Impact of Environmental Regulations on High-Quality Development of Energy: From the Perspective of Provincial Differences

Abstract

Environmental regulation plays an important role in the green development of energy, and there are different classifications of environmental regulations in academic circles. This paper attempts to divide environmental regulation into “pure” formal environmental regulations and informal environmental regulations. By selecting the official statistical data samples of 30 provinces, autonomous regions and municipalities in China from 2005 to 2020, and by referring to the Malmquist productivity index method and the mediation effect model, we explore pure environmental regulations. By analyzing the influence mechanism of “pure” formal environmental regulations and informal environmental regulation on energy green development, it is concluded that: (1) Informal environmental regulations have direct and indirect roles in promoting the energy green development index. (2) There is a “U-shaped” relationship between “pure” formal environmental regulations and green energy development. Based on the above conclusions, relevant suggestions are put forward. At the same time, through this division of environmental regulations, on the one hand, it provides a basis for the government to accurately formulate environmental policies, and on the other hand, it provides strong support for the government to scientifically implement environmental policies.

1. Introduction

In 2020, global fossil energy consumption amounted to 83.1% of primary energy consumption, with coal consumption accounting for 27.2% of primary energy consumption at 15.142 EJ. According to the internationally accepted energy forecasting model, oil will be exhausted within 40 years, natural gas will be depleted within 60 years, and coal will only be available for another 220 years. Energy depletion is a predicament that many countries and regions have suffered from or will face. Around the globe, many developed and developing countries have more or less exploited and consumed non-renewable resources excessively in the process of economic development, which causes environmental disruption. As a result, these countries have fallen into or are sliding into the “pollution-treatment” trap [1,2,3].

The same is true of China, a country rich in resources. Many of its cities with abundant resources sprang up with the expansion of resource-based industries at the beginning of the reform and opening up. Nevertheless, resources have been gradually depleted as a result of excessive excavation and waste, and the original production process fails to make full use of resources, resulting in serious energy waste and environmental problems. The locked industrial structures also limit cities’ abilities to transform, which eventually leads to the degradation of both the economy and the environment. Despite rapid economic progress since reform and opening up, China is trapped in a vicious circle of “pollution-treatment” due to its industrial development model, which features high investment, high consumption, high pollution, and high emissions. Although China has weathered the most difficult period of economic development, the long-standing problems of unreasonable industrial structures, waste of production capacity, and serious pollution are becoming increasingly apparent [4].

In 2017, China’s expansionary economic growth consumed a lot of natural resources and accelerated the deterioration of the ecological environment [5]. China proposed high-quality economic development while emphasizing sustainable development, which suggests that the extensive development will be phased down gradually, and the maximum output should be achieved with the least consumption. According to the “Environmental Performance Index (EPI) Report” released in 2018, China’s EPI ranks only 120th among the 180 observed countries and regions in the world [6]. To achieve this transformation, one factor that cannot be dismissed is the environment. In order to realize the sustainable development of the environment, the importance of environmental regulations cannot be overstated. At present, the academic community generally believes that environmental regulation refers to the revision policy of environmental pollution behavior, including formal environmental regulations (that is, corresponding to various mechanisms implemented by public institutions for monitoring pollution emissions) and informal environmental regulations (that is, with citizens, citizen groups, or NGOs or the market to change all types of corporate polluting behavior). It is universally acknowledged that pollution is a behavior with negative externalities, but it is also the most beneficial behavior for the actors. Therefore, it is necessary to rely on external forces to convert the negative externalities of pollution behaviors into internal costs, and environmental regulations play such a function. Compared with informal environmental regulations, formal environmental regulations are discussed and studied more due to their mandatory nature. Formal environmental regulations include collecting sewage charges, issuing emission reduction subsidies, investing in pollution control, installing environmental protection equipment, etc. [7,8]. The principle behind formal environmental regulations is to reduce pollution behavior by increasing the unit cost of pollution discharge. Although informal environmental regulations are not mandatory, they do play an important role in supervising pollution behaviors and can also minimize pollution emissions to a certain extent.

However, environmental regulations focus solely on the environment. Under the theme of high-quality economic development, another key word of concern is the economy. In other words, the economy cannot be undermined while protecting the environment. Green development of energy (GDE) is an essential indicator that reflects the efficiency of converting input energy and other resources into GDP. It not only emphasizes energy utilization efficiency but also takes pollution emissions as an indicator of undesired output. That is to say, it takes into account both economic and environmental factors, and hence it is a metric for determining the economic production model that meets the requirements of the times.

However, in the actual operation process, some informal environmental regulations are often transformed into formal environmental regulations. For example, the public supervises environmental pollution issues and exerts pressure on the government to make the government take some measures to control pollution. This shows that some of the formal environmental regulations generally considered in academic circles are transformed from informal environmental regulations.

Therefore, the contributions of this paper are presented as follows: (1) The GDE index is calculated in light of the Malmquist productivity index approach, and the conversion efficiency of energy is measured. In the process of calculation, pollution emission is taken as an undesired output based on the concept of green production, which serves to balance the economy and the environment. (2) The influence of “pure” formal environmental regulations and informal environmental regulations on GDE is investigated. (3) The direct and indirect paths of informal environmental regulations affecting the GDE index are discussed. (4) Attempts are made to put forward the concept of pure formal environmental regulation, that is, environmental regulation of pure government will, which is obtained by eliminating the public’s influence on the government.

