2.1 Research on environmental regulation itself

The idea that environmental pollution has strong negative externalities is considered to be a consensus view among academics and policymakers, and as a result, environmental regulation policies have become a hot spot for research.Generally speaking, environmental regulation involves government intervention in the production and operations of companies through administrative measures to achieve the goal of harmonizing environmental protection and sustainable growth.The meaning of environmental regulation is constantly evolving and changing over time. In addition to the rules and regulations implemented by the government through administrative means, it also includes market-based incentive supervision, as well as guidance and regulation of public behavior. This indicates that environmental regulation is not limited to traditional regulatory enforcement, but also includes influencing and guiding individual and organizational environmental behavior through economic incentives and social ideology [10]. Environmental regulation can be divided into three main categories based on its implementation methods: command based, market based, and voluntary. Command based regulation, also known as ER1, relies on mandatory policies implemented by administrative departments. Market based regulation, or ER2, is a method based on market mechanisms and economic incentives. Voluntary regulation, referred to as ER3, relies on public participation and supervision for implementation. This classification reflects the diversity of environmental regulatory strategies aimed at achieving environmental goals through different approaches and methods [11]. Environmental regulations can be classified in various ways, with one common classification being formal regulations and informal regulations. Formal regulations are those based on specific laws or regulations, while informal legal rules are based on non statutory standards or agreements. This classification method highlights the different sources and enforcement methods of environmental regulations, showcasing diverse approaches to environmental protection [12]. Regardless of the type, environmental control measures fundamentally guide enterprises to adjust their production and technological decisions, internalizing their external costs [13]. However, the current definition of environmental regulation is inconsistent, and there is still significant controversy over the methods and intensity of regulation, which has led to different empirical literature substituting regulatory measures. The table below mainly shows the variables of environmental control substitutes selected from existing studies.

In this study, the reason for choosing the proxy variables were as follows:

1. For ER1, according to the literature, the number of regulations and thenumber of administrative penalty cases are used as contingency agents for environmental regulation [14]. These proxy variables can reflect the government ’s regulation of environmental behavior and reveal the intensity and effect of environmental regulation.
2. For ER2, according to the literature, indicators such as pollution discharge fees, industrial pollution control investment, and the number of environmental acceptance projects are used as contingency agents for environmental regulation [15]. These proxy variables can reflect the economic incentives and normative pressures on environmental behavior of companies, and facilitate the implementation of more environmentally friendly measures by companies.
3. For ER3, according to the literature, the received environmental information, the number of visits and the real number of people involved in environmental protection are used as contingency agents for environmental supervision [16]. These proxy variables can reflect the degree of public concern andparticipation in environmental issues, and have an impact on the implementation of environmental regulatory policies.
4. For Formal ER, according to the literature, indicators such as pollutant emission removal rates, pollution control investment rates, and pollution are used as contingency agents for environmental regulation [17]. These proxy variables can effectively drive the adjustment of industrial structure.
5. For informal formal ER, according to the literature, contingency agents for environmental regulation are used with indicators such as environmental awareness, per capita income, and education level [18]. These proxy variables can enhance people’s attention and willingness to act on environmental issues, thus facilitating the effective implementation of environmental regulation.

Given the diversity of environmental regulation policymaking in China.In this article, the stimulus effect of government environmental laws on enterprise GTI is emphasized by us. In the empirical analysis section of this article, three policies will be combined with a separate policy to compare their impact on green environmental protection with ER1, ER2, and ER3.

2.2 Research on the relationship between environmental regulation and green technology innovation

Porter followed the logical framework of "endogenous growth" and for the first time combined technological innovation and environmental regulation into a unified theoretical framework, and based on this, proposed the Porter hypothesis. On this basis, many studies have been conducted on this hypothesis. Among these methods, the most direct approach to testing Porter’s hypothesis is by promoting green technological innovation through environmental regulations. However, the research findings regarding this approach have generated controversy. On the one hand, When studying the relationship between environmental regulation, innovation, and competitiveness, the perspective of maximizing profits should always be regarded as the main driving force for corporate strategic decision-making. This viewpoint holds a central position in any business decision-making consideration. Therefore, it is often unrealistic to believe that compliance costs or underestimating profit maximization factors can be ignored when companies decide whether to comply with specific regulatory stimulus measures or take proactive measures in the environment [19]. It was found that environmental regulations based on taxes and fees significantly hinder technological innovation by enterprises. Environmental regulations lead to an increase in production costs for enterprises, limiting their ability to engage in technological innovation [20]. In contrast, some studies suggest that environmental regulation does indeed have the compensation effect for innovation proposed by Porter. Environmental regulations have a significant positive impact on green technology innovation in enterprises. Environmental regulations lead to increased taxes fees, forcing large and medium-sized enterprises to engage in green technology innovation to reduce energy loss [21]. The starting point of this study is not to deny the existence of compliance costs, but to emphasize that innovation compensation exceeds compliance costs.Wang [22] examines the impact of environmental regulation (pollution fees) on green technology innovation from the perspective of compliance costs. Intended to demonstrate that innovation compensation exceeds compliance costs. The above two perspectives both explore the spillover effects of green technology from a linear relationship perspective, but do not consider the impact of environmental regulation levels on the spillover effects of green technology. From the above two aspects, the effect of environmental constraints on green technology innovation is linear, which means that in green technology innovation, the effect of environmental constraints on green technology innovation is marginal and fixed. Although this assumption is theoretically valid, it overlooks the possibility of the marginal effect limit of environmental laws on green technology progress, thus proposing research on the appropriate intensity of green technology progress. Based on the background of heterogeneous industries, Shen [23] identified the optimal degree of industrial regulation in China through non-linear analysis of environmental regulation and efficiency.Li Ling and Tao Feng [24] found that in China’s medium and low pollution industries, the impact of environmental regulation on green TFP and the impact of technological innovation on technological efficiency exhibit a U-shaped curve, and their speed of breaking through the U-shaped inflection point is faster than that of productivity and technological innovation. Bing et al [25] pointed out that different levels of environmental regulations have different effects on corporate green innovation.Tang et al [26] believe that compared to the other two types of environmental regulations, ER1, as the initial stage of green industry development, plays a much greater role in promoting innovation in the green industry. ER2 and ER3 experience stable growth during the high-income period of the economy,while ER1 serves as the foundation. Environmental restrictions have an inverted U-shaped impact mechanism on GTI [27] or a U-shaped impact mechanism [28], according to other research findings.

In conclusion, the existing literature has examined the relationship between environmental regulation and Green Technological Innovation (GTI) from a multitude of perspectives. While this exploration has yielded valuable insights, the findings remain a subject of contention. In this research, it is claimed that the following are the primary causes of the dispute in existing studies: 1.The game between the "follow cost" effect and the "innovation compensation" effect determines the inhibitory or promoting effect of environmental constraints on green technology innovation in enterprises. This is related to the way and intensity of environmental regulation. Different countries are at different economic stages in the country study, which may lead to variance in the empirical study outcomes. In terms of empirical variable selection, there is variability in the selection of environmental rules and GTIs as proxies, resulting in inconsistencies in the results, which is essentially an under-robustness of empirical findings. The main reasons for this are the endogeneity of environmental and pollution control, the difficulty in accurately evaluating green technology innovation, and the nonlinear effects of environmental regulation. Additionally, existing literature has primarily conducted empirical analyses at the regional or industry level, with a focus on industrial, manufacturing, and heavily polluting industries. However, there is limited research addressing the industry as a whole from a macroeconomic perspective, which may result in a degree of generality in the findings. Secondly, existing studies have mostly explored the mechanism of heterogeneous environmental standards on green innovation from the perspectives of both internal and external enterprises. yet it has overlooked the intricate interactions between government and businesses.This technique may not accurately portray the true situation, particularly when considering government incentives in many nations to encourage the development of green technology by businesses. Thus, this paper delves into the impact of diverse environmental regulations on firm Green Technological Innovation (GTI) and dissects its mechanisms from both internal and external coupling perspectives, with China’s listed A-share companies as the subject of study.

3.1 Environmental regulation and enterprise GTI

The "innovation compensation" effect and the "follow cost" effect are important mechanisms of environmental regulation on green innovation behavior, and they are the mechanisms of environmental regulation on green innovation behavior. However, at the micro level, different environmental control methods have different impacts on corporate green innovation behavior through different operating mechanisms. In this section, the theoretical impact of ER1, ER2, and ER3 on firm GTI is examined.

3.2 The moderating role of enterprise R&D investment and government support

In the preceding article, various kinds of environmental regulations that may force or guide enterprises to engage in GTI activities to some extent were examined.But the specific microscopic mechanism is still unclear. On this basis, from the perspectives of enterprise R&D investment and government support, this paper explores the micro mechanism of environmental legislation on green technology innovation.This technique was chosen because financing is the most important issue for organizations pursuing GTI as a research and development activity.

The technological innovation at the enterprise level is mainly influenced by R&D management drivers. Research has shown that R&D investment is an effective way to improve the environmental technology innovation capability of enterprises [40]. In the context of government enforcement of environmental regulations, enterprises have two options: increasing overall R&D investment and adjusting the structure of R&D investment.First, environmental rules encourage companies to increase R&D spending in order to take sustainability into account, and this upswing in total R&D spending consistently correlates with a rise in green R&D investments. Secondly, when facing environmental regulations, even if R&D investment remains unchanged, companies will adjust their own investment to encourage them to increase investment in green technology innovation; Meanwhile, increasing investment in green research and development can also enhance the awareness of enterprise researchers towards green technological innovation [41]. In addition, by increasing research and development investment, enterprises can implement green and low-carbon development strategies, improve pollution control levels, and minimize the source of pollution. This not only reflects the company’s high social responsibility, but also enhances its brand image. This can also help alleviate economic development and environmental issues.

Having established environmental legislation, government subsidies, tax incentives and other measures may have a more unique benefit in reducing the problem of companies underinvesting in green R&D. Corporate GTI is known for its high investment and high risk, and risk averse enterprises may not be interested or participate in GTI at all. From this perspective, short-term environmental constraints will only encourage companies to use carbon emissions trading to reduce production scale, rather than carbon emissions reduction. Currently, government assistance can to some extent bridge the gap between social welfare and personal well-being, reduce market failures in the green industry, and enhance the enthusiasm of enterprises for the development of the green industry [42]. On the other hand, obtaining government funding is a message conveyed by external investors that these enterprises typically possess strong innovation capabilities and development potential.These businesses hold a government-recognized status, which makes it easier for them to earn the trust of external investors. As a result, they can broaden their sources of funding, reduce funding expenses, enhance their negotiation leverage, bolster the risk-reduction capacity within the green innovation supply chain, and ultimately improve their Green Technological Innovation (GTI) capabilities.

4.1 Variable selection

4.1.2 Independent variables.

On this basis Three factors, namely ER1, ER2, and ER3, were selected to characterize the intensity of environmental regulation. (1) ER1: This study used the number of environmental protection legislation in each province and city, including the number of administrative penalty cases accepted, the number of environmental supervisors, and the amount of funding for environmental protection facility construction [44]. (2) ER2: This is measured from four aspects: pollution tax (environmental tax), the proportion of environmental expenditure in local fiscal expenditure, the proportion of environmental infrastructure investment in regional GDP, and the proportion of pollution control investment in regional GDP [45,46] (3) ER3: This article selects variable factors for quantitative evaluation of ER3, such as credibility, frequency of environmental inspections, proposals from the Two Sessions, automatic monitoring of key pollution sources, and nature reserves. Table 1 showed the specific description of independent variables.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Specific description of independent variables.

https://doi.org/10.1371/journal.pone.0296008.t001

Standardize the individual indicators: (1)

Where, represents the standardized value of index j, UE_ij represents the initial value of index j in province i, Min(UE_j) and Max(UE_j) respectively represents the minimum and maximum value of index j in all province, i represent each province, i = 1,2, …,30; j represents the indicators of each environmental regulation.

In order to eliminate the differences between similar indicators in various regions, the adjustment coefficient is calculated as follows: (2)

Among them, W_j represents the adjustment coefficient of the j index, and UE_j represents the average value of the j index in the sample.

Calculate the comprehensive index of ER1, ER2 and ER3 according to the following formulas: (3)

ER_it represents the comprehensive index of environmental regulation in province i in period t.

4.2 Data sources

The research samples of this paper are taken from the data of China ’s A-share listed companies On the basis of collecting the original data, the original data is processed: (1) Remove ST (listed companies with two consecutive years of losses) and * ST (listed companies with three consecutive years of losses) enterprises. (2) Due to the lack of data, remove the samples of Hong Kong, Macao, Taiwan and Tibet. (3) Delete company samples with severely missing data, which cannot be obtained and affects the operation of the software. (4) If the company lacks a small amount of data,After being supplemented through public databases, there is no significant difference. survey questionnaire data, or speculation based on other variables. (5) Adding 1 to the green patent data, and taking the log. As a result, the total sample size was3930observations.From an econometric point of view, the sample size chosen for this paper has been able to be compelling in a certain statistical sense.

In order to ensure the adequacy of the collected data, our data comes from: (1) Matching the company name with the China Research Data Service Platform (CNRDS) to obtain green patent data. (2) The company ’s name is linked to the accounting research database(CSMAR) and CNINFO website (www.CNINFO.com.cn) to obtain relevant financial data. (3) ’ China Statistical Yearbook ’, ’ China Environment Statistical Yearbook ’, ’ China Science and Technology Statistical Yearbook ’, ’ China Industry Statistical Yearbook ’ and ’ China Energy Statistical Yearbook ’.

4.3 Model construction

1. (1) Benchmark regression model

The benchmark regression in this paper is to use enterprise GTI to regression ER1, ER2 and ER3 respectively. The detailed model is shown in formula (4).

(4)

Among them: GTI_it represents the green technological innovation level of the i company in the t period, which is divided into green invention patents (GTI1) and green utility model patents (GTI2); ER_it represents the environmental regulations of the province where the i company is located during the t period, As mentioned above, this variable actually contains three indicators, ER1, ER2 and ER3. For the purpose of simplification, the three variables are marked with the same variable symbol in formula (4); Con_it represents a set of control variables, μ_i represents an individual fixed effect, and ε_it represents a random interference item.

1. (2) Moderating effect model

In order to test the moderating effect of R&D investment and government support on the relationship between environmental regulation and corporate GTI, this paper constructs two moderating effect models for regression. The concrete model is shown as Eq (5) and Eq (6).

(5)

(6)

Among them: RD_it represents the level of R&D investment of enterprise i in period t; Sub_it represents the government support received by enterprise i in period t; other variables are explained as above.

5.1 Descriptive statistical analysis

Table 2 shows the descriptive statistics of the variables. These variables include GTI1, GTI2, ER1, ER2, ER3, RD, Sub, InSize, Den, Pro, BG, Age, FDI, ML, GDP, and Atm. For the baseline regression of this article, each variable had 3930 observations. From an average perspective, the average value of GTI1 was 1.5223, GTI2 was 1.2015, ER1 was 0.6415, ER2 was 0.8263, ER3 was 0.7415, RD was 3.8456, Sub was 1.3263, InSize was 8.1253, Den was 2.6123, Pro was 0.6552, BG was 0.2012, Age was 10.362, FDI was 0.0326, ML was 0.3115, the average GDP was 0.0003, and the average Atm was 0.5163. For these variables, the standard deviation ranged from 0.003 to 12.3625. The minimum and maximum values provided the range of values for each variable. For example, the minimum value of GTI1 was 0.0002 and the maximum value was 8.3623. The minimum value of ER1 was 0.0003 and the maximum value was 3.7485. The minimum value of RD was 0.0003 and the maximum value was 224.1256. In general, the explained variables and the environmental regulation intensity variables (ER1, ER2, ER3) had relatively wide ranges, reflecting differences in environmental regulation policies across provinces within China, and pointed to the need for region-based heterogeneity analysis in the paper that follows.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Variable descriptive statistics.

https://doi.org/10.1371/journal.pone.0296008.t002

5.2 Benchmark regression results

In this paper, a fixed-effect model has been chosen for regression analysis. It incorporates controls at both the individual and time dimensions, which can help alleviate endogeneity issues arising from individual characteristics and time trends to a certain extent. The benchmark regression results are shown in Table 3. The regression findings with green innovation patents as the explanatory variables are shown in columns (1) through (4). Columns (5) through (8) of the green utility model are interpreted variables. Columns (1) and (5) show that ER1 has a good impact on both green invention patents and green utility model patents, indicating that ER1 may help the firm’s GTI. Each unit increase in ER1 strength was associated with a 0.084 and 0.096 unit increase in the level of green invention and utility model patents, respectively.However, ER1 is not significant for green invention patents, while ER1 passes the 5% significance level test for green utility model patents.Based on the previously mentioned findings, ER1, the most widely adopted regulatory approach in China, has been observed to play a significant role in driving efforts for the development of Green Technological Innovation (GTI).(2) and (6) demonstrate that ER2 has distinct effects on the green invention patent and the green utility model patent. Although the regression results indicate that the coefficient of influence of ER2 on green invention patents is 0.236 and has a beneficial effect, this finding is not statistically significant.The green utility model patent has also been severely damaged by ER2. For every unit increase in ER2 intensity, the patents for green utility models decrease by 0.299 units and pass the 10% significance level The research results show that China’s current environmental protection system still has a market-oriented environmental protection model, which restricts green and low-carbon development. From columns 3 and 7, it can be seen that ER3 has a significant impact on both green and green utility models, although in the opposite direction. The study found that the impact of ER3 index on green innovation patents is significant at 5%, and for every 1 unit increase in ER3 index, it decreases by 0.078.On the other hand, ER3 greatly boosted the green utility model patent at the5% significance level, with each unit increase in ER3 strength increasing the patent on the green utility model by 0.245.Research has found that companies are more inclined to adopt green utility model patents rather than green invention patents when facing voluntary environmental regulations. Even when using green utility models for patent creation, there is still an issue of excessive reliance on green invention patent resources. Alternatively, this result could be due to the tendency of firms to engage in short-term regulatory behavior such as end management in the face of public and media scrutiny, which not only wastes firm resources but also crowd out investment in green invention patenting. Introduce three different environmental regulation intensity variables into the regression analysis at the same time, and provide them in columns 4 and 8, respectively. Under the combined effect of the three adjustment methods, the regression coefficients of all three adjustment methods are relatively large. Using the green invention patent model as the explanatory variable, ER1 is 0.035, ER2 is 0.458, and ER3 is 0.336, respectively, and all are significant at the 5% significance level.Using the green utility model patent model as the explanatory variable, the regression results for ER1, ER2, and ER3 were 0.066, 0.699, and 0.563, respectively, and were significant at the 10% significance level. Furthermore, with the exception of ER1, the regression findings of the other two environmental regulation approaches outperform the regression results of the model with simply themselves as explanatory variables.These findings indicate that the interaction of the three environmental policies is crucial to the green technology innovation of enterprises. Considering the impact of each environmental regulation approach on green invention patents and green utility model patents alone does not stand out, and considering their interplay at the same time allows for a better understanding and optimization of the effects of environmental regulation policies.This policy interaction can increase the quantity and quality of green invention patents and green utility model patents, thereby promoting environmental protection and the development of enterprise GTI.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Benchmark regression results of environmental regulations and GTI.

https://doi.org/10.1371/journal.pone.0296008.t003

In addition, this project will also study the impact of environmental control policies on green invention patents and green practical new technologies, pointing out that environmental control policies have a higher level of support for green invention patents and green practical new technologies. Faced with environmental regulation, enterprises tend to adopt green and low-carbon technologies with low technological content rather than innovative benefits, which is consistent with the reality of most Chinese enterprises, especially small and medium-sized enterprises.

5.3 Robustness test

Outliers are cleaned here using Winsorize processing. Based on this, the robustness test is run using one of three methods:

1. Replace the explanatory variable with the number of green patent applications as a proxy for GTI. The number of green patent applications is used to replace the original explanatory variables and check whether the results are consistent with the benchmark regression resultsin the robustness test. This can verify whether the impact of green patent applications on GTI is robust.
2. Remove control variables at the enterprise level. In the benchmark regression, there are control variables at the enterprise level, such as enterprise scale, capital structure and so on. In the robustness test, these enterprise-level control variables are removed from the model to verify whether the results are still robust.
3. Add new control variables. In this robustness test, considering that the differences in education and industrial structure in China ’s provinces may have an impact on the GTI of enterprises, these two factors are added as control variables in the test. The industrial structure coefficient (Str) is calculated by the practical Eq (7), and these two factors are added and introduced into the model to verify whether the results are still robust.

(7)

where represents the proportion of the added value of the i industry (i = 1, 2, 3) in year t to the total output value.

Regression results in the total number of patent applications. The regression coefficient of variable ER1 was 0.811, but not significant. The regression coefficient of variable ER3 was -5.112, which passed the 10% significance level test.

Regression results after removing control variables at the enterprise level. The variable ER1 had a positive impact on the number of green invention patents, with a regression coefficient of 0.042, passing the significance level test of 5%. The variable ER1 also had a positive impact on the number of green utility model patents, with a regression coefficient of 0.085, passing the 10% significance level test. The coefficient of ER3 for green invention patents was -0.012, and was significantly negative at the 5% level. The coefficient of ER3 for green utility model patents was -0.136, and was significantly negative at the 5% level.

Regarding the regression results while simultaneously increasing education levels and industrial structure. Attention was still paid to the number of green invention patents and green utility model patents. The variable ER1 had a positive impact on green utility model patents, with a regression coefficient of 0.045, passing the 10% significance level test. The coefficient of ER2 for green invention patents was 0.212 and was significantly positive at the 10% level. The coefficient of ER2 for green utility model patents was 0.143, and was significantly positive at the 5% level. The coefficient of ER3 for green utility model patents was -0.146, and was significantly negative at the 5% level.

Table 4 showed the robustness test regression results. The aforementioned test findings are essentially compatible with the benchmark regression results, showing that the study’s conclusions are sound.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Robustness test regression results.

https://doi.org/10.1371/journal.pone.0296008.t004

5.4 Heterogeneity analysis

This article intends to study the mechanism of environmental protection legislation on the innovation of green industry clusters from four perspectives: region, ownership, factor density, and industry. For emerging economies like China, the study of heterogeneity is particularly important. Governments of various countries can formulate specific environmental policies by studying these issues.

5.4.1 Heterogeneity of regions.

There are significant differences in the economic development level, resource endowment, industrial structure, and other aspects among various provinces and cities in China. If the study of actual policies cannot take into account the significant differences between regions, then this conclusion will become less important. This article takes 30 provinces in China as an example and draws on the experience of China’s economic development to divide them into three parts: East, Central, and West. The economic development level in eastern China is relatively high, with a high degree of marketization, a well-developed industrial structure, and a high level of economic development. Next is the central region with the lowest level of economic development.

Based on this, the impact of environmental restrictions in various locations on firm GTI is explored, and the findings are provided in Table 5. In the green invention patent model in the eastern region, the coefficient of ER1 was 0.053, which passed the 10% significance level test, indicating that ER1 had a positive promoting effect on green invention patents; The coefficient of ER2 was 0.374, which passed the significance level test of 5%, indicating that ER2 had a significant positive impact on green invention patents; The coefficient of ER3 was -0.263, which passed the significance level test of 5%, indicating that the impact of ER3 on green invention patents was significantly negative. In the green utility model patent model in the eastern region, the coefficient of ER1 was 0.085, which passed the significance level test of 5%; The coefficient of ER2 was 0.526, which passed the significance level test of 5%, indicating that ER2 had a significant positive impact on green utility model patents; The coefficient of ER3 was -0.412, indicating a significant negative impact of ER3 on green utility model patents.According to the regression results, environmental regulation has a considerable impact on the GTI of companies in the Eastern sample, with ER1 and ER2 both favoring the growth of green invention patents and green utility model patents. In terms of intensity, ER2 has a stronger effect than ER1, although ER3 inhibits GTI activity significantly. In the central region, only ER1 and ER3 had a significant impact on green utility model patents. The coefficient of impact of ER1 on green utility model patents was -0.245, indicating that the impact of ER1 on green utility model patents was negative at the 10% significant level. The coefficient of influence of ER3 on green utility model patents was 0.445, indicating that the impact of ER3 on green utility model patents was positive at a significant level of 5%. In the western region, ER1, ER2, and ER3 were not significantly affected by green invention patents and green utility model patents, Indicating that none of the three environmental regulatory methods had a substantial impact on GTI.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Regression results of heterogeneity of regions.

https://doi.org/10.1371/journal.pone.0296008.t005

Through the comparison of these three regions, it can be found that the strict environmental rules in the eastern region help to promote innovation. Under the dual role of government supervision and market incentives, enterprises can better adapt to the production mode and strive to achieve green development. However, due to the backward industrial structure in the central and western regions, its environmental regulatory policies make it difficult to promote the growth of corporate GTI. The industrial development of the central and western regions mainly depends on the extensive resource utilization model, and with the transfer of high pollution and high energy-consuming industries to the central and western regions, enterprises need to continue to strive to get rid of the extensive development model. Therefore, enterprises in the central and western regions may face insufficient GTI investment under the implementation of environmental regulation policies, which may even lead to the migration of enterprises from the central and western regions to regions with weak environmental regulation.

By studying the impact of environmental restrictions in different regions of China on corporate GTI, it reveals the differences in economic development and environmental management among provinces in China, and points out that the eastern region has advantages in green innovation, while the central and western regions face some challenges due to industrial structure problems. This is of great significance for formulating targeted regional policies and promoting sustainable development.

5.4.2 Heterogeneity of ownership.

State-owned enterprises and non-state enterprises collectively constitute a substantial portion of China’s national economy. There are noteworthy distinctions between them concerning their nature, primary responsibilities, and business objectives, which represent a crucial matter that necessitates attention in the study of China’s economy.State owned enterprises are the representatives of state-owned assets and the concentrated embodiment of national will. They not only aim to meet the profits of enterprises, but also demonstrate the government’s control over all aspects of economic and social development. However, private enterprises aim to maximize economic benefits. For the purposes of this paper, the sample firms are divided into state-owned and non-state-owned firms in order to analyze the impact of environmental restrictions on the GTI heterogeneity of firms with different ownership characteristics. The regression results are shown in Table 6. In state-owned enterprises, the coefficient of influence of ER2 on green invention patents is 0.275, and it is significantly positive at the 5% level. The impact coefficient of ER2 on green utility model patents is 0.144, and it is significantly positive at the 10% level. The impact coefficient of ER3 on green utility model patents is -0.299, and it is significantly negative at the 10% level. In non-state-owned enterprises, the coefficient of influence of ER1 on green utility model patents is 0.094, and it is significantly positive at the 10% level. The impact coefficient of ER3 on green utility model patents is -0.263, and it is significantly negative at the 10% level.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 6. Regression results of heterogeneity of ownership.

https://doi.org/10.1371/journal.pone.0296008.t006

From the regression findings, it can be deduced that: (1) ER1 strongly enhances the patent level of the green utility model in a sample of non-state owned enterprises, but has no effect on green invention patents.The research results indicate that under ER1 conditions, non-state-owned companies are more inclined to adopt green technologies to compensate for the losses caused by environmental pollution. Compared with state-owned enterprises, private companies have greater flexibility in adjusting their own development strategies and have a greater promoting effect on green technology innovation. In the process of adopting green technology innovation, private enterprises often encounter green practical technologies with lower technical difficulty. (2) ER2 has a promoting effect on the technological innovation of state-owned enterprises (including green inventions and green utility models), indicating the promoting effect of ER2 on the technological innovation of state-owned enterprises. (3) The ER3 effect has had a negative impact on both types of enterprises, and green utility model products have a strong inhibitory effect on their patents. It can be shown that green utility models, regardless of the ownership of the business, are more vulnerable to external public opinion than green inventions.

5.4.3 Heterogeneity of factor intensity.

When facing environmental regulations, enterprises with significant differences in factor density exhibit heterogeneity in their choices of green technology innovation. For example, relying on resource intensive companies for transformation is usually better than relying on labor-intensive companies. Based on Dong and Guo’s research [47]. The article categorizes listed companies in China into technology intensive, capital intensive, labor-intensive, and resource intensive. Then, the impact of environmental regulations on the differences in green innovation capabilities of four types of enterprises was studied. The results are shown in Table 7. For technology intensive enterprises, in the regression model of green utility model patents, the coefficient of the independent variable ER1 is 0.063, which passes the 5% significance level test, indicating a positive relationship between ER1 and green utility model patents. The coefficient of ER3 is -0.185, which passes the significance level test of 5%, indicating that the impact of ER3 on green utility model patents is significantly negative.For capital intensive enterprises, the coefficient of influence of ER1 on green invention patents is 0.052, which passes the significance level test of 5%, indicating that ER1 has a significant positive impact on green invention patents. The coefficient of influence of ER2 on green invention patents is 0.56, which passes the significance level test of 1%, indicating that ER2 has a significant positive impact on green invention patents. The coefficient of influence of ER1 on green utility model patents is 0.055, which passes the 1% significance level test, indicating a positive relationship between ER1 and green utility model patents.For labor-intensive enterprises, in the regression model of green invention patents, indicating a negative relationship between ER1 and ER3 and green invention patents, and a positive relationship between ER2 and green invention patents.For resource intensive enterprises, there is no significant relationship between ER1, ER2, ER3 and green invention patents or green utility model patents.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 7. Regression results of heterogeneity of factor intensity.

https://doi.org/10.1371/journal.pone.0296008.t007

Through empirical analysis, it was found that: (1) ER1 plays an important role in enhancing the green innovation capabilities of high-tech and capital intensive enterprises, while it has a significant restraining effect on the improvement of the green innovation capabilities of labor-intensive enterprises. One of the reasons for this is that both technology and capital intensive enterprises have higher technological innovation and adaptability. When facing mandatory environmental control policies, they can quickly determine the technological innovation priority of green industry clusters, thereby gaining technological competitive advantages. In the production process, labor-intensive enterprises have low technological innovation capabilities and environmental awareness, and their willingness to purchase green machinery products is higher than that of green manufacturing enterprises. (2) ER2 has a significant promoting effect on green innovation in capital intensive and labor-intensive enterprises. The promotion effect of ER2 on capital intensive enterprises is similar to that of ER1. In capital intensive enterprises, in order to avoid taxes, their ability to seek flexible technological innovation will encourage them to engage in green technological innovation. Compared to ER1, ER2 bears much less pressure on labor-intensive enterprises and can provide a buffer time for such enterprises to some extent, thereby promoting the development of green technology innovation.(3) Research has found that in China, especially labor-intensive enterprises, they often adopt short-term terminal governance strategies when facing public opinion supervision. Overall, the activity of ER2 is continuously increasing, while the activity of ER3 is mainly inhibited, indicating that the environmental awareness of the Chinese people needs to be strengthened. The three environmental laws and regulations have little effect on promoting the green development of resource-based enterprises.One probable explanation is that the production activities of resource-intensive enterprises rely mostly on natural resources.In addition, environmental rules had minimal impact on the GTI due to the poor innovation base and the relative difficulty of modifying the production process.

6.1 The moderating effect of enterprise R&D investment

The model (5) developed in Section 4.3 is used to run the regression on the moderating effect of corporate R&D spending, and the results are shown in Table 8.The basic meaning of each column is consistent with the previous text. From the test results of three different types of environmental regulations, it can be seen from Table 8 that enterprise R&D investment positively regulates the relationship between ER2 and green inventions at a significance level of 5%, and positively regulates the relationship between ER2 and green utility models at a significance level of 5%.The regression results demonstrate that R&D spending considerably moderates the connection between ER2 and Green innovation patents and ER2 and Green utility model patents. These findings suggest that ER2 empowers enterprises to make their own judgements when it comes to internalising the cost of pollution. Companies have implemented GTI during this period by increasing investment in research and development, thereby improving resource usage, reducing energy consumption and reducing pollution control expenses.The possible reason is that market-oriented environmental regulations use emission permits, emission trading, and other economic tools; Establish pollutant emission prices or taxes, and encourage enterprises to adopt environmental protection behaviors. For some manufacturing enterprises, in order to reduce pollution emissions, they increase research and development investment, develop new technologies or improve existing technologies to reduce environmental impacts during the production process. Environmental regulations typically encourage companies to invest in green technology innovation to comply with these regulations and protect their market position. At a time when environmental protection and sustainable development have become key global issues, the relationship between environmental regulation (ER) and green technology innovation (GTI) in enterprises has become particularly important. R&D investment is often seen as a key factor driving technological innovation, while also having a moderating effect on the relationship between environmental regulations and corporate GTI. Therefore, the increase in R&D investment by enterprises has a positive regulatory effect on invention patents. In other words, increased R&D investment can not only directly improve the quantity or quality of green technology innovation, but also enhance the promoting effect of environmental regulations on GTI.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 8. Regression results of the moderating effect of corporate R&D investment.

https://doi.org/10.1371/journal.pone.0296008.t008

Karmaker [48] used a globally robust model with significant statistical capabilities to study the causal relationship between environmental taxes and environmental related technological innovation. This model uses panel cointegration analysis considering cross-sectional dependence to conduct a quantitative study on 42 high-income and middle-income countries from 1995 to 2018, and finds that the implementation of environmental taxes has stimulated technological innovation. Levying environmental taxes can help accelerate the reduction of carbon emissions and promote the development of green innovative technologies in high-income and middle-income countries. Meanwhile, Shang et al [31], who share the same view, conducted an empirical test on the impact of environmental tax collection on corporate technological innovation. Research has found that levying environmental taxes can stimulate the innovation drive of enterprises and significantly increase the input and output of technological innovation. By taxing pollution emissions from enterprises, they will be forced to think about how to reduce environmental pollution in the production process, which can guide them to invest more in green innovation. This will encourage companies to invest more resources in green innovation. Not only can it reduce taxes and fees generated by pollution emissions, but it can also promote green technology innovation in enterprises.

At the same time,the relationship between R&D investment on environmental regulations and green innovation patents can be adjusted in three ways. (1) Increase government supervision: strengthen the supervision of enterprises to comply with environmental regulations, ensure that enterprisesare in accordance with the provisions of R&D investment, and truly achieve green innovation. This can be achieved by increasing law enforcement and increasing penalties. (2) Provide incentive mechanism: encourage enterprises to increase R&D investment, through incentives and incentives to promote enterprises to actively carry out green innovation. For example, the government can set up a green innovation fund to provide financial support for enterprises that comply with environmental regulations and carry out green innovation. Third, strengthen information disclosure and public supervision: strengthen the information disclosure of corporate R&D investment and green innovation, so that the public can understand the behavior of enterprises, and promote enterprises to better fulfil their environmental responsibilities through public opinion supervision.

These results suggest that firms that actively promote substantial green change or innovation by expanding their R&D investments may be better positioned to meet compliance requirements under government-mandated environmental regulations.

6.2 The moderating effect of government support

The regression on the moderating effect of government support was run using the model (6) developed in Section 4.3, and the results are reported in Table 9.On this basis, this study found that under the interaction between environmental regulation and environmental regulation, the interaction between environmental regulation and environmental regulation is both positive, but not significant. The results of this study indicate that government support may have a partial impact on the correlation between environmental regulations and corporate green invention patents, although the magnitude of its positive regulatory effect is still unclear.Regarding the regression results, government support exhibits a significant negative moderating effect on the relationship between ER1 and firms’ green utility model patents, ER3 and enterprise green utility model patents. Simultaneously, government support significantly positively moderates the relationship between ER2 and enterprise green utility model patents.The empirical findings above suggest that, when accompanied by government support, environmental constraints might encourage firms to invest in green invention patents.For Chinese enterprises, government subsidies or financial support for their technological innovation generally play a significant role, especially for companies that can effectively improve green environmental protection technologies. This study will help explain the positive regulation of environmental regulations by the government, as well as the correlation between the feedback effect of manufacturers on environmental concept patents. On the other hand, green utility model patents indicate that companies have made a short-term technological investment in order to passively comply with regulations, and their nature is closer to the "compliance cost". Therefore, when the government subsidizes "green innovation" in enterprises, compliance pressure on businesses is alleviated, granting them more buffer space to address the pressures imposed by environmental regulations. Through this approach, the research and development efforts of green utility models can be reduced, while also increasing the development scale of green invention patents.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 9. Regression results of the moderating effect of government support.

https://doi.org/10.1371/journal.pone.0296008.t009

6.3 Transformation mechanism of innovation achievements

According to Porter’s theoretical hypothesis, environmental regulation has a certain impact on a company’s technological innovation, with the goal of improving the company’s production efficiency. Taking into account all the factors discussed above,it is natural to reach the conclusion that whether production efficiency can ultimately be improved is a critical criterion for enterprises seeking green technological innovation.Therefore, the purpose of this article is to explore the underlying theoretical logic of whether environmental regulations can promote the improvement of production efficiency and incentivize enterprises to implement green technology innovations from the perspectives of environmental regulation, production efficiency enhancement, and green technology innovation. Specifically, the impact of environmental restrictions on overall green factor productivity is investigated using empirical evidence. The first step is to construct the green productivity index and refer back to the methods of Wang and Zhang [49] and Kuang and Peng [50]. Among them, labor input (measured by the average number of employees in each province), capital investment (measured by net fixed assets), energy input (measured by converting regional total energy consumption into standard coal), and three output indices (measured by value added) and employment (measured by the actual number of employees in each province) are included (measured by the annual number of employees). Using the data envelopment analysis method, the total green production efficiency of each province was calculated. Finally, regression analysis was conducted on green TFP using environmental regulation methods. Table 10 showed the relationship between environmental regulations and green total factor productivity. Empirical analysis shows that ER2 has a significant driving effect on the improvement of total factor productivity in enterprises. Although both ER1 and ER3 have a significant positive relationship with green production efficiency, this result indicates that environmental regulation can improve green production efficiency by enhancing the productivity of enterprises. Another interpretation of this result is that while environmental regulations are conducive to an increase in green patents, a lack of green innovation output transformation may be an important underlying reason for inhibiting the implementation of green technology innovation by businesses, as reflected in ER1 and ER3.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 10. The relationship between environmental regulations and green total factor productivity.

https://doi.org/10.1371/journal.pone.0296008.t010

6.4 The temporal effects of environmental regulatory policies

When evaluating the exogenous effects of environmental regulation, the first thing to consider is its impact on policies, while also examining the speed of its impact. This section will explore the short-term and long-term effects of different environmental control measures on the company’s green technology innovation. The article takes the period of China’s Five Year Plan as an example to analyze five lagged variables. Table 11 shows the results of the regression analysis. Observe through regression coefficients.It can be determined that the influence of environmental regulations on the GTI of businesses has been decreasing over timeSpecifically, there was no statistically significant difference in the regression analysis of ER1 lag values, and in the second lag stage, there was a negative effect related to the forced ER1 trait. After adopting ER1, the company tends to quickly adjust its production process to avoid punishment, but this process will gradually weaken. The ER2 in the first and third stages has a significant promoting effect on the enthusiasm of companies for green investment, but in the fifth stage, it has a negative promoting effect on corporate green investment. Therefore, the role of ER2 in the company’s green technology innovation is likely to be only a short-term effect. ER3 has a negative effect on enterprise GTI, but under various time delayed variables, this effect gradually decreases over time. Therefore, It is speculated that the role of ER3 in promoting green technology innovation in enterprises still needs further long-term research, and the driving effect of ER3 on green technology innovation in enterprises will continue for a longer time. Overall, China’s environmental regulatory policies have a more short-term impact on the green development of enterprises. ER1 has the most rapid impact on enterprise GTI,while ER3 may play a more favorable role over a longer period of time.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 11. Results of the lag period of heterogeneous environmental regulations.

https://doi.org/10.1371/journal.pone.0296008.t011

8.1 Policy recommendations

Our research has specific guiding significance in various aspects, as shown below;

For decision makers, decision makers can further promote corporate GTI by strengthening the formulation and implementation of environmental regulatory policies. At the same time, decision makers can also formulate targeted environmental regulatory policies according to the heterogeneity of region, ownership structure and factor concentration, so as to better promote the development of GTI.

For enterprises, enterprises can add green environmental protection technology to the new strategy, increase R&D investment, and carry out economic transformation. At the same time, policy adjustments can be made according to heterogeneity, and long-term development strategies can be formulated to better adapt to the changes brought about by environmental regulatory policies.

For stakeholders, environmental protection organizations, consumers, etc., can supervise the environmental situation of enterprises, so as to improve the implementation of corporate environmental policies. This not only promotes the green development of enterprises, but also improves the sustainable development and environmental quality of the whole society.

8.2 Future resarch directions

Future research can be conducted in the following directions, as shown below;

1. Promoting the application of green innovation technology: Encourage and support enterprises to carry out green innovation technology and promote sustainable development. By employing measures such as government financial support and tax incentives, intensifying support for the development and application of green innovation technologies can, in turn, stimulate businesses to embrace technologies that are not only ecologically friendly but also highly efficient.
2. Economic incentives: Establish a sound economic incentive mechanism, such as environmental taxes, emissions trading, etc., to guide enterprises to reduce pollution and save resources, so that enterprises are economically aware of the impact of the environment, so as to take environmental protection measures.
3. Guiding public participation: strengthen the transparency of corporate information and encourage public participation in environmental decision-making and supervision. By instituting a comprehensive and transparent environmental data platform, the general populace gains access to insights into corporate environmental performance, along with offering channels and commensurate incentives for reporting, thereby fostering augment societal oversight and corporate accountability.
4. In line with international standards: Actively participate in international environmental regulation activities, share and learn experience and technology with other countries and regions, and jointly respond to global environmental challenges. At the same time, it should be in line with international standards, improve the standardization level of domestic environmental regulation, and promote the competitiveness of enterprises in the global market.

The research findings of this article will provide decision-making basis for the improvement of China’s environmental policies and the sustainable development of Chinese enterprises. This article is based on empirical research on China as a developing country and provides some useful suggestions for China’s environmental regulation practices. Since the reform and opening up, China has shifted from a planned economy to a market economy, and the market mechanism has played a good regulatory role in the Chinese economy. For other developing countries, when implementing market-based environmental regulation systems, they should adjust their enforcement concepts in a timely manner based on the actual operation of their own market economy, in order to avoid the abuse of market control tools due to imperfect market mechanisms and poor policy effects.