The Effect of Environmental Regulation on Corporate Environmental Governance Behavior and Its Mechanisms

Abstract

Environmental regulation is an institutional guarantee for achieving green and sustainable economic development, and the implementation effect of environmental regulation policies is a concern for all sectors of society. This paper empirically examines the impact of environmental regulation on firms’ environmental governance behavior using a double difference model with the help of a quasi-natural experiment, the new Environmental Protection Law, and analyzes the mediating effect of central environmental protection inspectors using data from listed companies in China from 2011 to 2020. It was found that environmental regulation significantly enhances firms’ motivation to participate in environmental governance and central environmental protection inspectors play a mediating role in the impact of environmental regulation on firms’ environmental governance behavior. Furthermore, heterogeneity analysis shows that the enhancement effect of environmental regulation on firms’ environmental governance behavior mainly exists in large-scale firms and nonstate enterprises.

1. Introduction

Since the reform and opening up, China has achieved world-renowned levels of economic development. At the same time, the environmental pollution problem brought about by rapid economic development has become increasingly serious. This threatens the general public’s safety and well-being and severely challenges the sustainable and healthy development of China’s economy [1,2]. Such a high environmental cost has triggered people’s profound reflection on the urgency of environmental protection legislation and related institutional systems. At the 19th National Congress of the Communist Party of China, General Secretary Xi Jinping stated that the country should actively participate in global environmental governance and focus on solving its environmental problems. All sectors of society are actively building an environmental governance system led by the government, with enterprises as the main stakeholders and the participation of social organizations and the public. The coordinated development of the natural environment, the economy, and society has become a social consensus, and the concept of “building a beautiful China” is now deeply rooted in the hearts of the people. However, enterprise investment in environmental protection is often less than revenue-generating functions, resulting in a low level of participation in environmental governance. Therefore, persuading enterprises to actively participate in environmental governance by strengthening the legal system has become an important issue facing the current government. On 1 January 2015, the revised “Environmental Protection Law of the People’s Republic of China” (hereinafter referred to as the new Environmental Protection Law) was officially launched, which clarified in detail the punishment for polluting enterprises and the environmental supervision responsibility of local governments at all levels, exerting unprecedented pressure on enterprises and governments. However, due to the short-sighted behavior of some enterprises and local governments, the enthusiasm for participating in environmental governance still falls short due to short-term interests. As the main object of environmental regulation, the response of enterprises to the new Environmental Protection Law directly affects the implementation effect of the policy. Therefore, in July 2015, the Central Leading Group for Comprehensively Deepening Reforms passed the “Environmental Protection Supervision Program (Trial)”, which established the environmental protection supervision work system, emphasized the strict implementation of the main responsibility of environmental protection, and clarified the responsibility of leadership and supervision. The environmental supervision system has become an important part of China’s environmental management transformation and system construction, which has a profound impact on the environmental governance behavior of enterprises. This paper selected the data of Shanghai and Shenzhen A-share listed companies from 2011 to 2020 as the research object. Under the policy background of the implementation of the new “Protection Law”, this paper studied the impact of environmental regulation on corporate environmental governance behavior, introduced environmental protection inspectors, and studied its role in determining whether there is an intermediary effect that can provide a reference for China to achieve sustainable and healthy economic development. The marginal contribution of this paper includes the following aspects. First, most of the existing studies use proxy variables to analyze the implementation effect of environmental regulation policies. This paper used the new Environmental Protection Law, an exogenous event, using a double difference model to construct dummy variables to investigate its impact on corporate environmental governance behavior; after using propensity score matching, lagged explanatory variables, and other methods, the experimental results were still robust, effectively solving the endogeneity problem. Second, for the first time, environmental regulation, corporate environmental governance behavior, and environmental protection supervision were placed in the same analytical framework, which enriched the relevant research on the implementation effect of environmental regulation policies and the influencing factors of corporate environmental governance behavior. Third, it further examined the impact of environmental regulation on different types of enterprises and provided a policy basis for adopting environmental protection measures of different strengths.

2. Institutional Background and Literature Review

Since the reform and opening up, the legal system of environmental protection has been continuously improved in promoting the green and healthy development of China’s economy [3]. The old Environmental Protection Law was first passed and implemented in December 1989. However, with the continuous development of the economy and society, people’s environmental protection concepts have been continuously strengthened, and environmental problems have become increasingly prominent [4]. At the same time, the limitations of the old Environmental Protection Law became increasingly apparent; it can no longer meet China’s requirements for environmental protection in the current situation. Therefore, environmental protection should continue to be strengthened. The formulation of the new Environmental Protection Law was the most urgent need for environmental protection in China at that time. The revision process of the Environmental Protection Law was launched in 2011 and has since been revised four times. First, the initial review in August 2012 received a positive response from the general public. As a sign of attention, the new drafting work was taken over by the National People’s Congress Law Committee, which is more concerned about environmental interests. Second, the attitude toward strengthening environmental protection legislation in the second review draft in June 2013 clearly showed a positive trend. Then, in the third trial in October 2013, the “Amendment” case of the Environmental Protection Law was changed to an “Amendment” case, which changed the way of thinking about the revised Environmental Protection Law and showed a strong determination to conduct a comprehensive and thorough revision of the Environmental Protection Law. Finally, in April 2014, the revision of the Environmental Protection Law was reviewed for the fourth time. On the 24th of that month, it was voted and passed by the Eighth Session of the Standing Committee of the Twelfth National People’s Congress and was officially implemented into law on 1 January 2015. Compared with the old version, the new Environmental Protection Law strengthens the environmental protection awareness of various economic entities, clarifies the legal responsibilities of different economic entities in terms of environmental protection supervision and environmental protection prevention and control, and increases information disclosure and environmental protection. For regulations on public participation, the new Environmental Protection Law mainly has the following three characteristics. First, it highlights government responsibility, supervision, and legal responsibility. At the same time, the new law stipulates that local people’s governments at all levels should be responsible for the environmental quality of their administrative regions, increase the environmental protection target responsibility system and the assessment and evaluation system, and stipulate that higher-level governments and competent departments have supervisory responsibility for the work of lower-level departments or staff. Second, it increases the punishment for illegal polluting enterprises. Environmental supervision departments can directly seal up, seize and dispose of enterprises, continuously punish illegal polluting enterprises that do not make corrections in a timely manner, and detain the persons directly responsible for violating laws and regulations, such as evading supervision, which directly increases the cost of noncompliance. Third, it emphasizes information disclosure and the participation of the public and the media. It clearly stipulates that the public has the right to know, participate and supervise environmental protection work. The government shall disclose environmental information to the public in a timely manner, making it convenient for the public and the media to participate in and supervise environmental protection.

How effective is China’s environmental protection system in its improvements and implementation? From the analysis of the literature, scholars at home and abroad usually analyze the effect of environmental regulation from the perspectives of corporate behavior and performance. For example, Wayne and Ronald, from the perspective of corporate behavior, found that the intensity of environmental regulation affects the investment and financing decisions of enterprises [5]. Marc et al. found that when enterprises face highly demanding environmental regulations, they significantly increase the scale of emission reduction investment [6]. The empirical results of Jia and Chen [7] show that the central environmental protection inspection system has played a significant role in the improvement of environmental quality. Hu and Yang [8] found that the implementation of the Ambient Air Quality Standards prompted local governments to pay more attention to environmental governance behaviors, thereby enabling companies with government experience to increase their investment strength. Pan et al. [9] investigated corporate environmental performance from the perspective of environmental protection interviews. Tang et al. [10] found that in the context of the implementation of the “two control zones” policy, environmental regulation has a significant role in promoting urban economic growth, and enterprise productivity plays a role in this impact.

Judging from the abovementioned institutional changes and existing research, environmental regulation, as an external governance condition, can effectively promote the adjustment of the investment structure of enterprises, increase investment in environmental protection, and promote the improvement of production efficiency and growth of the urban economy [11,12,13,14]. The implementation of the central environmental protection supervision system can ensure the improvement of environmental governance effects and effectively improve environmental quality. However, few studies place the three within the same analytical framework. Therefore, this paper took the implementation of environmental regulation policies as a quasi-natural experiment, analyzed its impact on environmental governance behavior, and studied the influencing mechanism of environmental protection inspectors, providing reference value for adhering to the way of thinking that economic development and ecological, environmental protection support each other, integrate and promote each other, better serve the public’s expectation of building a “beautiful China” in accordance with the law, and realize the grand blueprint of “lucid waters and lush mountains”.

3. Theoretical Analysis and Research Assumptions

3.1. Environmental Regulation and Corporate Environmental Governance Behavior

This paper analyzed the application of environmental regulation to corporate environmental governance behavior from the perspective of compliance motivation and competitive advantage motivation. In terms of compliance motivation, the new Environmental Protection Law increased the punishment of illegal polluting enterprises and local governments, which may lead to losses as a result of environmental violations being greater than the cost of environmental governance. Enterprises are more likely to actively participate in environmental governance to avoid punishment. At the same time, the current situation of China’s ecological environment is dire, the people’s awareness of environmental protection has been continuously strengthened, and participation in environmental governance has increasingly become an important manifestation of compliance with social norms. More investors pay attention not only to the economic benefits of enterprises but also to the social and environmental benefits generated by corporate behavior [15]. Therefore, to gain the favor of investors, companies are willing to actively participate in environmental governance. In addition, the new Environmental Protection Law not only intensifies the government’s environmental supervision and punishment of illegal enterprises but also emphasizes important issues such as information disclosure and public participation. Once a company has environmental pollution problems, negative public opinion affects investors’ risk expectations for the company, resulting in reduced investment or demands for higher returns [16], thereby reducing market value. In order to avoid this situation, companies strive to participate in environmental governance and control pollution emissions to a compliant level.

The analysis is also based on the perspective of competitive advantage motivation. The long-term, high-cost, and high-risk characteristics of environmental protection investment usually aggravate the degree of information asymmetry between enterprises and investors, resulting in financing difficulties. The new Environmental Protection Law strengthens government supervision and corporate environmental information disclosure, which is conducive to increasing information transparency, easing financing constraints, and helping companies obtain economic resources. Enterprises that further reduce emissions on the basis of environmental compliance are more likely to obtain financial, tax, price, and government procurement incentives, which help increase corporate cash flow, reduce possible irreversible losses and marginal costs of environmental protection investment and improve corporate environmental governance. According to the theory of resource dependence, if an enterprise wants to achieve long-term sustainable development, in addition to its own internal conditions, it also requires political resources to assist in understanding government policy and adjust its business strategy in a timely manner. The new Environmental Protection Law stipulates that meeting environmental protection goals is an important element in the evaluation of government officials. The new Environmental Protection Law has increased the weight of environmental performance in the performance assessment of enterprises by local governments. To establish a good relationship with local governments, enterprises naturally actively respond to government policies and actively engage in environmental governance.

In summary, this paper took listed companies from 2011 to 2020 as a sample to empirically analyze the impact of the implementation of the new Environmental Protection Law on corporate environmental governance behavior. Therefore, Hypothesis 1 of this paper was proposed.

Hypothesis 1.

Environmental regulation can effectively improve the participation and investment of enterprises in environmental governance.

3.2. Based on the Mediating Effect of Environmental Inspectors

In order to further strengthen the implementation effect of the new Environmental Protection Law and promote the transformation and upgrading of the manufacturing industry, on 1 July 2015, the central government passed the “Environmental Protection Inspection Plan (Trial).” Following a pilot inspection conducted in Hebei in January 2016, large-scale supervision work was carried out in batches, covering a total of 31 provinces across the country. The specific process is shown in

Table 1. Implementation of the first round of the central environmental protection supervision system.

Batch	Inspector Location	Inspector Time
Pilot	Hebei province	December 2015 to February 2016
First batch	Inner Mongolia, Heilongjiang, Jiangsu, Jiangxi, Henan, Guangxi, Yunnan, Ningxia 8 provinces (autonomous regions)	July 2016 to August 2016
Second batch	Beijing, Shanghai, Hubei, Guangdong, Chongqing, Shaanxi, Gansu 7 provinces (municipalities directly under the Central Government)	November 2016 to December 2016
Third batch	Tianjin, Shanxi, Liaoning, Anhui, Fujian, Hunan, Guizhou 7 provinces (municipalities directly under the Central Government)	April 2017 to May 2017
Fourth batch	Jilin, Zhejiang, Shandong, Hainan, Sichuan, Tibet, Qinghai, Xinjiang 8 provinces (autonomous regions)	August 2017 to September 2017

. The introduction of the environmental protection inspection system is necessary to strengthen environmental governance. On the one hand, under the original system, the local environmental protection department accepted the leadership of the superior department and the local government at the same level, distorting the information in the process of transmission of environmental protection policy information from top to bottom; this resulted in the central government not supervising the implementation of environmental policies by local governments. On the other hand, implementing some environmental protection policies involved multiple departments simultaneously; however, the powers, responsibilities, and boundaries between departments were unclear, resulting in low policy implementation efficiency [17]. In order to solve these problems, an environmental protection inspection team led by the Ministry of Environmental Protection of the Central Committee and representing the authority of the Party Central Committee and the State Council was established. These inspection teams were assigned to different parts of the country to conduct special inspections. As a result, local governments were under the dual pressure of environmental protection performance and political responsibility, changing the relatively weak position of environmental protection departments and effectively solving the “failure” of environmental governance [18,19]. The central government attaches great importance to the effect of environmental regulation and continues to increase the intensity of environmental regulation. How will enterprises respond to these policies? Such critical questions need scientific answers. Therefore, this study used environmental supervision as an intermediary variable to analyze the impact of environmental regulation on corporate environmental governance behavior. Based on the above analysis, this paper proposed the second hypothesis.

Hypothesis 2.

Environmental regulation effectively promotes favorable corporate environmental governance behaviors through environmental inspections.

3.3. Firm Size and Environmental Regulation

The larger the scale of the enterprise, the more production resources it has at its disposal, and it has strong economic strength. Such a large-scale and powerful enterprise has the ability to bear the environmental cost of complying with environmental regulations and even increase environmental protection investment to achieve a stronger competitive advantage in the same industry. At the same time, larger-scale enterprises often emit more pollutants, which play a decisive role in the environmental quality of the region. Environmental regulation puts local governments under strong political pressure and tends to urge heavily polluting enterprises to standardize their production through political and economical means such as taxation, project approval restrictions, water and power outages, and avoid large-scale expansion that produces more pollutants. Therefore, in order to avoid the adverse impact of substandard environmental standards on some business activities of large enterprises, the introduction of the new “Environmental Protection Law” policy promotes the enthusiasm of large-scale enterprises to participate in environmental governance. Based on the above analysis, this paper proposes hypothesis 3.

Hypothesis 3.

Compared with small-scale enterprises, environmental regulation has a more significant effect on the promotion of environmental governance behavior of large-scale enterprises.

3.4. Ownership Types and Environmental Regulation

In China, the environmental responsibilities of state-owned enterprises and non-state-owned enterprises are quite different. On the one hand, in terms of environmental protection and environmental governance, the public has higher expectations for state-owned enterprises. The pressure of corporate environmental legitimacy far exceeds that of private enterprises; on the other hand, due to the natural relationship between state-owned enterprises and the government, when the government attaches importance to environmental governance, it may give government subsidies to state-owned enterprises. Therefore, public expectations and government intervention in making SOEs fulfill their environmental governance responsibilities have become the norm, so there will be no significant changes due to the impact of exogenous events in the new Environmental Protection Law. On the contrary, private enterprises are quite different, and their social responsibilities depend to a large extent on external pressure. Under the background of the new Environmental Protection Law, private enterprises are more motivated to send favorable signals through environmental governance to seek market support. Therefore, under the influence of public pressure, private enterprises perform better in corporate environmental governance. Based on the above analysis, this paper proposes hypothesis 4.

Hypothesis 4.

Compared with state-owned enterprises, environmental regulation has a more significant effect on the promotion of environmental governance behavior of non-state-owned enterprises.

4. Research Design

4.1. Sample Selection and Data Sources

In order to test the above assumptions, this paper selected A-share listed companies in the Shanghai and Shenzhen stock exchanges from 2011 to 2020 as the research object. Before using the data, we processed the data as follows: removed financial listed companies; listed companies with missing major financial variables, data anomalies, and missing listed companies; and all non-dummy variable values were rounded by 1% to exclude the influence of extreme values. The initial sample data came from the Guotaian database, and 1054 observations were obtained after final processing, with a time span of 10 years.

4.2. Variable Definition and Interpretation

4.2.1. Explained Variables

Enterprise environmental governance behavior ($Epi$) draws on the practices of Wang [20], Lin et al. [21], considering the influence of different enterprise scales, using the total assets at the end of the research period. The current environmental protection investment amount after reduction was used to measure the corporate environmental governance behavior indicators. At the same time, in the robustness test, the company’s environmental governance behavior was measured by the relative number obtained by dividing the company’s new environmental protection investment in the current year by the total assets at the end of the period multiplied by 100.

4.2.2. Explanatory Variables

$Trea{t}_i*Tim{e}_t$ is the core explanatory variable of this paper, in which $Time$ is the year dummy variable, and the year of the sample company is 2015 and later, the value is 1; otherwise, it is 0. $Treat$ is the policy dummy variable when the sample company belongs to the heavyweight. When polluting the enterprise, the value is 1; otherwise, it is 0.

4.2.3. Mechanism Variables

The environmental inspection dummy variable ($Cepi$) was used to measure the policy effects at the government and enterprise levels before and after the establishment of the central environmental inspection team. The inspected province takes a value of 0 before being inspected, 1 after being inspected, and 0 for provinces that are not inspected.

4.2.4. Control Variables

This paper drew on the factors in the literature that may affect the environmental governance behavior of enterprises and selects the following as control variables: agency cost (Agencost), enterprise size (Size), growth (Growth), asset-liability ratio (Lev), enterprise age (Age), financing constraints (Credit), profitability (Roa), capital intensity (Tangibility), ownership concentration (Top1), and the nature of enterprise ownership (Soe). The specific definition of each variable is shown in

Table 2. Definition and measurement of variables.

Variable Symbol	Variable Name	Variable Description
$Epi$	corporate Environmental Governance Behavior	The amount of environmental protection investment divided by the total assets at the end of the period
$Treat$	whether it is a heavy polluting enterprise	0–1 dummy variable
Size	enterprise size	Ln (total assets at the end of the period)
Lev	assets and liabilities	Total liabilities at the end of the period/total assets at the end of the period
Age	business age	Sample Year–Business Registration Year
Roa	profitability	Net profit/Total assets
Agencost	agency cost	Administrative expenses divided by main business income
Cflow	cash flow from operating activities	Net cash flow from operating activities/total assets at the end of the period
Tangibility	capital intensity	Average annual balance of net fixed assets (thousand yuan)/total assets
Top1	ownership concentration	Shareholding ratio of the largest shareholder
Growth	growth	operating income growth rate
Soe	the nature of business ownership	0–1 dummy variable
Cepi	whether it belongs to the inspected province	0–1 dummy variable

, and the descriptive statistics of the total sample are shown in Table 2.

4.3. Model Building

The difference-in-differences model is often used to evaluate the implementation effect of a policy. The model is based on natural experiments. Its core idea is to compare the differences between the experimental group and the control group before and after the implementation of the policy. In order to establish whether the policy achieved the desired effect, the estimator of the difference-in-differences model was obtained by differencing twice. If the policy implementation has a significant effect, the coefficients of the difference-in-differences estimator will be statistically significantly different from zero. Therefore, this paper constructed a control group and an experimental group based on the differences in the implementation of environmental regulation policies and used the difference-in-difference model to test the impact of environmental regulation on corporate environmental governance behavior.

$$Ep{i}_{it}={\alpha }_0+{\alpha }_{1}Trea{t}_i*Tim{e}_t+{\alpha }_{2}Control{s}_{it}+{\epsilon }_{it}$$

(1)

To test for the existence of a mediating effect, construct the following model:

$$Ep{i}_{it}={\gamma }_0+{\gamma }_{1}Trea{t}_i*Tim{e}_t*Cep{i}_{it}+{\gamma }_{2}Trea{t}_i*Tim{e}_t+{\gamma }_{3}Cep{i}_{it}+{\gamma }_{4}Control{s}_{it}+{\epsilon }_{it}$$

(2)

In the above formula, $Ep{i}_{it}$ reflects the environmental governance behavior of sample company $i$ in observation period $t$, ${\alpha }_0$ is a constant term, and the cross term $Trea{t}_i*Tim{e}_t$ is the core explanatory variable of this paper. The average difference before and after policy implementation. If ${\alpha }_1$ > 0 after the implementation of the new Environmental Protection Law, the environmental governance behavior of heavily polluting enterprises was significantly improved compared with that of lightly polluting enterprises; that is, the implementation of the new Environmental Protection Law has prompted enterprises to strengthen their environmental governance. $Trea{t}_i$ is a policy dummy variable; $Trea{t}_i$ takes the value 1 when the sample enterprise is a heavily polluting enterprise and 0 otherwise. $Tim{e}_t$ is a dummy variable for the year. Before 2015, $Tim{e}_t$ takes the value 0, and the value of the current year and afterward is 1. $Cep{i}_{it}$ is the mediator variable. Controls is a control variable, which is a random disturbance term.

In the above model, first, Equation (1) was regressed to test the impact of the new Environmental Protection Law on corporate environmental governance behavior. If the coefficient is significantly positive, it means that the implementation of environmental regulation policies has significantly improved the participation of enterprises in environmental governance behavior. Then, Equation (2) was regressed to test Whether environmental protection inspector has played a mechanistic role in the impact of the new Environmental Protection Law on corporate environmental governance behavior. If the coefficient of the interaction term is significantly positive, it means that environmental protection inspector has played a mechanistic role.

5. Empirical Analysis

5.1. Descriptive Statistics

Table 3. Descriptive statistics for primary variables.

Variables	N	Mean	P50	Sd	Min	Max
$Epi$	1059	0.335	0.345	0.094	0.054	0.563
$Treat$*$Time$	1059	0.488	0.000	0.500	0.000	1.000
$Treat$	1059	0.541	1.000	0.499	0.000	1.000
Cepi	1059	0.186	0.000	0.389	0.000	1.000
Size	1059	22.930	22.810	1.492	20.290	27.200
Lev	1059	0.445	0.440	0.200	0.070	0.921
Age	1059	19.230	19.000	4.828	8.000	31.000
Roa	1059	0.046	0.040	0.051	−0.142	0.190
Agencost	1059	0.073	0.063	0.047	0.010	0.248
Cflow	1059	0.063	0.061	0.062	−0.097	0.250
Tangibility	1059	0.279	0.260	0.166	0.009	0.687
Top1	1059	3.516	3.574	0.469	2.057	4.355
Growth	1059	0.227	0.102	0.704	−0.617	5.544
Soa	1059	0.490	0.000	0.500	0.000	1.000

shows the descriptive statistical results of the main variables. The mean value of corporate environmental governance behavior ($Epi$) is 0.335, the minimum value is 0.054, the maximum value is 0.563, and the standard deviation is 0.094, indicating that the overall cost increase in enterprises fluctuates less, indicating that the overall data dispersion is less.

5.2. Analysis of Regression Results

Basic regression results. This paper adopted the double-difference method to empirically analyze the impact of the implementation of the new Environmental Protection Law on corporate environmental governance behavior. The basic regression results of Equation (1) are shown in

Table 4. Basic regression results.

Variable	$\mathit{E}\mathit{p}\mathit{i}$
	(1)	(2)
$Treat$*$Time$	0.027 ***	0.038 ***
	(3.16)	(4.35)
Size		−0.022 *
		(−1.73)
Lev		−0.050
		(−0.92)
Age		−0.008 ***
		(−3.14)
Roa		−0.012
		(−0.12)
Agencost		−0.066
		(−0.40)
Cflow		−0.117 *
		(−1.67)
Tangibility		−0.014
		(−0.26)
Top1		0.031
		(1.28)
Growth		0.012 **
		(2.23)
Soa		0.004
		(0.06)
Constant	0.322 ***	0.888 ***
	(66.97)	(3.25)
Observations	1059	1059
R-squared	0.019	0.095
F	4.912	4.912

Note: *, **, *** indicate significance at the 10%, 5%, and 1% levels, respectively. The numbers in brackets are the t-values of the parameter estimates.

.

Table 4 shows that Column (1) regresses only corporate governance behavior ($Epi$) and the core explanatory variable ($Trea{t}_i*Tim{e}_t$). It can be seen that the estimated coefficient of the cross term is significantly positive at the 1% level, indicating that the implementation of the Environmental Protection Law has significantly increased the participation of enterprises in environmental governance and enhanced their environmental awareness. In Column (2), further control variables were added. It was found that the estimated coefficient of the core explanatory variable $Trea{t}_i*Tim{e}_t$ is still significantly positive at the 1% level, indicating that after controlling for other influencing factors, the implementation of the new Environmental Protection Law still significantly improves the participation of enterprises in environmental governance. Among the control variables, the regression coefficient of growth (Growth) is significantly positive, and the regression coefficients of enterprise size (Size), enterprise age (Age), and operating cash flow (Cflow) are significantly negative. Therefore, the size of the enterprise, the age of the enterprise, and the cash flow of operating activities are inversely proportional to the enthusiasm of the enterprise to participate in environmental governance, and the growth of the enterprise is directly proportional to the enthusiasm of the enterprise to participate in environmental governance. The basic regression results verify Hypothesis H1; that is, environmental regulation will significantly increase the participation of enterprises in environmental governance.

5.3. Robustness Check

5.3.1. Parallel Trend Test

One of the basic premises of using the double difference method is that the experimental group and the control group meet the parallel trend assumption; otherwise, the regression results may be biased. This paper referred to the practice of predecessors and uses dummy variables $\mathrm{Pr}e{3}_{it}$~$After{3}_{it}$ to replace $Trea{t}_i*Tim{e}_t$ in Equation (1). The specific regression equation is as follows:

$$\begin{array}{l}Ep{i}_{it}={\alpha }_1+{\alpha }_{-3}\mathrm{Pr}e{3}_{it}+{\alpha }_{-2}\mathrm{Pr}e{2}_{it}+{\alpha }_{-1}\mathrm{Pr}e{1}_{it}+{\alpha }_{0}Curren{t}_{it}\\ +{\alpha }_1^{\prime }After{1}_{it}+{\alpha }_2^{\prime }After{2}_{it}+{\alpha }_3^{\prime }After{3}_{it}+{\alpha }_{2}Controls+{\epsilon }_{it}\end{array}$$

(3)

In Equation (3), $\mathrm{Pr}e{3}_{it}$ represents the third year before enterprise i is subject to environmental regulation in period $t$. When the enterprise is in the third year before environmental regulation, the variable takes a value of 1; otherwise, it takes a value of 0. $After{1}_{it}$ represents the first year after company i is subject to environmental regulation in period $t$. When the company is in the first year after environmental regulation, the variable takes the value of 1; otherwise, it takes the value of 0. $Curren{t}_{it}$ represents the current period when enterprise i is subject to environmental regulation in period $t$. When the enterprise is in the current period of environmental regulation, this variable takes the value of 1; otherwise, it takes 0. The definitions of the remaining variables are the same as the regression results, and the regression results are shown in

Table 5. Parallel trend test.

Variable	$\mathit{E}\mathit{p}\mathit{i}$
	(1) Parallel Trend Test	(2) Test 2012	(3) Test 2011
$\mathrm{Pr}e3$	0.033
	(1.09)
$\mathrm{Pr}e2$	0.031
	(1.01)
$\mathrm{Pr}e1$	0.024
	(1.04)
$Current$	0.010
	(0.58)
$After1$	0.029 **
	(2.52)
$After2$	0.037 ***
	(3.32)
$After3$	0.053 ***
	(4.73)
Placebox1		0.020	0.018
		(1.31)	(1.56)
Size	−0.023 *	−0.008	0.005
	(−1.86)	(−0.89)	(0.66)
Lev	−0.054	0.012	−0.012
	(−0.98)	(0.18)	(−0.22)
Age	−0.004	0.319	0.253
	(−1.48)	(1.28)	(1.64)
Roa	−0.056	0.001	−0.001
	(−0.54)	(0.49)	(−0.33)
Agencost	0.016	−0.150	−0.222
	(0.09)	(−0.86)	(−1.52)
Cflow	−0.103	0.059	0.000
	(−1.46)	(0.38)	(0.00)
Tangibility	0.014	0.148 **	0.192 ***
	(0.26)	(2.45)	(4.07)
Top1	0.025	0.037	0.030
	(1.04)	(1.50)	(1.52)
Growth	0.012 **	−0.082 ***	−0.069 ***
	(2.29)	(−4.64)	(−4.84)
Soa	0.001	0.002	−0.001
	(0.02)	(0.07)	(−0.03)
Constant	0.870 ***	0.323	0.084
	(3.19)	(1.63)	(0.50)
Observations	1059	86	178
R-squared	0.108	0.802	0.517
F	3.603	4.912	4.912

Note: *, **, *** indicate significance at the 20%, 10%, 5%, and 1% levels, respectively. The numbers in brackets are the t-values of the parameter estimates.

.

From the test results in columns (1) of Table 5, it can be seen that the regression coefficients of $\mathrm{Pr}e{3}_{it}$~$Curren{t}_{it}$ are not significant, the regression coefficients of $After{1}_{it}$~$After{3}_{it}$ are all significant, and with the passage of time, the regression coefficient value increases and the significance increases. The above results show that before the implementation of the new environmental protection law, there was no significant difference between the experimental group and the control group, which satisfies the parallel trend hypothesis. The implementation effect of the new environmental protection law in promoting enterprises to strengthen environmental governance is good, but the implementation of this policy has a certain lag; that is, with time, its influence has increased. On the whole, the research design of this paper conforms to the premise of using double difference.

5.3.2. Placebo Test

In order to ensure the robustness of the experimental results, this paper uses counterfactual testing to conduct a placebo test and draws on Topalova [22] to test whether similar economic results will occur if environmental regulations are not implemented and to ensure that the deviation of the cost-markup rate of the treatment and control groups is only caused by the new Environmental Protection Law policy. Specifically, we set the sample data before the implementation of the new Environmental Protection Law policy, assuming that the new Environmental Protection Law policy occurred in 2013 and 2014, and performed the regression analysis again. If the improvement of corporate environmental governance behavior is indeed caused by the implementation of the Environmental Protection Law in 2015, then in the regression results of the fictitious new Environmental Protection Law transition year, the coefficient of the interaction term should be insignificant. Columns (2) and (3) of Table 5 report the test results: among them, the coefficient of the interaction term is not significant. This means that the rise in corporate environmental governance behavior was indeed brought about by the implementation of the new Environmental Protection Law policy in 2015. It can be seen that environmental regulation significantly improves corporate environmental governance behavior.

5.3.3. Propensity Score Matching

The coefficient of the interaction term in the regression results above only represents an “average effect”, that is, the average impact of environmental regulation on corporate environmental governance behavior. It is difficult to prove the relationship between environmental regulation and corporate environmental governance behavior. There is a real causal relationship; at the same time, whether an enterprise is affected by environmental regulation policies is not random, and changes in corporate environmental governance behavior are related to factors such as enterprise scale and R&D investment. If the influence of these factors is not excluded, it inevitably leads to sample selectivity deviation. Therefore, this paper adopts the propensity score matching method to correct the sample selection bias to reduce the interference with the experimental results and ensure that the results are more robust.

Table 6. Propensity score matching.

Variable	$\mathit{E}\mathit{p}\mathit{i}$
	(1)	(2)	(3)	(4)	(5)	(6)
$Treat$*$Time$	0.044 ***	0.044 ***	0.049 ***	0.046 ***	0.027 ***	0.038 ***
	(7.06)	(7.28)	(8.49)	(7.93)	(3.16)	(4.35)
Size		0.007 **		0.006 **		−0.022 *
		(2.33)		(1.99)		(−1.73)
Lev		−0.007		−0.010		−0.050
		(−0.29)		(−0.46)		(−0.92)
Age		−0.002 ***		−0.001		−0.008 ***
		(−2.75)		(−0.99)		(−3.14)
Roa		0.078		0.020		−0.012
		(0.98)		(0.28)		(−0.12)
Agencost		−0.078		−0.046		−0.066
		(−1.04)		(−0.65)		(−0.40)
Cflow		−0.041		−0.078		−0.117 *
		(−0.73)		(−1.50)		(−1.67)
Tangibility		0.098 ***		0.102 ***		−0.014
		(4.58)		(4.89)		(−0.26)
Top1		−0.000		0.007		0.031
		(−0.01)		(0.96)		(1.28)
Growth		0.003		0.004		0.012 **
		(0.77)		(0.93)		(2.23)
Soa		0.018 **		0.015 *		0.004
		(2.27)		(1.87)		(0.06)
Constant	0.313 ***	0.158 **	0.307 ***	0.143 **	0.322 ***	0.888 ***
	(61.70)	(2.24)	(71.15)	(2.12)	(66.97)	(3.25)
Observations	859	859	1059	1059	1059	1059
R-squared	0.020	0.030	0.018	0.021	0.019	0.095
F	4.912	4.912	4.912	4.912	4.912	4.912

Note: *, **, *** indicate significance at the 10%, 5%, and 1% levels, respectively. The numbers in brackets are the t-values of the parameter estimates.

shows the regression results after propensity score matching.

Columns (1), (3), and (5) of Table 6 show the comparison between the corporate environmental governance behavior ($Ep{i}_{it}$) and the implementation of the new Environmental Protection Law ($Trea{t}_i*Tim{e}_t$) after passing the proximity matching, radius matching, and nuclear matching tests, respectively. From the regression results, it can be seen that the estimated coefficients of the core explanatory variables ($Trea{t}_i*Tim{e}_t$) are all significantly positive. In Columns (2), (4), and (6), control variables are further added, the estimated coefficient of the core explanatory variable ($Trea{t}_i*Tim{e}_t$) is still significantly positive, and the value of the coefficient increases compared with the basic regression coefficient, indicating that the new environmental protection law implements behaviors to improve corporate environmental governance. As seen from these tables, the coefficient signs and significance levels of the interaction terms are consistent with the previous analysis. This also shows once again that environmental regulation is conducive to improving corporate environmental governance behavior and ensures the robustness of the basic regression results.

5.3.4. Replacing the Explained Variable

In this paper, the measurement method of the explained variable was replaced and re-examined. The results are shown in the first and second columns of

Table 7. Robustness check.

Variable	$\mathit{E}\mathit{p}\mathit{i}$
	(1)	(2)	(3)	(4)
$Treat$*$Time$			0.025 ***	0.036 ***
			(2.77)	(3.88)
L.Treat*Time	0.048 ***	0.044 ***
	(5.07)	(4.27)
Size		0.003		−0.024 *
		(0.73)		(−1.80)
Lev		0.032		−0.053
		(0.94)		(−0.91)
Age		−0.002		−0.007 ***
		(−1.47)		(−2.76)
Roa		0.227 **		−0.033
		(2.06)		(−0.31)
Agencost		0.002		−0.023
		(0.02)		(−0.13)
Cflow		−0.145 *		−0.131 *
		(−1.65)		(−1.74)
Tangibility		0.078 **		−0.012
		(2.38)		(−0.21)
Top1		0.003		0.033
		(0.30)		(1.27)
Growth		−0.005		0.014 **
		(−0.63)		(2.41)
Soa		0.010		0.003
		(0.82)		(0.05)
Constant	0.308 ***	0.218 **	0.324 ***	0.924 ***
	(41.46)	(2.09)	(62.84)	(3.19)
Observations	409	409	1047	1047
R-squared	0.002	0.018	0.015	0.088
F	4.481	4.481	4.481	4.481

Note: *, **, *** indicate significance at the 10%, 5%, and 1% levels, respectively. The numbers in brackets are the t-values of the parameter estimates.

. The regression coefficient between environmental regulation and corporate environmental governance behavior is still significantly positive, and the regression results are consistent with the previous results.

5.3.5. Explanatory Variables with One Lag Period

By considering the possible lag effect in the implementation of the policy, this paper performed a basic regression on the core explanatory variables with one lag period. The results are shown in the third and fourth columns of Table 7. It can be seen that the regression results are still significantly positive after considering the lag effect, which verifies Hypothesis 1 of this paper.

5.4. Mechanism Test

As a special political mechanism, environmental supervision can exert a deterrent effect on the decision-making behavior of enterprise management [23]. Therefore, this paper argued that environmental regulation could be improved through central government supervision. Therefore, this paper believes that environmental regulation can improve the attention of enterprises to environmental governance through central supervision. By taking the central inspector as the mediating variable, the mediation effect model was used to test the mechanism of environmental regulation affecting corporate environmental governance behavior.

Table 8. Mediation test.

Variable	$\mathit{E}\mathit{p}\mathit{i}$
	(1)	(2)
$Treat$*$Time$*$Cepi$	0.023 ***	0.026 ***
	(2.64)	(2.97)
Roa		0.035
		(0.35)
Lev		−0.075
		(−1.45)
Cflow		−0.138 *
		(−1.93)
Tangibility		0.002
		(0.04)
Top1		0.052 **
		(2.14)
Soa		0.004
		(0.06)
Constant	0.321 ***	0.175 *
	(64.19)	(1.96)
Observations	1047	1047
R-squared	0.022	0.040
F	2.694	2.694

Note: *, **, *** indicate significance at the 10%, 5%, and 1% levels, respectively. The numbers in brackets are the t-values of the parameter estimates.

shows the regression results of the mediation effect.

Table 8 shows the regression results of the mechanism effect model. There is a significant positive correlation between the interaction term and environmental protection inspectors, indicating that after the implementation of the new Environmental Protection Law, there is an urgent need for a corresponding supervision mechanism to ensure that environmental governance is achieved. The expected effect is that the implementation of the new Environmental Protection Law promotes an environmental inspection system. The coefficients of the interaction term and the environmental protection inspector are both positive, and after adding the environmental protection inspection variable, the coefficient of the interaction term decreases but remains significant; that is, the environmental protection inspector plays a role in the impact of environmental regulation on corporate environmental governance behavior. Thus, H2 is validated.

6. Further Research

6.1. Firm Size Heterogeneity

A large number of studies have shown that the benefits of large-scale enterprises are generally higher than those of small-scale enterprises in many aspects [24]. In order to examine whether the differences in the effects of environmental regulation exist under different enterprise scale characteristics, we set up dummy variables of enterprise scale and divide the sample enterprises into large-scale enterprises and small-scale enterprises. When performing basic regression on the two groups of samples obtained, the regression results in

Table 9. Heterogeneity tests.

Variable	$\mathit{E}\mathit{p}\mathit{i}$
	(1) Large Scale	(2) Small Scale	(1) State-Owned Enterprise	(2) Nonstate Enterprise
$Treat$*$Time$	0.035 ***	0.017	0.025 **	0.036 ***
	(3.07)	(1.23)	(2.10)	(2.83)
Size	−0.072 ***	−0.080 ***	−0.032 **	−0.063 ***
	(−3.42)	(−3.37)	(−2.01)	(−3.56)
Lev	0.051	−0.036	−0.046	0.018
	(0.57)	(−0.46)	(−0.63)	(0.21)
Roa	0.171	−0.148	0.139	−0.363 **
	(0.97)	(−1.11)	(1.06)	(−2.15)
Agencost	0.212	−0.130	0.025	0.027
	(0.79)	(−0.58)	(0.11)	(0.12)
Cflow	−0.117	−0.100	−0.064	−0.273 **
	(−1.05)	(−1.02)	(−0.66)	(−2.59)
Tangibility	−0.024	−0.012	0.117	−0.180 **
	(−0.29)	(−0.14)	(1.58)	(−2.24)
Top1	0.015	−0.027	0.036	0.061
	(0.42)	(−0.54)	(1.25)	(1.32)
Growth	0.008	0.008	0.012	0.016 **
	(1.18)	(0.67)	(1.44)	(2.32)
Soa	−0.090	0.130	-	-
	(−1.07)	(1.53)	(-)	(-)
Constant	2.053 ***	2.140 ***	0.939 **	1.569 ***
	(3.76)	(3.47)	(2.46)	(3.45)
Observations	528	519	517	530
R-squared	0.118	0.087	0.053	0.157
F	2.077	2.077	4.992	4.992

Note: **, *** indicate significance at the 5%, and 1% levels, respectively. The numbers in brackets are the t-values of the parameter estimates.

show that environmental regulation has a significant positive effect on large-scale enterprises. The effect on small-scale enterprises is not significant. Thus, H3 is validated. A possible reason for this is that large enterprises often have the significant financial strength and the ability to resist risks and can embark on independent transformation with the help of central environmental regulations. However, many small-scale enterprises lack sufficient operation processes and management mechanisms. Thus, when the same environmental regulation policy is encountered, the impact may be more severe.

6.2. Heterogeneity of Corporate Property Rights

State-owned and privately held companies have different property rights and take different factors into consideration in decision-making [25]. Therefore, state-owned and nonstate-owned companies may have different responses to the implementation of the new Environmental Protection Law. We constructed dummy variables for the type of enterprise ownership; divided the sample enterprises into two groups of data, state-owned enterprises, and nonstate-owned enterprises; and conducted a double difference for the two groups of samples. In the estimation results of the double-difference model in Table 9, the estimated coefficient of Equation (1) is 0.025 and is significant at the 5% level, and the estimated coefficient of Equation (2) is 0.036 and is significant at the 1% level. It can be seen that the impact of environmental regulation on the environmental governance behavior of nonstate-owned enterprises is greater than that of state-owned enterprises. Thus, H4 is validated.

7. Research Conclusions and Policy Recommendations

7.1. Analysis Conclusion

By using the data of A-share listed companies in Shanghai and Shenzhen from 2011 to 2020, this paper uses the double-difference model and the mediation effect model to empirically test the impact of environmental regulation on corporate environmental governance behavior and determine its impact mechanism. The main conclusions are as follows: environmental regulation has significantly improved corporate environmental governance behavior, and the central inspection function played a partial intermediary role in the impact of environmental regulation on corporate environmental governance behavior. Furthermore, heterogeneity analysis shows that environmental regulation improves corporate environmental governance behavior mainly in large-scale enterprises and nonstate-owned enterprises.

7.2. Policy Suggestions

The research in this paper has certain policy implications.

①

Environmental inspectors should be normalized and institutionalized. With the active promotion of environmental regulation policies, the ecological environment was improved to a certain extent, but there are still policy “loopholes” that act on the behavior of stakeholders. Therefore, the central environmental protection inspector should give full play to its power to deter and punish, focusing on local governments that commit fraud and evade responsibility and establishing severe punishments once violations are detected. At the same time, it is also possible to better consolidate and expand the results of inspections through re-examination and other methods.

②

There should be a focus on exploring institutional mechanisms that are conducive to an ecological civilization. When formulating environmental regulation policies, the government should comprehensively consider the scale of enterprises and the nature of enterprise ownership, continuously deepen reforms, standardize corporate behavior, comprehensively consider various factors, and develop diversified environmental regulation methods to ensure the environmental governance effect of the policy.

③

Implement environmental protection measures of different strengths according to the differentiated characteristics of enterprises. When the government further optimizes environmental policies, it should take into account the heterogeneity of the regulated enterprises at the attribute level. Higher-level units can increase the weight of environmental protection when formulating performance appraisal indicators for state-owned enterprises or introduce a one-vote veto system for environmental issues to mobilize the enthusiasm of state-owned enterprises to participate in environmental governance.