Does Environmental Regulation Affect Circular Economy Performance? Evidence from China

Abstract

Environmental regulation is an important means to solve the environmental pollution challenges faced by economic development. Under the predicament of economic growth facing enormous environmental challenges, this paper aims to explore whether environmental regulation has a promoting effect on the circular economy performance, construct the mechanism through which environmental regulation impacts circular economy performance, as well as test regional heterogeneity on the impact of environmental regulation on circular economy performance under different levels of economic development. Exploring these factors is of great significance for meeting people’s demand for better living environments and promoting green and sustainable economic development. This paper takes interprovincial panel data of China from 2011 to 2022 as the sample and adopts structural equation modeling to explore the influence mechanism of environmental regulation on circular economy performance by constructing a measurement system of environmental regulation and circular economy performance. The results indicate the following: environmental regulation has a significant role in promoting circular economy performance, and there is a partial mediating effect of industrial structure upgrading within it; heterogeneity analysis shows that the promoting effect of environmental regulation on circular economy performance varies according to the different levels of regional economic development. Therefore, environmental regulation policies should be moderately adjusted to promote balanced economic and environmental development; environmental regulation policies should be enriched to promote industrial structure transformation and upgrading; regional synergistic governance should be improved to facilitate the promotion of the circular economy.

1. Introduction

China’s economic development has gradually changed from the pursuit of high speed to the pursuit of high quality. Continuous breakthroughs have been made in improving the economic scale and efficiency, and remarkable development accomplishments have been achieved [1]. However, under the traditional industrial economic development pattern characterized by energy consumption and serious pollution emissions, the accumulation of waste and pollutants beyond the carrying capacity of the natural ecosystem, resulting in serious environmental damage [2] and resource scarcity and other problems [3], have become the shackles that slow down the sustainable development of China’s economy [4,5]. Environmental pollution affects the ecological balance [6], hinders the achievement of sustainable development goals, and seriously restricts the high-quality development of China’s economy. At the same time, due to the public goods attribute of the ecological environment and the external diseconomies caused by environmental pollution, the serious environmental pollution problem is difficult to be significantly alleviated and improved through the self-regulation of the market mechanism; based on this, environmental regulation has become an important means to make up for the market failure [7,8]. The enactment and implementation of environmental regulation has the potential to mitigate the over-reliance of economic progress on natural resources and ecological sustenance [9]. Since the 17th National Congress of the Communist Party of China, the Party and the State have unequivocally articulated the imperative of “ecological civilization” construction, integrated environmental governance into the overall framework of national development, and steadfastly constructed the environmental governance system with the participation of multiple bodies, and they have also advocated for the comprehensive establishment of an efficient resource utilization regime and the stringent delineation of responsibilities for ecological environmental conservation [10,11], which shows the emphasis on ecological protection and green development, as well as the determination to strengthen the environmental regulation [12].

The roots of environmental problems are deeply entrenched in the irrationality of the economic development model, and the adjustment and optimization of the industrial structure, as an important component of the economic system, is the link between economic activities and ecological efficiency, which also is the key path to guide the high-quality development of a country’s economy [13]. Whether the industrial structure is reasonable or not shows the coordination of economic growth [14,15]. Some scholars believe that diversified environmental regulation policies can generate double pressure from local government regulation and public opinion, which plays the role of an external binding force and provides a driving mechanism for industrial structure transformation and upgrading [16], while the promoting effect of environmental regulation tools such as legislation, supervision and economic incentives will be gradually become more prominent with the improvement of the regional economic development level. Therefore, it is necessary to enrich the environmental regulation policy tools and seek precise regulation in the field of the environment, and support the upgrading of the industrial structure [17]. With the improvement and enhancement of environmental regulation policies, industrial structure, as the target of environmental regulation, can stimulate the expansion of economic scale and the growth of economic benefits through optimization and upgrading, as well as spatial agglomeration effects. This not only facilitates the provision of various public pollution control infrastructures for enterprises within clusters, achieving economies of scale in pollution control, but also fosters the development of specialized environmental industries, offering tailored pollution treatment solutions, thereby reducing per-unit pollution control costs for clustered enterprises [18], which can help to realize the mutual benefits of environmental protection and sustainable economic development [19,20]. Furthermore, industrial structure upgrading can also reduce the excessive dependence on energy consumption by increasing the development and utilization of clean energy, while encouraging clean production and consumption, promoting coal removal and emission reduction [19], in order to promote the rapid development of the circular economy.

Therefore, under the predicament of economic growth facing enormous environmental challenges, the policies of environmental regulation have been continuously enriched. In this context, it is worth exploring whether environmental regulation has a promoting effect on the performance of the circular economy, as well as the specific mechanism that determines how environmental regulation impacts circular economy performance; it is of great significance in order to realize the compatibility between the protection of environmental interests and the high-quality green development of the economy, as well as to satisfy the people’s demand for a better living environment, and so on.

