45. Luigi, A.; Mohsen, B.; Xihui, C.; Concetto, P.V. Knowledge spillovers and technical efficiency for cleaner production: An economic analysis from agriculture innovation. J. Clean. Prod. 2021, 320, 128830. [Google Scholar] [CrossRef]
  46. Feng, Z.; Zeng, B.; Ming, Q. Environmental Regulation, Two-Way Foreign Direct Investment, and Green Innovation Efficiency in China’s Manufacturing Industry. Int. J. Environ. Res. Public Health 2018, 15, 2292. [Google Scholar] [CrossRef] [PubMed]
  47. Liang, P.; Xie, S.; Qi, F.; Huang, Y.; Wu, X. Environmental Regulation and Green Technology Innovation under the Carbon Neutrality Goal: Dual Regulation of Human Capital and Industrial Structure. Sustainability 2023, 15, 2001. [Google Scholar] [CrossRef]
  48. Anselin, L.; Bera, A.K.; Florax, R.; Mann, J.; Yoon, M.J. Simple Diagnostic Tests for Spatial Dependence. Reg. Sci. Urban Econ. 1996, 26, 77–104. [Google Scholar] [CrossRef]
  49. Tao, C.; Yang, H. Spatial Econometric Model Selection and Simulation Analysis. Stat. Res. 2014, 31, 88–96. Available online: http://www.cqvip.com/qk/96472x/201408/662086769.html (accessed on 20 March 2022).
  50. Scherngell, T.; Borowiecki, M.; Hu, Y. Effects of Knowledge Capital on Total Factor Productivity in China: A Spatial Econometric Perspective. China Econ. Rev. 2014, 29, 82–94. [Google Scholar] [CrossRef]
  51. Lesage, J.; Pace, R.K. Introduction to Spatial Econometrics; CRC Press: Boca Raton, FL, USA, 2009. [Google Scholar] [CrossRef]
  52. Chen, D.L.; Hu, W.B.; Li, Y.Y.; Zhang, C.Z.; Lu, X.H.; Cheng, H. Exploring the Temporal and Spatial Effects of City Size on Regional Economic Integration: Evidence from the Yangtze River Economic Belt in China. Land Use Policy 2023, 132, 106770. [Google Scholar] [CrossRef]
  53. Wang, L.; Long, Y.; Li, C. Research on the impact mechanism of heterogeneous environmental regulation on enterprise green technology innovation. J. Environ. Manag. 2022, 322, 116127. [Google Scholar] [CrossRef]
  54. Qiang, O.; Tian-Tian, W.; Ying, D.; Zhu-Ping, L.; Jahanger, A. The Impact of Environmental Regulations on Export Trade at Provincial Level in China: Evidence from Panel Quantile Regression. Environ. Sci. Pollut. Res. 2022, 29, 24098–24111. [Google Scholar] [CrossRef]
  55. Xie, R.; Fu, W.; Yao, S.L.; Zhang, Q. Effects of Financial Agglomeration on Green Total Factor Productivity in Chinese Cities: Insights from an Empirical Spatial Durbin Model. Energy Econ. 2021, 101, 105449. [Google Scholar] [CrossRef]
  56. Wen, Z.; Zhang, L.; Hou, J.; Liu, H. Mediation Effect Test Procedure and Its Application. J. Psychol. 2004, 36, 614–620. [Google Scholar] [CrossRef]
Sustainability 15 14133 g001
Figure 1. The number of green policies and green patent applications in different stages in China. Note: according to the “Green Finance Policy Collection” published by the Ministry of Ecology and Environment of China.
Figure 1. The number of green policies and green patent applications in different stages in China. Note: according to the “Green Finance Policy Collection” published by the Ministry of Ecology and Environment of China.
Sustainability 15 14133 g001
Sustainability 15 14133 g002