The remainder of this paper is organized as follows. Section 2 presents a discussion of the relevant literature and theoretical mechanisms. Section 3 introduces the research methods. Section 4 provides the study’s data and econometric methods. Section 5 provides the empirical results. Section 7 conducts the robustness test. The final section offers conclusions, policy implications.

2. Literature Review and Hypotheses

Environmental pollution is a problem that has threatened or is threatening many countries. In light of the theory of environmental economics, pollution is a public topic that will exhibit obvious negative externalities in the absence of long-term supervision, so environmental regulations are needed at this moment to prevent economic path dependence [9,10]. Environmental regulation research is well-developed, which indicates that environmental regulation is a normative policy system as well as the embodiment of social responsibility and environmental awareness [11,12]. Formal environmental regulations (mandatory environmental regulations) and informal environmental regulations are the two types of environmental regulations [13]. The former mainly relate to compulsory measures such as collecting sewage charges and promulgating legal provisions. By increasing the cost of sewage discharge, formal environmental regulations prompt enterprises to weigh the costs and benefits of sewage discharge and spontaneously lower the amount of sewage discharge in order to maximize profits [14]. The latter essentially includes spontaneous environmental protection activities that realize environmental governance by raising people’s environmental protection consciousness [15]. Regarding the measurement of environmental regulation intensity, the pollution reduction cost was the most commonly used measure by researchers in the United States and Europe [16,17,18]. However, these data were later found to be inaccurate because they could not precisely distinguish between pollution treatment-based spending and profit making-based spending. Aside from the pollution reduction cost, the energy price is also employed to measure the intensity of environmental regulations and explore its impact on trade competitiveness, productivity, and polluting gas emissions, etc. [19,20,21]. In addition, from the perspective of government intervention, the intensity of environmental regulations is measured by means of pollution control investment, the number and standards of environmental regulations, the enforcement of environmental regulations, emissions trading, and pollution taxes [22,23,24,25].

Energy research is primarily concerned with the innovation of energy efficiency computation methods, which includes the combination of the TOPSIS method and FAHP, as well as the multi-index pinch point analysis [26,27]. In recent studies, Cheng et al. looked into the relationship between carbon tax and energy innovation, used the QQR method to perform regression, and found a path to achieve carbon neutrality [28]. Energy innovation refers to the extended study of energy efficiency that focuses on the ratio of energy input and economic output [29,30,31]. Suo and Tang adopted the entropy TOPSIS method to calculate the high-quality development level of energy and performed cluster analysis based on regional differences [32]. Based on the above studies, it was discovered that in terms of the measurement of indicators such as energy efficiency, energy innovation, and high-quality development of energy, single-factor energy efficiency only considers the economic output that can be obtained from energy input. In subsequent studies, the meaning and scope of energy efficiency saw wider expansion. Specifically, apart from energy input, other production factors that play a vital role in economic output such as labor and capital were added; hence, the total factor energy productivity comes into being [33]. As environmental issues draw increasing attention from academia, the calculation method of incorporating environmental indicators into total factor energy productivity has been more and more widely used [34]. This notion is also applied in the measurement of the GDE index in this paper. In addition to the total factor energy productivity, environmental factors are also taken into account so as to reflect the utilization efficiency and cleanliness degree of energy in the economic development model.

There are many studies on the relationship between environmental regulation and green energy development at home and abroad, but environmental regulation is divided into pure formal environmental regulation and informal environmental regulation, and there is almost no research on its impact on green energy development. First, the acting path of informal environmental regulations must be looked into. As the main participants of informal environmental regulations are the general public, their enforcement is becoming increasingly crucial in pollution prevention and control as people’s environmental awareness and environmental demands grow [35,36]. Relevant studies have shown that public participation has played an essential and positive role in the implementation of environmental regulations by relevant departments [37,38]. The public’s demands for the environment prompt the government to strengthen supervision and governance or provide other services [39]. The public can express environmental protection appeals to the local government through petitions, telephone calls, online reports, environmental protection hearings, etc., as well as effectively supervise and restrict the local government’s implementation of environmental regulations, which can “force” the local government to intensify the enforcement of environmental policies and minimize disparities in environmental regulations for the sake of regional interests [40]. Moreover, the public can even vent their dissatisfaction by resorting to some extreme unconventional practices such as migration. Tiebout argued in his research that if the public is displeased with their cities’ public services, they will have a willingness to migrate [41]. This act is also referred to as an “exit threat” by Albert Hirschman [42]. From the viewpoint of the indirect effect of environmental regulations, some scholars have stated that the informal environmental regulations adopted by the public out of a desire to protect the environment have put pressure on relevant government departments. For instance, Dasgupta pointed out that public participation in environmental governance greatly affects auditing [43]. Based on the preceding analysis, the following hypothesis are proposed in this paper:

Hypothesis 1.

Informal environmental regulations have a direct impact on the GDE index.

Hypothesis 2.

Informal environmental regulations can indirectly affect the GDE index by increasing the intensity of formal environmental regulations.

Assuming the aforementioned hypothesis are correct, the impact of formal environmental regulations incorporates that exerted by some informal environmental regulations. That is to say, some formal environmental regulations are transformed from informal environmental regulations. Therefore, to disclose the relationship between PFERs and GDE, eliminating the influence of informal environmental regulations is necessary. To examine the relationship between PFERs and GDE, the relevant literature on the impact of formal environmental regulations on energy efficiency must first be analyzed. Xin et al. adopted the Metafrontier-Global-SBM super-efficient data envelopment analysis (DEA) model to measure economic green growth indicators and concluded that environmental regulations related to energy efficiency have a U-shaped association with economic green growth indicators [44]. Cui et al. evaluated the energy eco-efficiency of 31 mining cities in China and discovered that mandatory environmental regulations and market-based environmental regulations both inhibited energy eco-efficiency, and that mandatory environmental regulations had a threshold effect on energy eco-efficiency with the aid of Tobit regression and threshold regression models [45]. Therefore, the following hypothesis is also advanced in this paper:

Hypothesis 3:

Pure formal environmental regulation has a nonlinear relationship with green energy development.

According to the research of related scholars in academia (Suo, J. [32], Lieflaender, K.A., Bogner, F.X. [35], Quan, M. [39], etc.), formal environmental regulation has a non-linear relationship with energy green development. Since the pure formal environmental regulation proposed in this paper is the part of the formal environmental regulation that excludes the public will, the pure formal environmental regulation is subordinate to the formal environmental regulation, and so this hypothesis is proposed.

3. Research Methods

3.1. Variable Design

The explained variable in this paper is the egd (Energy Green Development Index), and it is calculated with the Malmquist productivity index approach. The specific model is described as follows: (${\mathrm{x}}_{\mathrm{t}}, {\mathrm{y}}_{\mathrm{t}}$) and (${\mathrm{x}}_{\mathrm{t}+1,} {\mathrm{y}}_{\mathrm{t}+1}$) are assumed to represent the input and output functions in t and t + 1 periods, respectively. According to Fare et al. [46], an output-oriented Malmquist production index model is extended and established therefrom, as illustrated in Formula (1):

$$\left\{\begin{array}{c}{\mathrm{M}}_{\mathrm{t}}=\frac{{\mathrm{D}}_{}^{\mathrm{t}}\left({\mathrm{x}}_{\mathrm{t}+1},{\mathrm{y}}_{\mathrm{t}+1}\right)}{{\mathrm{D}}_{}^{\mathrm{t}}\left({\mathrm{x}}_{\mathrm{t}},{\mathrm{y}}_{\mathrm{t}}\right)}\\ {\mathrm{M}}_{\mathrm{t}+1}=\frac{{\mathrm{D}}_{}^{\mathrm{t}+1}\left({\mathrm{x}}_{\mathrm{t}+1},{\mathrm{y}}_{\mathrm{t}+1}\right)}{{\mathrm{D}}_{}^{\mathrm{t}+1}\left({\mathrm{x}}_{\mathrm{t}},{\mathrm{y}}_{\mathrm{t}}\right)}\end{array}\right.$$

(1)

In light of the economical symmetry of the productivity indices ${\mathrm{M}}_{\mathrm{t}}$ and ${\mathrm{M}}_{\mathrm{t}+1}$ in the $\mathrm{t}$ and $\mathrm{t}+1$ periods, the Malmquist index that takes both the t and $\mathrm{t}+1$ periods into account is established, as shown in Formula (2):

$${\mathrm{M}}_{\mathrm{t}}({\mathrm{x}}_{\mathrm{t}},{\mathrm{y}}_{\mathrm{t}},{\mathrm{x}}_{\mathrm{t}+1},{\mathrm{y}}_{\mathrm{t}+1})=({\mathrm{M}}_{\mathrm{t}},{\mathrm{M}}_{\mathrm{t}+1}){}^{\frac{1}{2}}={\left[\frac{{\mathrm{D}}_{\mathrm{u}}^{\mathrm{t}}\left({\mathrm{x}}_{\mathrm{t}+1},{\mathrm{y}}_{\mathrm{t}+1}\right)}{{\mathrm{D}}_{\mathrm{u}}^{\mathrm{t}}\left({\mathrm{x}}_{\mathrm{t}},{\mathrm{y}}_{\mathrm{t}}\right)}\times \frac{{\mathrm{D}}_{\mathrm{u}}^{\mathrm{t}+1}\left({\mathrm{x}}_{\mathrm{t}+1},{\mathrm{y}}_{\mathrm{t}+1}\right)}{{\mathrm{D}}_{\mathrm{u}}^{\mathrm{t}+1}\left({\mathrm{x}}_{\mathrm{t}},{\mathrm{y}}_{\mathrm{t}}\right)}\right]}^{\frac{1}{2}}$$

(2)