Regarding the relationship between environmental regulation and the circular economy, some scholars have explored the impact of environmental regulation on enterprises engaging in circular economy practices. Wang (2021) [21] believes that environmental regulation can stimulate the environmental responsibility of manufacturing enterprises and encourage them to develop green technology innovations, and thus environmental regulation can promote the development of the circular economy and environmental sustainability at the enterprise level. Barón Dorado’s (2022) [22] research also reinforces this opinion. Dorado believes that environmental regulation is an important way to promote sustainability and develop effective ecological innovation within enterprises, which can maintain and potentially improve their environmental performance, motivate enterprises to engage in more circular economy practices, emphasize the importance of formulating environmental regulation, and this research also points out that an effective environmental management system is a guarantee for the implementation of environmental regulation policies. Other scholars have also explored and verified the role of environmental regulations in promoting circular economy performance. Shang (2022) [23] provides evidence of the relationship between environmental regulations and circular economy performance, based on the decomposition of China’s circular economy performance and the growth rate of the circular economy, that is, environmental regulations can play a linear role through the “catch-up effect”, while the “innovation effect” and “demonstration effect” have not yet played an effective role, so the role of “innovation effect” and “demonstration effect” should be further strengthened to improve circular economy performance on the basis of measuring the value of provincial circular economy performance.

The above studies provide useful references for analyzing the relationship between environmental regulations and circular economy performance, but there is little literature that delves into the specific path of interaction between the two. Based on this, this paper conducts an empirical analysis using provincial-level panel data in China from 2011 to 2020, focusing on thirty provinces in China as research subjects, in order to explore the impact mechanism of environmental regulation on circular economy performance. This paper also conducts a heterogeneity analysis and proposes targeted policy recommendations. The possible marginal contributions of this paper are: (1) constructing a higher-order constructive measurement model for measuring environmental regulation and circular economy performance, which complements the methodology for quantifying environmental regulation and circular economy performance; (2) applying structural equation modeling, constructing and verifying the mechanism connection between environmental regulation, industrial structure upgrading and circular economy performance; and (3) testing whether there is regional heterogeneity in the impact of environmental regulation on circular economy performance under different levels of economic development.

2. Literature Review and Hypothesis

A circular economy is described as the process economy of “resource-product-renewable resource”, which promotes sustainable development [24,25]. It advocates achieving maximal economic benefits through investment savings, waste recycling, and other methods, while also conserving the use of scarce resources and non-renewable energy, as well as the safe treatment and the efficient flow of waste, so as to achieve the regeneration of “waste” to “resource” [26]. The root of the circular economy is the coordinated development among economy, environment and resources [27]. Circular economy performance is the comprehensive relationship between the input costs of operation and the results obtained during the process of circular economy development [28].

The circular economy plays a vital role in ensuring the reuse of waste created and, therefore, reducing the waste of limited resources, which is the primary goal of the general economic concept [29]. The circular economy has gained a great deal of attention among scholars. Still, the transition to the circular economy is a complex task [30]. The circular economy is frequently advocated and acknowledged as a tool or strategy for achieving sustainable development through encompassing the three pillars of environmental, economic and social development [27,31]. The pursuit of sustained economic growth cannot be at the expense of environmental degradation; therefore, in order to realize sustainable development, one dimension should not be pursued at the expense of others. Millar et al. (2019) [32] emphasize that the circular economy could serve as an effective tool to achieve sustainable development. Schroeder et al. (2018) [33] suggest that the circular economy can contribute directly to attain a variety of sustainable development goals. Hondroyiannis et al. (2024) [26] regard that the circular economy can contribute to the promotion of sustainable development issues by improving economic, environmental and social goals.

To address the environmental challenges stemming from the conventional industrial development model characterized by high-input, high-energy consumption, and high levels of pollution, Du (2023) points out that environmental regulation has emerged as a pivotal instrument in national governance for mitigating environmental pollution and reshaping economic development [34]. Environmental regulation focuses on rectifying the negative externalities of environmental pollution, through mechanisms such as the enactment and enforcement of governmental environmental measures, the incentives of market regulation, and the exertion of public supervision, which involves the collective participation of various subjects to collectively adjust and constrain activities detrimental to ecosystems. It effectively alleviates environmental issues resulting from extensive economic development [35], internalizes the external diseconomies caused by pollution, and makes recycling all kinds of resources and improving utilization efficiency a priority choice for reducing emission costs. This approach significantly elevates the level of environmental governance, improves the trajectory of economic growth, and facilitates the advancement of green production [36]. It serves as a crucial pillar and safeguard for enhancing circular economy performance.

The “Pollution Haven Hypothesis” posits that environmental regulation will affect the transfer of different types of industries between countries or regions, which in turn will have an impact on the industrial structure adjustments of both the transferring and receiving countries or regions. Specifically, countries or regions with stringent environmental constraints tend to relocate pollution-intensive industries to areas with more relaxed environmental regulations for production activities [37,38]. Environmental regulations will constrain or gradually phase out the enterprises with high pollution levels or inadequate potential for transformation, and at the same time establish high environmental standards as barriers to market entry, thereby providing greater development opportunities for energy-saving industries [39], and accelerating the advancement of the circular economy.

The “Porter hypothesis” contends that in the short term, enterprises may be subjected to greater pressure because of the “compliance costs” brought by environmental regulation, leading to a compression of profit margins [40]. However, in the long term, environmental regulation will stimulate the innovation potential and technological investment of enterprises [41,42], the progress and innovation of production technology will substantially reduce the cost of environmental regulation, yielding the “innovation compensation” effect. The lingering belief that environmental regulations erode competitiveness has resulted in a stalemate. Porter and van der Linde C. (1995) [43] proposed that the relationship between environmental regulations and economic competitiveness should not be viewed as mutually exclusive. Stricter environmental regulations can actually motivate businesses to adopt innovative environmental technologies and management practices, which can lead to increased resource efficiency, reduced waste and lower operating costs. Such environmental regulation can actually incentivize businesses to seek innovations in environmental protection and to improve efficiency, which in turn enhances their competitiveness.

The progress of production technology and innovation will largely reduce the cost of environmental management, that is, to produce the “innovation compensation” effect. Under the supervision of environmental regulation, enterprises will strengthen the control of pollutant emissions during the production process by adopting energy-saving technologies and improving production methods, which aims to elevate energy conservation and environmental protection standards, enhance production efficiency, and ultimately achieve environmental preservation objectives. Consequently, it incentivizes a shift in industrial structure towards industries with higher added value and more rational resource allocation, thus providing high-level technological support for economic green development. Moreover, while environmental regulation restricts the development of high-polluting and low-end manufacturing industry industries [44], it will concurrently provide greater opportunities for the growth of low-carbon and environmentally friendly industries, which can enhance their comparative advantages, attract more capital inflows, and expand their production scales [45]. For instance, governmental subsidies and support measures tend to favor low-carbon and green environmentally friendly industries, thereby facilitating the optimization and adjustment of the industrial structure.

Environmental regulation, besides its coercive binding force, serves as a vehicle for disseminating green, low-carbon, and eco-friendly consumption ideologies, thus fostering consumer awareness regarding the importance of environmental preservation and resource conservation [46]. Consumer demand structures undergo corresponding changes with different stages of economic development. Presently, due to continuous advancements in economic development, there is a heightened emphasis on personalized and diversified consumption demand among the public. Particularly in this era of heightened environmental consciousness, public consumption preferences are shifting towards eco-friendly products, manifesting a preference for green and environmentally sustainable goods. Consequently, this trend amplifies the demand for eco-friendly products while reducing the demand for non-environmentally friendly alternatives. Fluctuations in individual consumption patterns impel enterprises to adapt their production strategies accordingly, thereby instigating continual adjustments and upgrades in industrial structures [47]. At the same time, the rising demand for environmental protection among the public necessitates greater transparency in environmental information, which exerts supervisory pressure on corporate pollution behaviors [48], which, in turn, urges enterprises to actively assume environmental responsibilities and accelerate green transformation efforts, so as to mitigate the adverse repercussions associated with the exposure of polluting practices.

The adjustment and optimization of industrial structure, beyond being the primary targets of environmental regulations, play an indispensable role in reducing resource dependency and improving environmental quality throughout the process of economic development. Specifically, under the constraints of environmental regulation, the industrial structure upgrading primarily involves enhancing ecological efficiency, reducing pollutant emissions, and promoting sustainable economic development, thereby impacting the circular economy. Firstly, in the process of industrial structural upgrading, the center of gravity of economic structure shifts gradually towards the tertiary industry, while the proportion among three industries aligns more closely with the demands of high-quality economic development. Given the low energy consumption, minimal pollutant emissions, and high economic returns associated with the tertiary sector, its increased proportion contributes to environmental quality optimization, alleviating and ameliorating structural pollution issues. As the industrial structure shifts from lower to higher levels, industrial division becomes more refined, and the industrial chain extends towards industries with higher added value [49], stimulating the rapid development of high-end industries characterized by high knowledge and technological content, efficient utilization of material resources, environmental friendliness, and promising growth prospects. This, in turn, injects new strengths and dynamics into economic growth [50].

Furthermore, the industrial structure upgrading deepens the coordination and integration among industries, leading to a more balanced allocation and utilization of resources across sectors [51], and the factors of production such as labor and capital in low-efficiency sectors undergo a swift transition towards more efficient sectors through mechanisms such as market-driven adjustments, fostering the expansion of sectors with high productivity within the overall industrial structure. Consequently, the overall economic efficiency of the industries will also obtain significant enhancement. Simultaneously, the efficient utilization of energy will further elevate green productivity, which will in turn improve ecological welfare performance and the level of economic sustainability.

Moreover, the process of industrial structural upgrading is accompanied by increased investment in enterprise innovation, leading to the development and implementation of smarter, environmentally friendly production processes and technologies, thereby creating a conducive environment for the advancement of the circular economy and providing complementary technical support. With the adoption of new production technologies, outdated and excessive production capacity will be phased out. The energy consumption structure transitions from predominantly non-renewable energy consumption to an increasing proportion of clean energy usage [52], which effectively reduces pollution per unit of output, thus the economy tends towards greener and higher-quality development [53], facilitating the enhancement of circular economy performance.

In summary, the following hypotheses are proposed in this paper:

Hypothesis 1 (H1). 

Environmental regulation has a positive impact on circular economy performance.

Hypothesis 2 (H2). 

Environmental regulation has a positive impact on industrial structural upgrading.

Hypothesis 3 (H3). 

Industrial structural upgrading has a positive impact on circular economy performance.

Hypothesis 4 (H4). 

Industrial structural upgrading serves as a mediator in the impact of environmental regulation on circular economy performance.

The Environmental Kuznets Curve (EKC) theory posits that there is an inverted U-shaped relationship between the degree of environmental pollution and the economic development of a country or region [1,54]. Therefore, disparities in economic development levels affect environmental quality, which in turn influences the adoption of environmental regulatory policies and the realization of their binding effect. Specifically, the advancement of economic development levels is typically accompanied by increased investment in capital and R&D investment, which can be allocated towards supporting green technology innovation, infrastructure development, and production process upgrades. This can foster the development and utilization of new technological achievements to enhance resource allocation efficiency, alleviate resource consumption pressures, and increase the rate of production waste recycling, thereby invigorating the healthy development of eco-industries and consequently improving the performance and development of the circular economy.