Figure 2. Spatial and temporal distribution of green innovation. Note: the base map is from the standard map service website of the Ministry of Natural Resources.
Figure 2. Spatial and temporal distribution of green innovation. Note: the base map is from the standard map service website of the Ministry of Natural Resources.
Sustainability 15 14133 g002
Sustainability 15 14133 g003
Figure 3. Moran scatter.
Figure 3. Moran scatter.
Sustainability 15 14133 g003
Table 1. Descriptive statistics of the main variables.
Table 1. Descriptive statistics of the main variables.
VariableObsAveStdMinMax
Green technology innovation (GTP)33960.0930.0410.0220.203
Market-motivated environmental regulation intensity (MER)339610.0540.5914.81311.673
Command-controlled environmental regulation intensity (CER)33960.6220.4410.0011.832
industrial structure (IS)33960.8830.4320.2612.791
Economic development level (ED)33964.2722.9910.69315.552
Openness (OP)33960.1230.1640.0000.803
Marketization degree (MD)33960.9640.5920.1303.211
Infrastructure (INF)33961.1221.2010.0326.143
Greening rate of built-up area (GB)33962.3523.2010.00015.521
Investment in scientific research (RD)33960.0000.0000.0000.011
Financial development degree (FD)33960.8620.4920.2712.900
Table 2. Global Moran index.
Table 2. Global Moran index.
Moran IndexGeographic Distance MatrixEconomic Distance Matrix
20070.103 ***0.234 ***
20080.102 ***0.250 ***
20090.121 ***0.280 ***
20100.114 ***0.270 ***
20110.120 ***0.271 ***
20120.119 ***0.265 ***
20130.106 ***0.274 ***
20140.112 ***0.270 ***
20150.120 ***0.267 ***
20160.117 ***0.288 ***
20170.124 ***0.267 ***
20180.132 ***0.263 ***
20190.109 ***0.269 ***
Note: *** represent 1% significance levels, respectively.
Table 3. Benchmark regression results.
Table 3. Benchmark regression results.
VariablesWeight Matrix of Geographical DistanceWeight Matrix of Economic Distance
(1)(2)(3)(4)
ρ0.873 ***
(0.025)		0.859 ***
(0.029)		0.027
(0.043)		0.031
(0.042)
MERt−10.003
(0.032)		 −0.018
(0.034)
MERt−12−0.000
(0.003)		 0.002
(0.003)
W × MERt−1−0.996 ***
(0.379)		 −0.210 **
(0.085)
W × MERt−120.086 **
(0.036)		 0.020 **
(0.008)
CERt−1 0.0384
(0.094)		 0.113
(0.098)
CERt−12 −0.020
(0.051)		 −0.068
(0.054)
W × CERt−1 3.507 ***
(0.998)		 0.000
(0.299)
W × CERt−12 −2.411 ***
(0.583)		 −0.071
(0.156)
ED0.005
(0.013)		0.008
(0.013)		0.011
(0.014)		0.009
(0.014)
OP0.216
(0.318)		0.143
(0.314)		0.386
(0.328)		0.355
(0.324)
MD0.040
(0.026)		0.042
(0.026)		0.038
(0.029)		0.034
(0.028)
INF−0.034 *
(0.018)		−0.041 **
(0.018)		−0.025
(0.017)		−0.024
(0.017)
GB0.007 *
(0.004)		0.008 **
(0.004)		0.007 *
(0.004)		0.008 *
(0.004)
RD90.100 ***
(13.570)		93.770 ***
(13.800)		103.300 ***
(13.850)		102.600 ***
(13.950)
FD−0.015
(0.051)		−0.023
(0.049)		−0.041
(0.055)		−0.044
(0.054)
Year and individual fixed effectsYesYesYesYes
R20.2530.2510.3190.314
Obs3113311331133113
Note: *, ** and *** represent 10%, 5%, and 1% significance levels, respectively, and Z value is in brackets. The following table is the same.
Table 4. Spatial spillover effect of green technology innovation (MER).
Table 4. Spatial spillover effect of green technology innovation (MER).
VariablesDirect EffectIndirect EffectTotal Effect
(1)(2)(3)
MERt-1−0.026
(0.035)		−8.082 **
(3.549)		−8.108 **
(3.562)
MERt−120.002
(0.003)		0.707 **
(0.330)		0.709 **
(0.332)
ρ0.873 ***