As the returns to scale of VRS are variable, Formula (2) can be decomposed where technological change efficiency and technological progress are introduced. The decomposition process is illustrated in Formula (3):

$${\mathrm{M}}_{\mathrm{t}}({\mathrm{x}}_{\mathrm{t}},{\mathrm{y}}_{\mathrm{t}},{\mathrm{x}}_{\mathrm{t}+1},{\mathrm{y}}_{\mathrm{t}+1})=\frac{{\mathrm{D}}_{\mathrm{u}}^{\mathrm{t}}\left({\mathrm{x}}_{\mathrm{t}+1},{\mathrm{y}}_{\mathrm{t}+1}\right)}{{\mathrm{D}}_{\mathrm{u}}^{\mathrm{t}}\left({\mathrm{x}}_{\mathrm{t}},{\mathrm{y}}_{\mathrm{t}}\right)}\times {\left[\frac{{\mathrm{D}}_{\mathrm{u}}^{\mathrm{t}}\left({\mathrm{x}}_{\mathrm{t}},{\mathrm{y}}_{\mathrm{t}}\right)}{{\mathrm{D}}_{\mathrm{u}}^{\mathrm{t}+1}\left({\mathrm{x}}_{\mathrm{t}},{\mathrm{y}}_{\mathrm{t}}\right)}\times \frac{{\mathrm{D}}_{\mathrm{u}}^{\mathrm{t}}\left({\mathrm{x}}_{\mathrm{t}+1},{\mathrm{y}}_{\mathrm{t}+1}\right)}{{\mathrm{D}}_{\mathrm{u}}^{\mathrm{t}+1}\left({\mathrm{x}}_{\mathrm{t}+1},{\mathrm{y}}_{\mathrm{t}+1}\right)}\right]}^{\frac{1}{2}}$$

(3)

When the technological change efficiency is greater than 1, it indicates that the technological level of the research object has improved, and vice versa; when technological progress is larger than 1, it means that the total factor productivity has increased, and vice versa.

Labor force, capital stock, and energy consumption are selected as input indicators. Among them, the labor force is calculated by the number of employees in each region at the end of the year, and the capital stock is computed based on the social fixed asset investment and estimated using the perpetual inventory method, as depicted in Formula (4), where $K_t$ is the capital stock, $\delta$ is the depreciation rate, and $I_t$ is the investment amount.

$$K_{i,t}=\left(1-\delta \right)K_{i,t-1}+I_{i,t}$$

(4)

Among them, $K_{i,t}$ and $K_{i,t-1}$ represent the capital stock of the region i at time $t$ and $t-1$, respectively, $I_{i,t}$ represents the fixed asset investment in the region $i$ at time $t$, and $\delta$ represents the depreciation rate of fixed assets in the region i in time $t$. Referring to Shan’s treatment method, the depreciation rate $\delta$ is set to 10.96% [47].

In terms of the measurement of energy consumption, the electricity consumption of each city is chosen as a proxy variable. The GDP of each region is chosen as the desirable output index, whereas the discharge amounts of wastewater, sulfur dioxide, smoke, and dust are taken as non-desirable output indicators.

Explanatory variables: ① Intensity of formal environmental regulations (fer). Taking the availability of data and the government’s determination to protect the environment into account, the fer is calculated based on the proportion of industrial pollution control investment in GDP in each region [48].

② Informal environmental regulation intensity (per). This variable serves to measure the concern and supervision for environmental issues in the region. In this paper, the per is calculated according to the intensity of informal environmental regulation by the degree of attention to the word “environmental pollution” in the Baidu Index [49].

③ Intensity of PFERs (fer₂). Regression is performed with formal environmental regulations as the dependent variable and informal environmental regulations as the independent variable.

$$fe{r}_{i,t}={\zeta }_0+{\zeta }_{1}pe{r}_{i,t}+\epsilon 4_{i,t}$$

(5)

After the regression, adjusted R² is obtained. First, the minimum values are taken for the variables fer and per, respectively. Then, the minimum values are multiplied by the adjusted R² to obtain the overlapping parts of fer and per. Next, the overlapping part is subtracted by fer, and the outcome is the intensity of PERs [50].

$$fe{r}_{2i}=fe{r}_i-\min \left(fe{r}_i,pe{r}_i\right)R^2$$

(6)

Control variables include: ① financial development level (fin). Circulating funds act as the foundation for the normal operation of various economic entities, as well as the source of the economy’s vitality. Gaining insights from most domestic research, this paper uses the ratio of the total loans of regional financial institutions to GDP to measure the level of financial development.

② Labor market (rk). Labor is an essential factor for economic progress. When automation is underdeveloped, adequate manpower is a necessary condition for production. In this paper, population density (10,000 peoples/square kilometer) is taken to measure the labor market potential.

③ Foreign direct investment (invest). The inflow of foreign capital can increase the capital stock of the region and promote the popularization and application of energy-saving technologies in the region and beyond. In this paper, the average exchange rate over the years is used to convert the amount of foreign direct investment, and the result is further used to calculate the level of foreign direct investment based on its proportion in GDP.

④ Transportation capacity (trans). Coal and oil dominate China’s energy consumption structure, accounting for roughly 80% of total consumption, and they are generally circulated by means of freight transport. Transportation is a crucial part of the energy consumption process, and strong transportation capacity is a great guarantee for energy supply in various regions. In this paper, freight volume per square kilometer is used to measure the transportation capacity of each region.

3.2. Construction of Econometric Models

First, according to the previous analysis and assumptions, the following mediating effect model was constructed:

$$eg{d}_{i,t}={\alpha }_0+{\alpha }_{1}pe{r}_{i,t}+{\alpha }_2{X}_{i,t}+\epsilon 1_{i,t}$$

(7)

$$fe{r}_{i,t}={\beta }_0+{\beta }_{1}pe{r}_{i,t}+{\beta }_2{X}_{i,t}+\epsilon 2_{i,t}$$

(8)

$$eg{d}_{i,t}={\theta }_0+{\theta }_{1}pe{r}_{i,t}+{\theta }_{2}fe{r}_{i,t}+{\theta }_3{X}_{i,t}+\epsilon 3_{i,t}$$

(9)

Secondly, in line with the assumed nonlinear relationship between PFERs and GDE, the following econometric model was established:

$$eg{d}_{i,t}={\lambda }_0+{\lambda }_{1}fe{r}_{2,it}+{\lambda }_{2}fe{r}_{2,it}^2+{\lambda }_3{X}_{i,t}+\epsilon 4_{i,t}$$

(10)

$$X_{i,t}={\gamma }_0+{\gamma }_{1}lnfin+{\gamma }_{2}lntrans+{\gamma }_{3}lninvest+{\gamma }_{4}lnrk$$

(11)

where $eg{d}_{i,t}$ represents the GDE index in the year t of Province i; ferit denotes the intensity of formal environmental regulations in the year t of Province i; $fe{r}_{2,it}$ indicates the intensity of PFERs in year t of Province i; and X symbolizes the control variable.

4. Data Sources and Descriptive Statistics on Indexes

4.1. Data Sources

This paper selected 30 provinces, autonomous regions, and municipalities in China, based on the official statistical data from 2005 to 2020, among which the number of employees in each region at the end of the year and fixed asset investment were derived from the statistical yearbooks of various provinces and cities; the energy consumption was collected from the China Energy Statistical Yearbook; the total loans of financial institutions, GDP, foreign direct investment, population, education level, and age structure were taken from the statistical yearbooks of various provinces and the China Population and Employment Statistical Yearbook; the freight volume per square kilometer was sourced from the China Transportation Statistical Yearbook and the statistical yearbooks of various provinces and cities; and the discharge amount of wastewater, sulfur dioxide, and smoke and dust were gathered based on the China Environmental Statistics Yearbook.

4.2. Descriptive Statistics of Variables

According to the relevant data of research samples from 30 provinces, autonomous regions, and municipalities directly under the Central Government in China from 2005 to 2020, through statistical software analysis and sorting, the descriptive statistics of each variable were obtained, as shown in

Table 1. Descriptive statistics of variables.

Variable	Obs	Mean	Std. Dev.	Min	Max
egd	420	0.89955	0.118825	0.428	1
lnfin	420	4.723304	0.313581	3.992133	5.531104
lntrans	420	−0.56785	1.371916	−4.61463	2.829482
lninvest	420	0.414895	1.061563	−4.53641	2.300374
lnrk	420	−3.76233	1.292067	−7.09954	−0.96071
per	420	76.91975	50.02011	0.814941	240
fer	420	0.114555	0.062838	0.052706	0.502211
fer2	420	0.098626	0.062838	0.036776	0.486281
er	420	120.0936	270.5809	3.171111	1572.046
er2	420	116.9225	270.5809	0	1568.875
per2	420	0.466756	0.110479	0.158541	0.724552

.

5. Analysis of Empirical Results

5.1. Stationarity Analysis and Collinearity Test

The explanatory variables and the explained variables selected in this paper were tested for correlation using stata software, and the test results were sorted out. The results are shown in

Table 2. Variable correlation test results.

Variable	Adjusted t *	VIF
egd	−2.601 ***
lnfin	−11.0035 ***	1.88
lntrans	−6.2696 ***	3.35
lninvest	−4.7937 ***	1.75
lnrk	−15.7783 ***	2.38
per	−10.5558 ***	1.45
fer	−3.4813 ***	1.67

Note: * means significance at the 10% level; *** means significance at the 1% level.

.

A unit root test and a collinearity test were performed on each variable with the aid of STATA software. The results revealed that the unit root of each variable was stable, the VIF value of each variable was less than 10, and no collinearity existed.

5.2. Impact of Informal Environmental Regulations on GDE

As discussed before, informal environmental regulations affect GDE in two ways: by exerting a direct effect, and by playing an indirect role via formal environmental regulations. As demonstrated by

Table 3. Regression results of the mediating effect model.

Variable	(7)	(8)	(9)
	Egd	Fer	Egd
per	0.0004 *** (2.90)	0.0001 * (1.93)	0.004 *** (1.72)
fer			0.197 * (2.73)
Lnfin	0.033 * (1.66)	0.121 *** (14.03)	0.096 (0.39)
Lntrans	0.017 ** (2.39)	−0.018 *** (−5.74)	0.021 *** (2.77)
Lninvest	0.001 (0.11)	0.013 *** (4.37)	−0.002 (−0.25)
Lnrk	−0.020 *** (−3.19)	0.019 *** (7.08)	−0.024 *** (−3.58)
C	0.616 *** (6.95)	−0.409 *** (−10.29)	0.726 *** (6.99)
R2	0.083	0.399	0.089
F	7.64 ***	55.11 ***	6.73 ***
N	420	420	420

Note: * means significance at the 10% level; ** means significance at the 5% level; *** means significance at the 1% level.

, the regression results of Model 7, Model 8, and Model 9 were obtained by means of stepwise regression, based on which it could be preliminarily confirmed that H1 and H2 are valid. The regression results of Model 7 revealed that informal environmental regulations have a positive and direct stimulating effect on the GDE index, with the coefficient between them being 0.0004, which is significant at the 1% level. Therefore, it was concluded that the informal environmental regulations conducted by the public, who act as important participants in environmental protection, can positively and significantly influence energy efficiency and clean development. The regression results of Model 8 and Model 9 confirm that formal environmental regulations have a mediating effect between informal environmental regulations and the GDE index. Among them, Model 8 mainly demonstrated that informal environmental regulations engaging the public can impose pressure on relevant departments, consequently accelerating the implementation of formal environmental regulations. The regression results of Model 8 indicated that there was a positive correlation between pressure and the intensity of informal environmental regulations that engage the public. Specifically, when the pressure coefficient was 0.0001, i.e., the intensity of environmental regulations involving the public increased by one unit, the intensity of formal environmental regulations would grow by 0.0001 units as a result of the pressure exerted. Further, formal environmental regulations could contribute to increasing the GDE index. Therefore, H2 was confirmed.

In addition, based on the regression results in Table 3, it could be calculated that the total effect of informal environmental regulations on the GDE index was about 0.00042, and the indirect effect was 0.00002. From this, it was clear that the direct effect of informal environmental regulations on the GDE index was greater than the indirect effect. Furthermore, it can be seen in Table 3 that the coefficient between formal environmental regulations and the GDE index was 0.197, and it was significant at the 10% level, inferring that formal environmental regulations have a greater boosting influence on the GDE index than informal environmental regulations. The reason behind this disparity is that formal environmental regulations are a series of mandatory laws and regulations aimed against pollution that are promulgated by the government, so they enjoy the advantages of wider coverage and stronger normativity.

Among the control variables, the level of financial development influenced the GDE index positively, given the fact that capital is the lifeblood of the economy, as well as the primary driving force for a dynamic economy. The higher the financial development level in a region, the more monetary needs of various economic entities can be satisfied. Moreover, a higher level of financial development can better accelerate the circulation of social materials and currency, stimulate economic momentum, boost economic development, and speed up the elimination of backward production capacity, consequently raising energy efficiency. Furthermore, the level of financial development and the intensity of formal environmental regulations also exhibited a positive correlation since a high level of financial development promotes the capital flows of governments and enterprises and supports government investment in pollution management.

The correlation coefficient between transportation capacity and the GDE index was positive and significant. The fundamental reason for this is that transportation is an indispensable part of the energy consumption process, and high transportation capacity can guarantee energy supply in each region. Moreover, not only energy, but all kinds of products need to be transported to various marketing destinations. In this sense, transportation provides a strong assurance for economic operations. In conclusion, the stronger the transportation capacity, the greater the energy efficiency, and the higher the GDE index. However, the correlation between transportation capacity and formal environmental regulations was negative and significant at the 1% level. The primary reason for this is that the variable was calculated based on the proportion of pollution investment in GDP, where the GDP was raised due to the increased circulation rate of goods, while the emissions from transportation equipment were difficult to quantify.

Foreign direct investment contributes positively to formal environmental regulations. For one thing, foreign investment can help boost the local economy, raise government revenue, and assist relevant departments in carrying out environmental regulations more effectively. Another reason is that regions with more foreign direct investment are usually economically developed and have greater needs and efforts to manage the environment, resulting in a higher intensity of environmental regulation.

There was a negative correlation between the labor market and GDE, and it was very significant. This is due to the fact that the labor market was quantified based on population density. Specifically, as the population grows, so does the amount of energy consumed, as well as the difficulty and complexity of GDE. In terms of the correlation between the labor market and formal environmental regulations, it was a positive one. It is believed that as the population density increases, people have a higher demand for environmental welfare, and the environmental pressure that may be exerted is greater, which can drive the introduction of formal environmental regulations.

The regression results in Table 3 demonstrate that the coefficients α1, β1, and θ1 were significant, which initially verified the existence of the mediating effect. Furthermore, the Sobel test and the Bootstrap test were carried out. The Sobel test results revealed that there was a significant mediating effect, and the Bootstrap test results also demonstrated that fer acted as an active mediating variable between per and EGD. After 500 repetitions of the Bootstrap test, the following results were obtained and are listed in

Table 4. Bootstrap test.

Number of obs = 420 Replications = 500 _bs_1: r(ind_eff) _bs_2: r(dir_eff)
	Observed coed	Bootstrap std. Err.	z	p > │z│	Normal-based (95% conf. Internal)
_bs_1	0.000056	0.0000249	2.25	0.024	0.00000732	0.000105
_bs_2	0.000456	0.00001273	3.59	0.000	0.0002069	0.000706

; the p of the indirect effect was significant; the confidence interval excluded 0; and the indirect effect was noticeable. This further verified that informal environmental regulations can influence GDE through formal environmental regulations.

5.3. Impact of PFERs on GDE

Table 5. Regression results of Model 10.

Variable	(10)
Fer2	−0.414 (−1.13)
Fer22	1.332 * (1.84)
Lnfin	0.037 (1.47)
Lntrans	0.026 *** (3.68)
Lninvest	−0.004 (−0.6)
Lnrk	−0.023 *** (−3.47)
C	0.678 *** (6.51)
R2	0.08
F	6.00 ***
N	420

Note: * means significance at the 10% level; *** means significance at the 1% level.

presents the regression results of PFERs and the GDE index, which were also analyzed using STATA software. According to the regression results in Table 5, PFERs exerted a significant influence on the GDE index, and there was a “U-shaped” relationship between them. That is to say, the intensity of PFERs showed a negative relationship with the GDE index when it was lower than the inflection point value and exhibited a positive relationship with the GDE index when it was higher than the inflection point value. Based on the above analysis, it can be determined that a nonlinear relationship exists between the two, implying that H3 is valid.

The regression results of the control variables were the same as the results in Table 3. Concretely, the correlation relation between the level of financial development and the GDE index was positive, and the correlation coefficient of the transportation capacity and the GDE index was positive and extremely significant, which means that for every 1% increase in transportation capacity, the GDE index rose by 0.026%; the impact of foreign direct investment on the GDE index was not remarkable; and the labor market had a negative and noticeable correlation with GDE, suggesting that for every 1% rise in population density, the GDE index dropped by 0.023%.

6. Robustness Test

The intensities of formal environmental regulations and informal environmental regulations were calculated based on explanatory variables as described below:

Pure formal environmental regulation (er): pure formal environmental regulation refers to the environmental regulation of the government’s pure will, that is, the government supervises and restricts the emission behavior of enterprises through coercive means without the will of the public. In this test, the intensity of formal environmental regulations was determined based on the discharge volumes of industrial wastewater, industrial sulfur dioxide, and smoke (powder) dust [51], and the formula is as follows:

$$e{r}_{i,t}=\frac{1}{3}{\displaystyle \sum }_{l=1}^3\frac{e_{l,i,t}/y_{i,t}}{{{\displaystyle \Sigma }}_{i=1}^{179}\left(e_{l,i,t}/y_{i,t}\right)}$$

(12)

where i represents the city; l denotes the type of emissions (industrial wastewater, industrial sulfur dioxide, and industrial smoke (powder) dust in this paper); y symbolizes the industrial output value of the city. Since the investigation object was the intensity of urban pollution emissions, the data were processed in reverse to facilitate subsequent calculations.

Informal environmental regulations (per₂): the technique of factor analysis and the method proposed by Pargal et al. [52] were applied to combine per capita income, population density, education level, and age structure into a single index to reflect the intensity of informal environmental regulations.

The results shown in

Table 6. Robustness test 1.

Variable	(7)	(8)	(9)
	Egd	Er	Egd
Per2	0.211 *** (3.24)	410.714 *** (3.45)	0.0001 *** (2.54)
er			0.1183 *** (2.79)
Lnfin	0.051 *** (2.63)	376.746 *** (10.73)	0.025 (1.16)
Lntrans	0.016 ** (2.24)	−34.594 *** (−2.63)	0.018 ** (2.56)
Lninvest	−0.003 (−0.47)	46.066 *** (3.70)	−0.006 (−0.92)
Lnrk	−0.023 *** (−3.53)	90.087 *** (7.73)	−0.029 *** (−4.23)
C	0.488 *** (5.00)	−1550.92 *** (−8.68)	0.594 *** (5.63)
R2	0.087	0.410	0.101
F	7.91 ***	57.63 ***	7.75 ***
N	420	420	420

Note: ** means significance at the 5% level; *** means significance at the 1% level.

are nearly consistent with those in Table 3, which further demonstrates that the existence of the mediating effect was confirmed by stepwise regression in Model 7, Model 8, and Model 9, and the direct effect of informal environmental regulations outweighed the indirect effect.

As per the regression results of the control variables, the financial development level had a substantial promoting influence on the GDE index and formal environmental regulations. Transportation capacity had the potential to dramatically boost GDE while inhibiting pure formal environmental regulations. At the 1% level, foreign direct investment had a considerable promoting effect on formal environmental regulations. The labor market negatively influenced GDE, but it enjoyed a significant and positive correlation with pure formal environmental regulations.

The stepwise regression results in Table 6 preliminarily verified that the mediating effect exists. Then, the Sobel test and the Bootstrap test were performed. The Sobel test results revealed that there was a significant mediating effect, and the Bootstrap test results also showed that the mediating variable was very active. After 500 repetitions of the Bootstrap test, the following results were obtained, as shown in

Table 7. Bootstrap test 2.

Number of obs = 420 Replications = 500 _bs_1: r(ind_eff) _bs_2: r(dir_eff)
	Observed coed	Bootstrap std.Err.	z	p > │z│	Normal-based (95% conf. Internal)
_bs_1	0.051411	0.0129521	3.97	0.000	0.0260251	0.076796
_bs_2	0.135840	0.0554468	2.45	0.014	0.0271663	0.244514

; the p of both the direct effect and indirect effect were significant, and the confidence interval was [0.027, 0.245] excluding 0. This further verifies that informal environmental regulations can influence GDE through formal environmental regulations.

The robustness test was performed on the nonlinear relationship between PFERs and GDE, and the findings are listed in

Table 8. Regression results (2) of Model 10.

Variable	Model 10
er2	−0.0002 (−1.41)
er22	0.0000002 ** (2.21)
Lnfin	0.024 (1.10)
Lntrans	0.029 *** (4.07)
Lninvest	−0.004 (−0.63)
Lnrk	−0.024 *** (−3.49)
C	0.718 *** (7.35)
R2	0.095
F	7.22 ***
N	420

Note: ** means significance at the 5% level; *** means significance at the 1% level.

. As revealed by the findings, H3 was proven to be true again. The regression results of Model 10 indicated that there was a “U-shaped” relationship between PFERs and the GDE index, and the value of the inflection point was 2500 based on the calculation standard of er. It could also be found that the intensity of the PFERs gave a boost to the GDE index when it was higher than the inflection point value and exerted an inhibition effect on the GDE index when it was lower than the inflection point value. The regression results of the control variables were consistent with the regression results above.

7. Conclusions and Suggestions

In this study, the impact of informal environmental regulations and PFERs on GDE was explored with the aid of the STATA software, and the following conclusions were drawn:

(1) Informal environmental regulations exert an expansionary effect on the GDE index in two ways: directly promoting GDE, and indirectly boosting GDE via formal environmental regulations;

(2) The relationship between PFERs and GDE is “U-shaped”, i.e., when the intensity of PFERs is lower than the inflection point value, it presents a negative correlation with the GDE index, and when the intensity of PFERs is higher than the inflection point value, it increases the GDE index positively.

(3) The level of financial development influences the GDE index positively, mainly because capital acts as the lifeblood of the economy as well as the primary driving force for the economy to retain vitality. In addition, there is also a positive correlation between the level of financial development and the intensity of formal environmental regulations. This is primarily due to the fact that a high level of financial development expedites government and enterprise capital flows and boosts government investment in pollution control.

(4) The correlation coefficient between transportation capacity and the GDE index is positive and significant. The main reason underlying this phenomenon is that transportation is an essential part of the energy consumption process, and robust transportation capacity can guarantee the energy supply and economic operation of various regions. However, transportation capacity shows a negative correlation with formal environmental regulations, and this correlation is significant at the 1% level. This is determined by the calculation method of variables.

(5) Foreign direct investment has the potential to stimulate formal environmental regulations. Foreign investment helps boost local economic development, increases government revenue, and assists relevant departments in implementing environmental regulations more effectively.

(6) The labor market negatively and remarkably influences GDE. The primary reason behind this is that as the population grows, so does the amount of energy consumed, as well as the difficulty and complexity of GDE. As for the correlation between the labor market and formal environmental regulations, it is a positive one. It is believed that as population density increases, people have a higher demand for environmental welfare and the environmental pressure that may be exerted is greater, which can drive the introduction of formal environmental regulations.

Based on the above conclusions, the following suggestions are raised:

(1) Guide public behavior in an orderly manner and standardize informal environmental regulations. The most fundamental way to safeguard the environment is to raise people’s ideological awareness. To raise people’s ideological consciousness on environmental protection, the following measures are to be taken: strengthen environmental protection education in areas where the public are unwilling to protect the environment; create more communication channels between the public and the government, and popularize these channels simultaneously; respond to public concerns in time, and educate the public on how to submit their opinions in an orderly and standardized manner so as to prevent false information from reducing work efficiency.

(2) Improve the intensity of pure formal environmental regulations according to local conditions. Relevant departments should formulate policies that are favorable for local economic development and environmental protection in light of local economic conditions. The intensity of environmental regulations should be kept within the most reasonable range in order to acquire the maximum economic output with the least input and consumption.

(3) Raise the level of financial development while improving the convenience, safety, and speed of loans. Specifically, the following efforts can be made: strengthen the construction of official loan platforms, improve relevant regulations, and standardize relevant procedures to minimize unnecessary steps; establish an information assessment mechanism; expand more access to credit, except for transactions with larger amounts and higher risks; try to reduce physical mortgages to increase loan efficiency and capital operation speed; regulate the financial market order; and strictly investigate and prohibit non-standard high-interest lending institutions.

(4) Strengthen transportation and logistics infrastructure and improve the convenience of transportation. To this end, road construction must meet with more investment, the supply of materials and energy should be encouraged, and the liquidity of commodities and funds ought to be expedited so as to deliver a more dynamic economy.

8. Research Prospect

This paper gives insights into the factors that influence the GDE index based on the econometric model by measuring the GDE index, and it further verifies the impact of two types of environmental regulations on the GDE index. In addition, a robustness test is also performed. The innovations of this paper include:

(1) Adopt the Malmquist productivity index approach to calculate the GDE index and measure the conversion efficiency of energy, where the notion of green production is reflected;

(2) Clarify two paths through which informal environmental regulations affect GDE;

(3) Put forward the concept of PFERs and isolate the informal environmental regulations from the formal environmental regulations so as to examine the impact of PFERs on GDE.

However, there are still some limitations on account of constrained conditions:

(1) The calculation methods of variables correspond to some limitations. In particular, the calculation methods of informal environmental regulations have their limits. In this paper, the intensity of informal environmental regulations is measured based on how much public attention is paid to environmental issues, which neglects a transitional step, i.e., the transition between public attention to environmental issues and the intensity of informal environmental regulations. Not all environmental protection intentions can be implemented as informal environmental regulations, so this calculation approach can only reflect the intensity of informal environmental regulations in a roundabout way.

(2) The calculation of the intensity of pure formal environmental regulations has certain shortcomings as well. Taking into account the availability of data and the determination of relevant departments for environmental control, this paper uses the proportion of pollution control investment to GDP in each region to calculate the intensity of formal environmental regulations. However, this index fails to comprehensively measure formal environmental regulations.