In regions with varying levels of economic development, governmental capacity for environmental governance and the stringency of environmental regulations differ. Typically, in regions with higher economic level, governments tend to pay greater emphasis on environmental protection and sustainable economic development [55], the intervention in environmental regulation may be intensified, such as introducing more stringent environmental laws and policy measures to restrict and prohibit the entry and development of pollution-intensive industries, augmenting penalties for environmental degradation, and fostering green innovation and development within enterprises. Such interventions can influence the intensity and effectiveness of environmental regulation and the performance of the circular economy.

Furthermore, as economic development levels rise, public attitudes towards green consumption and individual environmental awareness evolve, paying more attention to environmental protection and sustainable development, leading to increased demand for eco-friendly products and services, thereby stimulating the development of the circular economy industry chain. Additionally, advancements in education and heightened social awareness prompt individuals to prioritize environmental quality, resulting in stronger recognition and enforcement of environmental policies, along with spontaneous monitoring and rejection of industries that neglect ecological conservation.

In summary, the following hypothesis is proposed:

Hypothesis 5 (H5). 

There is heterogeneity in the impact of environmental regulation on circular economy performance across regions with different levels of economic development.

Based on the above analysis and hypothesis, the conceptual model of this paper is shown in Figure 1.

3. Data and Research Methodology

3.1. Data

In this paper, panel data from 30 provinces in China from 2011 to 2020 were chosen as research samples to investigate the impact of environmental regulation on circular economy performance, as well as the mediating role of industrial structural upgrading. The data were obtained from the “China Statistical Yearbook,” the “China Environmental Statistical Yearbook,” the “China Energy Statistical Yearbook,” and the statistical database of China Economic Net. Missing values in the collected data are imputed using the mean imputation method. Additionally, due to disparities in units and dimensions among different variables, the imputed data are subjected to dimensionless standardization processing. The rationale for utilizing data from yearbooks and statistical databases lies in the fact that these datasets are compiled and disseminated by official statistical authorities, thus ensuring their accuracy and comprehensiveness. Additionally, these data are comparatively easily accessible, guaranteeing the continuity and credibility of the data.

3.1.1. Environmental Regulation

The diverse array of measures encompassed within environmental regulation and the different entities responsible for their implementation, as well as the selection of measurement indicators necessitates a comprehensive consideration of their nature and disparities, given the complexity and diversity of regulatory means and types. Thus, this paper proposes to assess environmental regulations from three dimensions: executive order (EO), market incentive (MI), and public participation (PP) [56,57,58,59]. The evaluation indicators are shown in

Table 1. Environmental regulation measurement index system.

Latent Variable	Dimension	Measurement Variable Description	Index
Environmental regulation (ER)	Executive order (EO)	Administrative penalty decision amount	EO1
		Number of suggestions from the National People’s Congress	EO2
		Number of personnel in environmental monitoring agencies	EO3
	Market incentive (MI)	Pollution discharge fees	MI1
		Total investment in environmental pollution control	MI2
		Environmental protection expenditure	MI3
		Urban maintenance and construction tax	MI4
	Public participation (PP)	Number of environmental complaint letters	PP1
		Number of environmental reports and complaints	PP2
		Environmental Pollution Baidu Search Index	PP3

.

3.1.2. Circular Economy Performance

Performance is the synthesis of inputs and outputs; circular economy performance encapsulates the comprehensive relationship between the input costs incurred during the operation and the outcomes achieved through the process of circular economy development. Therefore, the circular economy performance measurement index system is constructed from the three dimensions: Energy Consumption and Utilization (ECU), Ecological Pollution (EP), and Economic and Social Development (ESD) [27,60,61,62], reflecting the intricate interplay of resources, environmental impact, and socio-economic advancement inherent in the circular economy paradigm. The evaluation indicators are shown in

Table 2. Circular economy performance measurement index system.

Latent Variable	Dimension	Measurement Variable Description	Index
Circular economy performance (CEP)	Energy Consumption and Utilization (ECU)	Comprehensive utilization of general industrial solid waste per unit GDP	ECU1
		Electricity consumption per unit GDP	ECU2
		Energy consumption per unit GDP	ECU3
	Ecological Pollution (EP)	Industrial wastewater discharge COD per unit GDP	EP1
		Industrial sulfur dioxide emissions per unit GDP	EP2
		Industrial smoke and dust emissions per unit GDP	EP3
		Harmless treatment capacity of domestic garbage	EP4
	Economic and Social Development (ESD)	GDP	ESD1
		Per capital disposable income	ESD2
		R&D investment intensity	ESD3
		Urbanization level	ESD4

.

3.1.3. Industrial Structure Upgrading

Industrial structure upgrading entails more than mere shifts in the proportion of output among various industries, it encompasses the dynamic evolution of specialization across sectors, so the level of industrial structure upgrading is measured in terms of industrial structure rationalization [53], industrial structure sophistication [63] and industrial upgrading rate [17]. The evaluation indicators are shown in

Table 3. Industrial structure upgrading measurement index system.

Latent Variable	Measurement Variable Description	Index
Industrial Structure Upgrading (ISU)	Industrial structure rationalization	ISU1
	Industrial structure sophistication	ISU2
	Industrial upgrading rate	ISU3

.

3.2. Research Methodology

Structural Equation Modeling (SEM) is a statistical analysis tool frequently employed in modern research for examining and predicting direct or indirect relationships among variables. It facilitates the analysis and interpretation of complex relationships among multiple variables [64], quantifying the strength of relationships between variables as path coefficients for direct understanding and observation. Combining factor analysis from traditional statistics with regression analysis based on linear models, SEM can concurrently deal with multiple equations and test the validity of various theoretical models. SEM encompasses the covariance-based structural equation modeling (CB-SEM) and the partial least squares-based structural equation modeling (PLS-SEM). Compared with CB-SEM, PLS-SEM offers several advantages:

• It is more suitable for situations with small samples, enabling effective analysis even with fewer samples without model identification issues [65,66];
• It has relatively lenient requirements regarding data distribution, capable of handling non-normally distributed data [67];
• It is suitable for exploratory research as well as the establishment and prediction of complex models, allowing for the consideration of both direct and indirect effects among multiple latent variables, which can provide a more comprehensive reflection of relationships between variables, making it applicable to studies involving hierarchical structures or multiple influencing factors [68,69].

Therefore, the structural equation model adopted in this study is PLS-SEM, and the statistical analysis software utilized is Smart-PLS 4.0.

The structural equation model constructed in this study not only includes first-order latent variables, but also comprises two higher-order latent variable measurement models. Therefore, it is necessary to first examine and analyze the two higher-order measurement models in this study separately as complete measurement models, assessing the reliability, validity, and other aspects of the higher-order latent variables. Currently, the methods for handling higher-order latent variables in Smart-PLS include the repeated indicators approach and the two-stage approach [70]. Although these two methods yield similar results when the sample size is sufficient, the repeated indicators approach requires the number of indicators for lower-order constructs to be consistent, otherwise, bias may occur [70]. Since the number of indicators for each sub-configuration of the two higher-order models in this study varies, to avoid the influence of bias, the two-stage approach is employed.

The Structural Equation Model comprises measurement and structural models, wherein the measurement model is employed to describe the relationships between latent variables and the corresponding measurement variables. Its purpose is to assess the extent to which measurement variables contribute to latent variables; it is the degree to which a set of measurement variables can substitute for the latent variables they represent. The equation for this is as follows:

X = ∧x ξ + δ

(1)

Y = ∧y η + ε

(2)

X and Y are the vector of exogenous and endogenous observed variables, respectively; ∧x and ∧y represent the factor loading matrix (standardized regression coefficients), which describes the relationship between latent variables and observed variables. δ and ε represent the random errors in measurement.

The structural model is used to describe the path relationship between exogenous and endogenous latent variables and is formulated as follows:

η = Γ ξ + ζ

(3)

η = β η + Γ ξ + ζ

(4)

Equation (3) represents the structural model, Equation (4) denotes the structural model with mediators. η and ξ represent endogenous latent variables and exogenous latent variables, respectively. β is the regression coefficients among endogenous latent variables, reflecting their interrelations. Γ denotes the regression coefficients between exogenous latent variables and endogenous latent variables, elucidating their associations. ζ is the error terms of the structural model.

In structural equation modelling, in order to assess the reliability of our constructs, reliability tests are performed, including internal consistency tests and combined reliability tests, measured by Cronbach’s alpha and composite reliability (CR). Cronbach’s alpha refers to a measure of internal consistency, indicating the degree of related closeness the items comprising a group are. CR measures the sum of the latent variables’ factor loading relative to the sum of the factor loading plus error variance.

4. Results

4.1. Higher-Order Construct Measurement Model Results

The results in

Table 4. Reliability and convergent validity analysis.

Higher-Order Construct	First-Order Construct	Observed Variable	Factor Loading	Cronbach’s Alpha	C.R.	AVE
ER	EO	EO1	0.760	0.725	0.846	0.648
		EO2	0.774
		EO3	0.875
	MI	MI1	0.757	0.858	0.904	0.702
		MI2	0.880
		MI3	0.850
		MI4	0.859
	PP	PP1	0.819	0.752	0.857	0.667
		PP2	0.801
		PP3	0.829

demonstrate that the factor loading coefficients of each observed variable for environmental regulation are all above 0.60, which meets the requirements of the study, and the basic fit of the model is good. Cronbach’s Alpha coefficient was used to test the internal consistency of the measurement indicators [71], and this study obtained that the Cronbach’s Alpha coefficients of each first-order construct are all above 0.8, with the combined reliability (CR) values exceeding 0.7, which proves a high level of reliability.

Validity assessment aims to verify whether the observed variables can accurately measure latent variables, including the convergent validity test and discriminant validity test [72]. As shown in Table 4, the average variance extract (AVE) of each construct is greater than 0.5, which indicates the model has high convergent validity [73]. As shown in

Table 5. Discriminant validity analysis.

	EO	MI	PP
EO	0.805
MI	0.752	0.838
PP	0.689	0.772	0.817

Notes: Diagonal numbers represent the square root of AVE values, while numbers in the lower triangle represent the correlations of other constructs. The same applies as follows.

, according to the Fornell criterion, the square root of the AVE values of each first-order construct is greater than the correlations of other constructs, validating the satisfactory discriminant validity among the first-order constructs of environmental regulation constructed in this study.

The results in

Table 6. Reliability and convergent validity analysis.

Higher-Order Construct	First-Order Construct	Observed Variable	Factor Loading	Cronbach’s Alpha	C.R.	AVE
CEP	ECU	ECU1	0.797	0.755	0.858	0.668
		ECU2	0.786
		ECU3	0.866
	EP	EP1	0.850	0.841	0.896	0.687
		EP2	0.914
		EP3	0.876
		EP4	0.650
	ESD	ESD1	0.628	0.855	0.906	0.710
		ESD2	0.918
		ESD3	0.903
		ESD4	0.889

demonstrate that the factor loading coefficients of each observed variable for circular economy performance are all above 0.60, which meets the requirements of the study, and the basic fit of the model is good. Cronbach’s Alpha coefficient was used to test the internal consistency of the measurement indicators, and this study obtained that the Cronbach’s Alpha coefficients of each first-order construct are all above 0.8, with the combined reliability (CR) values exceeding 0.7, which proves a high level of reliability.

As shown in Table 6, the average variance extract (AVE) of each construct is greater than 0.5, which indicates the model has high convergent validity. As shown in

Table 7. Discriminant validity analysis.

	ECU	EP	ESD
ECU	0.817
EP	0.727	0.829
ESD	0.690	0.680	0.843

, according to the Fornell criterion, the square root of the AVE values of each first-order construct is greater than the correlations of other constructs, validating the satisfactory discriminant validity among the first-order constructs of circular economy performance constructed in this study.

Based on the test and evaluation of the higher-order measurement model for environmental regulation and circular economy performance as mentioned above, it is confirmed that each lower-order construct demonstrates good explanatory power and reliability concerning its corresponding higher-order construct. This validation underscores the effectiveness of constructing a measurement framework for assessing environmental regulation and circular economy performance from a multidimensional perspective.

4.2. Measurement Model Results

The measurement model serves as a test of the relationship between latent variables and observed variables in the overall model, reflecting the intrinsic quality of the model. According to the results presented in

Table 8. Reliability and convergent validity analysis.

Latent Variable	Observed Variable	Factor loading	Cronbach’s Alpha	CR	AVE
ER	EO	0.846	0.894	0.931	0.818
	EI	0.937
	PP	0.928
ISU	IS1	0.811	0.797	0.878	0.706
	IS2	0.802
	IS3	0.905
CEP	ECU	0.886	0.874	0.922	0.798
	EP	0.892
	ESD	0.902

, the factor loadings of each measurement variable are all above 0.60, while the Cronbach’s Alpha coefficients of each latent variable exceed 0.8, and the composite reliability (CR) values are all above 0.7, demonstrating high levels of reliability.

The purpose of validity testing is to ascertain whether the observed variables accurately measure the latent variables, encompassing convergent and discriminant validity. As shown in Table 8, the AVE values of each construct exceed 0.5, indicating high levels of convergent validity. As shown in

Table 9. Discriminant validity analysis.

	ER	ISU	CEP
ER	0.905
ISU	0.309	0.840
CEP	0.650	0.753	0.893

, the square roots of the AVE values are greater than the correlations between the various constructs, thereby there is good discriminant validity among the latent constructs in the measurement model of this paper.

4.3. Structural Model Results

The bootstrapping method is used to test the structural model constructed in this paper. Path coefficients elucidate the extent and direction of the impact between latent variables. The results displayed in

Table 10. Path coefficient and hypothesis testing.

Path	Path Coefficient	T-Value	p-Value	Hypothesis	Support
ER→CEP	0.461 ***	18.840	0.000	H1	Yes
ER→ISU	0.309 ***	8.819	0.000	H2	Yes
ISU→CEP	0.611 ***	23.832	0.000	H3	Yes

Notes: Significance levels are expressed as *** for 1%. These notation conventions apply consistently to subsequent tables.

show that the path coefficients of environmental regulation on industrial structure upgrading and circular economy performance are 0.309 (t = 8.819) and 0.461 (t = 18.840), respectively, validating hypotheses H1 and H2. These findings suggest that environmental regulations can facilitate the upgrading of industrial structures and significantly promote the development of the circular economy. Furthermore, the path coefficient from industrial structure upgrading to circular economy performance is 0.611 (t = 23.832), validating hypothesis H3, which indicates that the industrial structure upgrading can significantly enhance circular economy performance.

4.4. Mediating Effect Analysis

The purpose of the mediating effect analysis is to ascertain the capacity of an independent variable to exert influence on a dependent variable through mediator variables. Variance Accounted For (VAF) value serves as a metric for discerning the nature of mediating effect, with its magnitude reflecting the proportion of indirect effects within the overall effect [74]. When the VAF is lower than 0.2, it means that there is no mediator; when the VAF is between 0.2 and 0.8 it means that there is a partially mediated effect; and when the VAF is greater than 0.8 it means that there is complete mediation. The results in

Table 11. Mediating effect test.

	Path	T-Value	p-Value	95% CI		Hypothesis	Support
Direct effect
ER→CEP	0.461 ***	18.840	0.000	[0.417;0.506]		H1	Yes
ER→ISU	0.309 ***	8.819	0.000	[0.233;0.368]		H2	Yes
ISU→CEP	0.611 ***	23.832	0.000	[0.556;0.653]		H3	Yes
Indirect effect					VAF
ER→ISU→CEP	0.189 ***	9.242	0.000	[0.144;0.224]	0.291	H4	Yes
Total effect
ER→CEP	0.650 ***			[0.592;0.698]

Notes: Significance levels are expressed as *** for 1%. These notation conventions apply consistently to subsequent tables.

elucidate that the impact of environmental regulations on the performance of the circular economy can be channeled through the industrial structural upgrading, yielding an indirect effect of 0.189, and the VAF value of 0.291, which proves the existence of a partial mediation effect; hypothesis H4 is supported.

The structural path diagram of the impact of environmental regulation on the circular economy performance is shown in Figure 2:

4.5. Heterogeneity Analysis

The regional heterogeneity in the impact of environmental regulation on circular economy performance can be examined through the establishment of a multi-group model for analysis. Based on the varying levels of economic development across different regions, the 30 provinces of China were categorized into three groups: high, medium, and low levels of economic development. The results of the analysis are shown in

Table 12. Heterogeneity test.

Path	Path Coefficient			p-Value
	(1) High Level	(2) Medium Level	(3) Low Level	(1) vs. (2)	(1) vs. (3)	(2) vs. (3)
ER→CEP	0.429 ***	0.685 ***	0.745 ***	0.003	0.001	0.254

Notes: Significance levels are expressed as *** for 1%. These notation conventions apply consistently to subsequent tables.

. In regions with high, medium, and low levels of economic development, the path coefficients of environmental regulations on the circular economy are 0.429, 0.685, and 0.745, respectively. This indicates that in regions with lower levels of economic development, environmental regulation has a greater impact on the development of the circular economy.

The p-values for the differences between groups are less than 0.05, which represents the existence of heterogeneity at the 5% significance level, so according to Table 12, significant differences exist among regions with high economic development and those with medium and low development regarding the impact of environmental regulations on circular economy performance. Therefore, hypothesis H5 is supported.

The possible reasons for the above results may lie in the comparatively outdated production methods and technologies in regions with medium to low economic development, in contrast to those with higher economic levels. Consequently, these regions exhibit lower production and energy conversion efficiencies, leading to more pronounced instances of resource wastage and environmental pollution. Therefore, the marginal constraining effects of environmental regulation in these areas are more pronounced, which will exert greater pressure on enterprises to intensify the effect toward waste disposal, adopt sustainable development strategies, and enforce the intensity of the crackdown against activities such as illegal emissions, inadequate environmental infrastructure, and significant energy wastage, which contradict the concept of circular economy development. Moreover, in economically underdeveloped regions, environmental awareness and the level of green innovative technologies are relatively low, reflecting a lower starting point for development. Consequently, the introduction of stringent environmental regulations necessitates significant adjustments and improvements in the production and lifestyle practices of enterprises and residents, in order to comply with regulatory requirements, thus leading to comparatively higher benefits in terms of enhancing efficiency in the circular economy.

In reality, economic sustainability may be undermined by the excessive extraction of natural resources and significant burden of pollution load. Gansu province is located in the northwest arid and semiarid areas in China, and its economic development level is much lower than the national average, belonging to the region of low economic development level. Wang (2015) [75] points out in his study that Gansu province is in the middle stage of industrialization, and the pressure of production on environmental damage is high. However, through the implementation of a series of environmental policies focused on ecological construction projects, as well as the introduction of advanced production technology, the upgrading and adjustment of its industrial structure has accelerated. Particularly considering that the industrial structure of Gansu province exhibits typical characteristics of high resource consumption and high pollution, powerful environmental regulation and technological advancements have great potential to improve environmental quality in low-income economy areas, realizing the harmony of ecology and economy. Therefore, Gansu province should strengthen its ecological protection via government and public support, and fulfil the function of market mechanism, in order to promote a sustainable economic development strategy.

5. Discussion

Initially, a two-stage approach is employed to assess and analyze the two higher-order constructs in this study. In the first stage, a measurement model for environmental regulation, as a higher-order construct, is constructed from three dimensions (first-order construct): administrative directives, market incentives, and public participation. Simultaneously, a measurement model for circular economy performance, as another higher-order construct, is constructed from three dimensions (first-order construct): energy consumption and utilization, environmental pollution, and economic and social development. The results of the higher-order measurement model test indicate satisfactory reliability and validity for the measurement items of each first-order construct, which proves the effectiveness of constructing the measurement system for assessing environmental regulation and circular economy performance from multidimensional perspectives, thereby mitigating the shortcomings of some previous studies that relied solely on a single indicator for measurement.

Subsequently, utilizing the scores obtained from the first-order constructs is the first stage to downgrade the corresponding higher-order constructs, followed by testing the structural equation modeling established in this study. The results show that the path coefficient of ER→CEP is 0.461, indicating a direct facilitating effect of environmental regulations on circular economy performance, and the path coefficient of ER→ISU→CEP is 0.189, indicating that the environmental regulation can indirectly promote the circular economy performance through the intermediary of industrial structure upgrading.

Environmental regulations exert normative and regulatory constraints on behaviors detrimental to ecological integrity, such as excessive emissions and inadequate hazardous waste management, through intensified supervision, management, and punitive measures. Additionally, they stimulate voluntary attention and monitoring by fostering public environmental consciousness. Through administrative directives, environmental regulations normatively guide enterprises to comply with environmental standards, thereby mitigating actions harmful to the ecosystem. Furthermore, by internalizing the costs of pollution through mechanisms like environmental protection taxes and emission charges, environmental regulations induce enterprises to adjust their energy consumption structures and invest in technological innovation for energy conservation and environmental protection. Consequently, this facilitates not only the reduction of environmental pollution costs but also the attainment of higher production efficiency and economic benefits. This integration of market economy principles with ecological imperatives enhances circular economy performance, fostering sustainable development.

The transformation and upgrading of industrial structures under environmental regulations entail a shift towards more technologically intensive industries and the relocation of resources to sectors characterized by higher efficiency and environmental friendliness. This transition augments the benefits of economic sustainability and ecological civilization, thereby promoting the development of the circular economy.

6. Conclusions and Implication

6.1. Conclusions

Based on the panel data of 30 provinces in China from 2011 to 2020, this paper establishes an evaluation index system for environmental regulations and circular economy performance by employing structural equation modeling to construct the impact mechanism of environmental regulation on circular economy performance, and conduct empirical analysis. The main conclusions are as follows: environmental regulation has a significant positive impact on circular economy performance; industrial structure upgrading has a mediating role in the path of environmental regulation on circular economy performance; the heterogeneity study shows that the mechanism of environmental regulation on circular economy performance varies according to the level of regional economic development. The heterogeneity study reveals variations in the mechanism through which environmental regulations affect circular economy performance due to disparities in regional economic development levels.

6.2. Policy Recommendations

First, environmental regulatory policies should be adjusted appropriately to foster a harmonious equilibrium between the economy and the environment. Given the existence of regional disparities, targeted formulation and the amalgamation of various environmental regulatory policies are necessary to circumvent adverse impacts resulting from blindly increasing or relaxing the intensity of environmental regulation. Therefore, regional governments need to scientifically and systematically assess the actual development situation, implement differentiated environmental regulatory policies to oversee and rectify polluting behaviors, and gradually refine measures such as environmental expenditure and investments in pollution control, in order to moderately alleviate the burden of enterprises transitioning towards greener practices. Initiatives to establish carbon emissions trading and pollutant discharge rights markets should be explored to enhance market-regulated capacity, prevent the escalation of additional costs for enterprises, and promote the development and utilization of solar energy, wind power, and other clean, low-carbon, renewable energy, thereby injecting fresh impetus into the advancement of the circular economy.

Secondly, it is imperative to enrich environmental regulatory policies to propel industrial transformation and upgrading. Environmental regulations encompass various types, which entail measures such as administrative penalties, environmental monitoring, taxation, and pollution reporting and complaints. These measures can complement and coordinate with each other to form a multi-agent, systematic environmental monitoring and management system, collectively advancing industrial structural transformation and upgrading. Enhancements in taxation and penalties for highly polluting industries, coupled with subsidies for environmental clean industries, compel enterprises to assume clear responsibilities and address pollution behaviors at its source. By harnessing the stimulating effect of environmental regulations on enterprise innovation, companies are encouraged to take proactive measures in green production, thereby accelerating the pace of technological innovation, and fostering a cleaner and more efficient industrial development pattern. Expanding the scope of environmental information disclosure, establishing convenient channels for pollution reporting, and exposing corporate pollution behaviors are essential steps. Conducting public hearings, citizen voting, and other forms of public participation in environmental governance intensifies public engagement, and enhances the fairness and transparency of environmental governance, combining with the collective strength of public entities to regulate corporate production behaviors.

Thirdly, enhancing regional collaborative governance is vital to fostering the development of the circular economy. According to the “Pollution Haven Hypothesis”, inter-regional cooperation is required to prevent certain regions from becoming havens for outdated and polluting industries. Strengthening cooperation and coordination among diverse regions facilitates resource sharing and complementary advantages, thereby promoting mutual industrial development and synergies, that will drive the development and construction of industrial chains within the circular economy. Utilizing digitization, cloud computing, and other technologies to establish cross-regional platforms for environmental information sharing is essential. These platforms facilitate the timely dissemination of effective environmental regulatory policies and pollution control experiences. Regional governments engage in joint monitoring, law enforcement, and punitive actions against transboundary pollution to prevent polluting enterprises from evading regulations through production relocation. Furthermore, promoting investments in cross-regional environmental governance and research and development of environmental protection technologies optimizes industrial structures, enhances ecological benefits, and elevates the overall performance of the circular economy.

6.3. Limitations and Prospects

The scope of research warrants further expansion. Due to issues such as data scarcity and statistical interruptions, this paper only encompasses data from 2011 to 2020, with a focus limited to 30 provinces. Therefore, future research endeavors may extend the time series and delve into more targeted investigations to enhance the specificity of the findings.

Due to variations in the implementing entities and methods of environmental regulations, future research can classify them to explore the mechanisms through which heterogeneous environmental regulations affect circular economy performance. This entails delving into the characteristics and specific pathways through which different types of environmental regulations promote the growth of circular economy performance, along with conducting comparative analyses. Simultaneously, in-depth investigation into the mediating or regulating variables through which environmental regulations impact circular economy performance should be conducted. These will enrich and refine the framework illustrating the influence mechanisms of environmental regulations on circular economy performance, thereby furnishing theoretical support and policy recommendations for further promoting the development of the circular economy. In a future study, with the completion of data collection, the measurement dimensions of circular economy performance can be further enriched to improve the comprehensiveness of the measurement index system.