(0.025)		0.873 ***
(0.025)		0.873 ***
(0.025)
ControlYesYesYes
Year and individual fixed effectsYesYesYes
R20.3740.3740.374
Obs311331133113
Note: ** and *** represent 5%, and 1% significance levels, respectively. The control variables are the level of economic development, the degree of openness, the degree of marketization, infrastructure, the greening rate of built-up areas, the intensity of scientific research investment, and the degree of financial development. The same is true for the following table.
Table 5. Spatial spillover effect of green technology innovation (CER).
Table 5. Spatial spillover effect of green technology innovation (CER).
VariablesDirect EffectIndirect EffectTotal Effect
(1)(2)(3)
CERt−10.144
(0.100)		27.080 **
(10.990)		27.230 **
(11.020)
CERt−12−0.092 *
(0.055)		−18.640 ***
(7.152)		−18.730 ***
(7.171)
ρ0.859 ***
(0.029)		0.859 ***
(0.029)		0.859 ***
(0.029)
ControlYesYesYes
Year and individual fixed effectsYesYesYes
R20.1610.1610.161
Obs311331133113
Note: *, ** and *** represent 10%, 5%, and 1% significance levels, respectively.
Table 6. Test of the intermediary effect of industrial structure.
Table 6. Test of the intermediary effect of industrial structure.
VariablesGTPISGTP
(1)(2)(3)
CERt−10.144
(0.100)		−0.090 **
(0.043)		0.115
(0.103)
CERt-12−0.092 *
(0.055)		0.048 **
(0.023)		−0.074
(0.057)
Industrial structuret−1 −0.229 ***
(0.084)
ρ0.832 ***
(0.038)		0.829 ***
(0.036)		0.852 ***
(0.032)
ControlYesYesYes
Year and individual fixed effectsYesYesYes
R20.1610.4520.160
Obs311331133113
Note: *, ** and *** represent 10%, 5%, and 1% significance levels, respectively.
Table 7. Robustness test (replacement weight matrix).
Table 7. Robustness test (replacement weight matrix).
VariablesAdjacency MatrixTalent Matrix
(1)(2)(3)(4)
ρ0.320 ***
(0.033)		0.320 ***
(0.034)		−0.039
(0.052)		−0.037
(0.053)
ControlYesYesYesYes
Year and individual fixed effectsYesYesYesYes
R20.2620.2420.3450.337
Obs3113311331133113
Note: *** represent 1% significance levels.
Table 8. Robustness test (add time lag item).
Table 8. Robustness test (add time lag item).
Variables(1)(2)
W × MERt−11.572 ***
(0.332)
W × MERt−12−0.159 ***
(0.032)
W × CERt−1 −9.051 ***
(0.826)
W × CERt−12 8.123 ***
(0.501)
ρ1.875 ***
(0.088)		15.490 ***
(0.088)
GTPt−10.116 ***
(0.041)		−0.028
(0.041)
ControlYesYes
Year and individual fixed effectsYesYes
R20.1720.163
Obs31133113
Note: *** represent 1% significance levels.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Zhang, X.; Xu, F. Environmental Regulation and Spatial Spillover Effect of Green Technology Innovation: An Empirical Study on the Spatial Durbin Model. Sustainability 2023, 15, 14133. https://doi.org/10.3390/su151914133

AMA Style

Zhang X, Xu F. Environmental Regulation and Spatial Spillover Effect of Green Technology Innovation: An Empirical Study on the Spatial Durbin Model. Sustainability. 2023; 15(19):14133. https://doi.org/10.3390/su151914133

Chicago/Turabian Style

Zhang, Xin, and Feng Xu. 2023. "Environmental Regulation and Spatial Spillover Effect of Green Technology Innovation: An Empirical Study on the Spatial Durbin Model" Sustainability 15, no. 19: 14133. https://doi.org/10.3390/su151914133

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop