Green Revolution vs. Digital Leap: Decoding the Impact of Environmental Regulation on New Quality Productive Forces in China’s Yangtze River Basin

Abstract

The development of New Quality Productive Forces (NQPF), fueled by simultaneous progress of informatization, digitalization, and ecologization, creates a transformative sustainability framework that connects economic growth and environmental protection. People usually think that environmental regulation enhances regional ecologization, thereby boosting total NQPF. Does this hold true for China’s Yangtze River Basin? Utilizing panel data from 2015 to 2022, this study examines the impact of environmental regulation on NQPF across 86 prefecture-level cities in the basin. Our empirical results corroborate that environmental regulation exerts a statistically significant positive effect on digital NQPF development, which in turn contributes substantially to overall NQPF enhancement—This finding remains robust across alternative estimation methods. Our analysis further identifies three primary mechanisms driving this effect: industrial upgrading, technological innovation, and GDP growth. The effect is nonlinear and characterized by a threshold: in less developed areas, environmental regulation somewhat helps, whereas in more developed regions, reaching a certain strength significantly enhances both digital and overall productivity. Furthermore, environmental regulation demonstrates notable spillover effects: they enhance local outcomes while simultaneously improving digital and overall NQPF in neighboring regions. These findings offer strong evidence and valuable policy insights for advancing the digital transformation and high-quality sustainable development of the Yangtze River Basin.

1. Introduction

Striking a sustainable balance between ecological protection and economic growth now stands as a critical imperative for nations worldwide. As a crucial region for China’s economic growth, the Yangtze River Basin has garnered significant attention for its ecological condition and economic transformation. In recent years, the basin has contributed over 40% of the nation’s GDP while covering just 20% of its land area [1]. Its cumulative investment in ecological management has surpassed CNY 700 billion [2]. In 2023, water quality assessments of the Yangtze River’s main streams and tributaries were overwhelmingly positive, with 98.5% of sections rated between Class I and III. Moreover, 2441 shoreline utilization projects suspected of legal violations were identified and rectified along the main streams, resulting in the restoration of 162 km of shoreline. Additionally, monitoring efforts recorded 227 species of indigenous fish in the basin—an increase of 34 species compared to 2022 [3]. These achievements underscore the strategic importance of prioritizing ecological preservation and green development for regional economic transformation. On the one hand, the basin has invested over CNY 700 billion in ecological management, with 98.5% of its water sections meeting high-quality standards [2,3], underscoring the effectiveness of environmental policies. On the other hand, understanding how environmental regulation influences NQPF thus holds the key to unlocking a “win-win” scenario: stricter regulation could accelerate the transition to sustainable, innovation-driven growth, while weak or misaligned policies might stifle productivity. General Secretary Xi Jinping’s emphasis on advancing the green development of the Yangtze River Economic Belt underscores the critical importance of examining the relationship between environmental policies and economic dynamics in the region [4]. As a fundamental engine of modern economic development, New Quality Productive Forces (NQPF)—characterized by the systemic integration of scientific innovation, green technologies, and digital transformation [5]—serve as a transformative catalyst for both economic restructuring and ecological sustainability [6,7,8,9,10]. This advanced productive paradigm demonstrates multidimensional impacts: it significantly accelerates rural transformation through technological empowerment [7], enhances agricultural green production efficiency through sustainable innovation [8], improves sustainable development of corporate by strict environmental rules [9], and fosters synergistic development with forestry economic resilience to establish China’s global leadership in sustainable forestry practices [10]. This progression enables the Yangtze River Basin to overcome the limitations of traditional industries and achieve the optimization and upgrading of its economic framework. An in-depth study of the impact of environmental regulation on the NQPF of the Yangtze River Basin would shed light on how environmental policies can drive green transformation and high-quality economic development in the region. Such research would provide a solid foundation for precise policy formulation and support the realization of a win–win outcome that balances economic advancement with environmental protection [11].

Research on the direct impact of environmental regulation on NQPF remains limited. However, numerous scholars have examined how environmental regulations affect various components or dimensions of NQPF. For example, Hao et al. found that China’s green credit policy substantially improves the quality of green innovation [12]. Li et al. showed that green credit policy fosters an effective synergy between environmental regulation and financial resource allocation, significantly promoting the development of NQPF in resource-based enterprises [13]. Using a quasi-natural experiment, a study reveals the positive effects of China’s aggregate emission control policy on green technology innovation [14]. Additionally, some scholars suggest that environmental regulations influence NQPF through various mechanisms, such as the compensation effect of technological innovation and the industrial linkage effect, which collectively drive enterprise innovation, upgrading, and the advancement of NQPF. Liu and He investigated how collaborative agglomeration facilitates high-quality development in manufacturing and enhances NQPF from the perspectives of industrial structure upgrading and innovation ecosystem construction [15]. Similarly, Zou and Wang analyzed the role of environmental regulation in China’s energy transition, offering another perspective on the connection between environmental regulation and NQPF [16]. Li et al. examined the impact of environmental regulation on NQPF from a provincial perspective, using environmental regulation policy as a starting point and employing a two-way fixed effects model [13]. Additionally, Zhang et al. analyzed city-level cases and demonstrated that market-based and voluntary regulations are more effective than mandatory directives in promoting green innovation [17], revealing a nonlinear inverted U-shaped relationship between environmental regulations and green innovation efficiency [18].

Some studies have also examined the varying relationship between environmental regulation and NQPF from a micro-level firm perspective. Wang et al. employed multivariate modeling with firm-level data to assess the role of environmental regulation in enhancing NQPF, highlighting differences related to property rights and firms’ resource bases [19]. Focusing specifically on heavily polluting firms and the heterogeneity among them, Guo et al. found that environmental regulation significantly stimulates green innovation within these enterprises [20].

Despite the abundance of existing research, several limitations remain. On the one hand, most studies concentrate on the provincial or enterprise level, with relatively few offering detailed analyses at the prefecture-level city scale—particularly for cities within the Yangtze River Basin. This gap hinders a precise understanding of both the short- and long-term effects, as well as spatial variations within the region. While Zhang et al. provide city-level case studies, their coverage does not systematically encompass multiple prefecture-level cities across the Yangtze River Basin [17]. On the other hand, although theoretical discussions have elaborated on the mechanisms through which environmental regulations influence NQPF, empirical analyses remain insufficiently thorough and comprehensive. Mao and Wang explore this issue from the perspectives of technological innovation and environmental concern but fail to sufficiently examine other potential pathways [21]. Similarly, Guo et al. note that existing studies have yet to fully uncover the mechanisms by which environmental regulations impact green innovation [20]. Zhou et al. use a difference-in-differences approach to identify the effect of environmental regulations on green innovation, but do not examine the role of mediating mechanisms [22]. Moreover, from a methodological standpoint, most prior research does not fully leverage advanced measurement techniques to address complexities such as threshold effects, limiting the ability to comprehensively and accurately capture the nuanced relationship between environmental regulation and NQPF, which in turn affects the robustness and reliability of the findings. For instance, Li et al. do not detail their approach to handling complex issues like threshold effects in their empirical work, underscoring a lack of methodological sophistication in current studies [23]. Fu focuses on the influence of data factor allocation on NQPF but similarly falls short of investigating strategies to manage threshold effects [24]. Together, these shortcomings highlight the need for more comprehensive research to better illuminate the intricate relationship between environmental regulation and NQPF.

This study examines 86 prefecture-level cities in the Yangtze River Basin, using panel data from 2015 to 2022 to conduct a comprehensive analysis of the impact of environmental regulations on NQPF. By measuring the level of NQPF, it investigates the direction, key driving forces, channels of influence, nonlinear effects, and spatial spillover effects of environmental regulations. Furthermore, the study uncovers the underlying mechanisms at work and validates the robustness of its findings.

The marginal contributions of this study can be summarized in three key points. This paper is the first to comprehensively investigate, through theoretical analysis and multi-model empirical evidence, which specific dimension of NQPF in the Yangtze River Basin is predominantly reshaped by environmental regulations. The study delivers consistent and robust conclusions, making a valuable contribution to existing research. While previous studies have primarily focused on the aggregate index of NQPF, they have not identified which particular dimension within a region is most significantly influenced by environmental regulation and thus drives the overall index improvement [13,15,16,19,25,26,27]. Moreover, the impact of environmental regulation on the three dimensions of NQPF exhibits heterogeneity across different geographical contexts. Consequently, the findings of this study offer important insights for advancing NQPF in comparable regions.

Second, robust empirical evidence in this study contradicts perceptual judgments, with mechanisms and implications systematically examined. Existing research has preliminarily explored how environmental regulation ultimately enhances overall NQPF by promoting green technological innovation [14,21]. Building on this, our study adds the effects of industrial structure upgrading and GDP growth. Stringent environmental regulation raises compliance costs for high-pollution industries, incentivizing their exit or transformation, while fostering the growth of low-carbon, high-tech sectors. This structural shift creates a more favorable ecosystem for NQPF, as emerging industries are more receptive to digital and green technologies. And effective environmental regulation redirects resources toward efficient, eco-friendly sectors, boosting sustainable GDP growth. Higher GDP provides the capital for digital infrastructure like 5G base stations and R&D investment, which are foundational to NQPF. Furthermore, by decomposing overall NQPF, the study finds that environmental regulation primarily boosts digital NQPF through various mediating variables, which in turn drives the improvement of overall NQPF—this conclusion is supported by robust tests of the mediation effects. The findings are innovative and differ from traditional views that environmental regulation enhances green NQPF, thereby boosting overall NQPF, offering fresh insights.

Third, in terms of methodology, this study employs various models and comprehensive data analysis to enhance the reliability and robustness of the findings. Recognizing that most existing studies rely on traditional econometric models that may inadequately address threshold and spatial effects, this research adopts a threshold regression model to test for potential threshold effects in the relationship between environmental regulation and NQPF, and a spatial Durbin model to innovatively explores the spatial impacts and spillover effects of environmental regulation on NQPF by incorporating a spatial weighting matrix to capture spatial correlations and proximity among prefecture-level cities. The linear model, threshold model, and spatial Durbin model all indicate that the significant impact of environmental regulation on NQPF is primarily reflected in its effect on digital NQPF, a conclusion that has not been confirmed in previous studies.

The following sections of the paper are organized as follows. Section 2 combines the existing literature, conducts theoretical analysis, and proposes hypotheses. Section 3 introduces a detailed description of the variables, data, and research methods. Section 4 conducts base regression analysis, robustness tests, mechanism tests, heterogeneity tests, and further exploration. Section 5 summarizes the research conclusions and puts forward policy recommendations.

2. Theoretical Analysis and Research Hypothesis

2.1. Environmental Regulation and NQPF

The core logic behind the direct impact of environmental regulation on NQPF is that such regulation can alter the production orientation and cost structure of enterprises, thereby motivating them to pursue technological innovation and changes in production methods to achieve a win–win outcome for both economic and environmental benefits. Specifically, when the government enforces environmental regulation policies, firms face stricter environmental standards and lower pollution emission limits, which incentivize them to enhance production efficiency and reduce unit product costs through technological advancements and digital management optimization, ultimately maintaining or strengthening their market competitiveness. Zhang et al. find that market-based and voluntary regulations are more effective than mandatory regulations in stimulating green innovation [17]. Furthermore, referring to [23], it is noted that digital transformation is influenced significantly by environmental regulation. Green and technological breakthrough as well as digital transformation is a core aspect of NQPF [28,29,30]. In comparison with green and technological breakthrough dimension, the high permeability and synergy of digital technologies facilitate the rapid integration of clean energy and optimization of production processes, directly boosting resource efficiency and innovation output. In contrast, the green and technological dimensions depend more on long-term R&D investments and infrastructure upgrades, leading to slower short-term responses. Consequently, environmental regulations tend to more strongly stimulate the digital dimension—such as intelligent emission reduction and circular economy platforms—unlocking efficiency gains that drive significant improvements in overall NQPF. Based on the above analysis, this paper proposes Hypothesis 1.

Hypothesis 1 (H1):

Environmental regulation can effectively enhance the level of digital NQPF, ultimately improving overall NQPF.

2.2. Mechanisms of Environmental Regulation Affecting NQPF

Drawing on the existing study, the implementation of environmental regulation policies plays a crucial role in promoting industrial structure upgrading [31,32,33,34], which is a key pathway to enhancing NQPF. Under the constraints of environmental regulations, traditional high-pollution, low-efficiency industries face increasing survival pressures, while emerging sectors such as high-tech industries and modern service industries enjoy greater development opportunities due to their low pollution levels and high value-added characteristics. The transformation and upgrading of traditional industries, coupled with the rise in new industries, jointly drive the optimization and advancement of industrial structure. This industrial upgrading enhances NQPF through multiple channels. On the one hand, emerging industries generally possess stronger technological innovation capabilities and more flexible organizational structures. Their growth stimulates technological progress and boosts production efficiency in related sectors, thereby elevating the overall scientific and technological level of NQPF across the economic system. On the other hand, traditional industries actively adopt new technologies, processes, and equipment during their transformation and upgrading to pursue green and digital transitions. The primary reason industrial transformation and upgrading mainly enhance the digital dimension of new quality productivity is that digital technologies—including big data analytics, artificial intelligence, cloud computing, and the Internet of Things—enable more precise, efficient, and flexible production processes. These digital tools facilitate real-time, data-driven decision-making, optimize resource allocation, and improve coordination within and across production chains. Consequently, digital upgrading serves as a key enabler for simultaneously reducing waste, cutting emissions, and improving product quality. This digital advancement in traditional industries not only helps mitigate environmental pollution but also enhances production efficiency and product quality, achieving a green and digital upgrade in their NQPF. Through this dual pathway—innovation-driven growth in emerging industries and digital-enabled transformation in traditional ones —the overall new quality productivity of the economy is significantly improved, supporting sustainable and high-quality development goals.

Second, drawing on the studies of [19,35,36,37], environmental regulation “pushes” enterprises to engage in green technological innovation, which in turn drives the development of NQPF. Thus, the role of environmental regulation in fostering technological innovation is a key mechanism influencing the advancement of NQPF. Strict environmental regulations mandate that enterprises reduce pollution emissions, compelling them to increase investments in environmental technology research and development, the adoption of cleaner production technologies, and related areas. To meet these regulatory requirements, enterprises must continuously explore innovative technological solutions. This not only stimulates their own technological innovation—enhancing competitiveness and economic efficiency—but also drives technological progress across the entire industry. Such advancements foster the growth of related sectors and generate technological spillover effects that further accelerate the enhancement of scientific and digital NQPF. Moreover, environmental regulatory policies effectively encourage enterprises to increase R&D investments and strengthen independent innovation capabilities by establishing technological standards and innovation incentives, providing robust support for the growth of scientific and digital NQPF. Green technology innovation plays a pivotal role in driving these improvements by leveraging advanced digital tools—such as big data analytics, the Internet of Things (IoT), and artificial intelligence (AI)—to optimize resource use, monitor environmental performance in real time, and enable smarter, cleaner production processes. By integrating digital technologies into green innovation, enterprises can simultaneously achieve higher efficiency and sustainability, significantly contributing to the overall enhancement of new quality productivity. This digital foundation ensures that green innovations transcend isolated technical advances, transforming into systemic productivity gains across industries and ultimately fostering comprehensive improvements in both economic and environmental performance.

Thirdly, environmental regulatory policies can foster sustainable economic development by steering resource allocation toward efficient and eco-friendly sectors. An increase in GDP creates a favorable macroeconomic environment that supports the enhancement of NQPF. As GDP grows, enterprises acquire greater capital and capacity to pursue digital transformation and innovation, while governments are able to increase investments in education, scientific research, and infrastructure development. These collective efforts provide a strong material foundation and intellectual support for advancing digital NQPF. Furthermore, GDP growth heavily depends on the progress of digital new quality productivity, as digital technologies drive efficiency, innovation, and value creation across industries. By strengthening digital capabilities, enterprises can optimize processes, develop innovative products, and better respond to market demands, thereby significantly raising the overall level of new quality productivity. This virtuous cycle ensures that improvements in digital productivity serve as the primary engine for sustained economic growth and development. Based on this analysis, this paper proposes Hypothesis 2.

Hypothesis 2 (H2):

Environmental regulation can influence digital NQPF and overall NQPF by promoting industrial structure upgrading, facilitating technological innovation, and enhancing GDP levels.

2.3. Threshold Effects of Environmental Regulations on NQPF

By referring to [23,32,38], it has been demonstrated that there is a “U”-shaped relationship between environmental regulation and NQPF. This indicates that the impact of environmental regulation on NQPF is nonlinear and subject to a threshold effect: environmental regulation can enhance NQPF up to a certain point, but beyond this threshold, it begins to inhibit productivity growth. Analyzing the reasons behind this threshold effect, when the intensity of environmental regulation is low, the cost pressures on enterprises are relatively minor. In this stage, environmental regulation primarily promotes technological research and development and encourages changes in production methods by incentivizing innovation, thereby boosting NQPF. However, once the intensity of regulation surpasses a certain threshold, the cost burden on enterprises increases significantly, which in turn suppresses the development of NQPF. Guo et al. also find that environmental regulation significantly influences green innovation among heavily polluting enterprises [20]. When regulation intensity is moderate, green innovation effects are more pronounced, especially in firms with weaker internal and external governance, further underscoring the threshold nature of environmental regulation’s impact. Moreover, because enterprises have limited capacities for technological innovation and absorption, moderate levels of environmental regulation enable them to effectively leverage internal resources and external technology spillovers to drive innovation and transform production modes. Conversely, excessively stringent regulations overwhelm their resource capacity, reducing innovation efficiency and hindering improvements in NQPF. Additionally, environmental regulation’s influence on industrial restructuring also exhibits a threshold effect. Moderate regulation can facilitate the transformation and upgrading of high-pollution, low-efficiency traditional industries while fostering the growth of emerging sectors such as high-tech and modern services, thereby optimizing industrial structure and enhancing NQPF. However, excessively strong regulation may accelerate the decline of traditional industries without sufficient growth in new sectors to compensate, leading to structural imbalances and inhibiting NQPF development. Based on the above analysis, this paper proposes Hypothesis 3.

Hypothesis 3 (H3):

The impact of environmental regulation on NQPF has a threshold effect.

3. Empirical Model and Variables

3.1. Model Building

In order to elucidate the impact of environmental regulation on NQPF, Hypothesis 1 is tested and a panel data benchmark regression model is constructed:

$${\mathrm{N}\mathrm{e}\mathrm{w}\mathrm{P}\mathrm{r}\mathrm{o}\mathrm{d}}_{it}={\alpha }_0+{\alpha }_1{\mathrm{E}\mathrm{n}\mathrm{v}\mathrm{R}\mathrm{e}\mathrm{g}}_{it}+\sum {\alpha }_i{\mathrm{X}}_{it}+{\mu }_i+{\delta }_t+{\epsilon }_{it},$$

(1)

where ${\mathrm{N}\mathrm{e}\mathrm{w}\mathrm{P}\mathrm{r}\mathrm{o}\mathrm{d}}_{it}$ is the level of NQPF in year t in prefecture i; ${\mathrm{E}\mathrm{n}\mathrm{v}\mathrm{R}\mathrm{e}\mathrm{g}}_{\mathrm{i}\mathrm{t}}$ is the environmental regulation in year t in prefecture i; ${\mathrm{X}}_{it}$ is each control variable; ${\alpha }_0$ is the intercept term; ${\alpha }_1\mathrm{a}\mathrm{n}\mathrm{d}{\alpha }_i$ denote each regression coefficient to be estimated; ${\mu }_i$ and ${\delta }_t$ denote prefecture and year fixed effects, respectively; ${\epsilon }_{it}$ denotes the random disturbance term.

In order to explore the indirect effect of environmental regulation on NQPF and to test Hypothesis 2, a mediation effect model is constructed based on Equation (1) with reference to Baron and Kenny [39]:

$${\mathrm{M}}_{it}={\beta }_0+{\beta }_1{\mathrm{E}\mathrm{n}\mathrm{v}\mathrm{R}\mathrm{e}\mathrm{g}}_{it}+\sum {\beta }_i{\mathrm{X}}_{it}+{\mu }_i+{\delta }_t+{\epsilon }_{it},$$

(2)

where ${\mathrm{M}}_{it}$ represents the mediating variable; ${\beta }_0$ represents the intercept term; ${\beta }_1$ represents the effect of NQPF on the mediating variable; and ${\beta }_i$ denotes regression coefficient to be estimated.

To test research Hypothesis 3, panel data threshold regression model proposed by Hansen is drawn upon to construct a panel threshold model with Ln (GDP) as the threshold variable [40], respectively:

$$\begin{array}{c}{\mathrm{N}\mathrm{e}\mathrm{w}\mathrm{P}\mathrm{r}\mathrm{o}\mathrm{d}}_{it}={\phi }_0+{\phi }_1{\mathrm{E}\mathrm{n}\mathrm{v}\mathrm{R}\mathrm{e}\mathrm{g}}_{it}\times I\left({th}_{it}<\theta \right)+{\phi }_2{\mathrm{E}\mathrm{n}\mathrm{v}\mathrm{R}\mathrm{e}\mathrm{g}}_{it}\times I\left({th}_{it}\ge \theta \right)\\ +\sum {\phi }_i{\mathrm{X}}_{it}+{\mu }_i+{\delta }_t+{\epsilon }_{it},\end{array}$$

(3)

where ${th}_{it}$ represents the Ln (GDP) of the i-th prefecture-level city in year t; $\theta$ represents the threshold value; $I$ is the indicator function, which takes the value of 1 if the conditions in parentheses are met, and 0 otherwise; ${\phi }_1$ and ${\phi }_2$ represent the impact of environmental regulations on NQPF when the threshold variable lies in different interval; ${\phi }_0$ represents the intercept term; and ${\phi }_i$ is the regression coefficient to be estimated. Equation (3) is for the single-threshold case and can be expanded to the multi-threshold case based on an econometric test of the sample data.

Given the potential spatial correlation between environmental regulation and new-type productive forces across prefecture-level cities, this study incorporates spatial factors into the examination of the relationship between environmental regulation and new-type productive forces [41]. By combining a spatial weight matrix with relevant variables, a spatial econometric model is constructed. Panel spatial models primarily include spatial lag models (SLM), spatial error models (SEM), and spatial Durbin models (SDM). Compared to the other two models, the spatial Durbin model better reflects the influence of spatial lag factors of the dependent and independent variables on the dependent variable. Therefore, this paper establishes a spatial Durbin panel model to describe the spatial impact of environmental regulation on new-type productive forces [42,43,44,45] as follows:

$$\begin{array}{c}{\mathrm{N}\mathrm{e}\mathrm{w}\mathrm{P}\mathrm{r}\mathrm{o}\mathrm{d}}_{it}={\beta }_0+\rho {\displaystyle \sum _{j=1}^n}{w}_{ij}{\mathrm{N}\mathrm{e}\mathrm{w}\mathrm{P}\mathrm{r}\mathrm{o}\mathrm{d}}_{jt}+{\beta }_1{\mathrm{E}\mathrm{n}\mathrm{v}\mathrm{R}\mathrm{e}\mathrm{g}}_{it}+{\beta }_2{\mathrm{X}}_{it}+{\theta }_1{\displaystyle \sum _{j=1}^n}{w}_{ij}{\mathrm{E}\mathrm{n}\mathrm{v}\mathrm{R}\mathrm{e}\mathrm{g}}_{jt}\\ +{\theta }_2\sum _{j=1}^n{w}_{ij}{\mathrm{X}}_{jt}+{\mu }_i+v_t+{\epsilon }_{it},{\epsilon }_{it}=\sigma \sum _{j=1}^n{w}_{ij}{\epsilon }_{jt}+{\phi }_{it},\end{array}$$

(4)

where $w_{ij}$ is the spatial weight matrix; β₁ and β₂ denote the direct effects of the explanatory and control variables; θ₁ and θ₂ denote the spatial spillover effects of the explanatory and control variables; ${\mu }_i$ is the individual effect; $v_t$ is the time effect; ${\epsilon }_{it}$ is the random error term; and ρ is the spatial autocorrelation regression coefficient.

3.2. Sample Selection and Data Sources

This study selects panel data from 86 prefecture-level cities across 11 provinces, including two municipalities directly under the central government, in China’s Yangtze River Basin covering the period from 2015 to 2022. The data are primarily sourced from the China Statistical Yearbook, China Environmental Statistical Yearbook, China Urban Statistical Yearbook, Provincial Statistical Yearbooks, and Prefecture-level Municipal Statistical Yearbooks. Any missing individual data points are supplemented using interpolation.

3.3. Description of Variables

3.3.1. Explained Variable

The explanatory variable in this paper is NQPF (NewProd). According to [46,47,48,49], NQPF is divided into three key dimensions: science and technology, green development, and digital innovation. A comprehensive evaluation system is constructed based on three primary indicators: scientific and technological NQPF, green NQPF, and digital NQPF. The entropy weight method is employed to measure the NQPF levels of 86 prefecture-level cities in China from 2015 to 2022. Specifically, scientific and technological NQPF (T_NewProd) is measured using the number of graduates from general higher education institutions (persons), the number of students enrolled in general higher education institutions (persons), the number of research and development (R&D) personnel (persons), the number of published scientific and technological papers (articles), and the number of valid invention patents in China (patents). Green NQPF (G_NewProd) is assessed based on urban green space area (hectares), green coverage rate of built-up areas (%), forest coverage rate (%), and the area of national nature reserves (10,000 hectares). Digital NQPF (D_NewProd) is measured by the length of fiber-optic cable routes (kilometers), the number of internet broadband access ports (10,000 ports), software business revenues (CNY billion), employment in the information transmission, software, and information technology services sector (persons), and the number of 5G base stations (10,000 stations). The detailed indicators and measurement methods for the three dimensions of NQPF are provided in

Table A1. The sign, calculation methods, and data sources for New Quality Productive Forces.

	Variable	Label	Calculation Method	Data Source
New Quality Productive Forces	Technology NQPF	T_NewProd	The number of graduates from general higher education institutions (persons), the number of students enrolled in general higher education institutions (persons), the number of research and development (R&D) personnel (persons), the number of published scientific and technological papers (articles), and the number of valid invention patents in China (patents) are combined by the entropy weight method	China Statistical Yearbook, China Environmental Statistical Yearbook, China Urban Statistical Yearbook, Provincial Statistical Yearbooks and Prefectural Statistical Yearbooks
	Green NQPF	G_NewProd	The urban green space area (hectares), green coverage rate of built-up areas (%), forest coverage rate (%), and the area of national nature reserves (10,000 hectares) are combined by the entropy weight method	China Statistical Yearbook, China Environmental Statistical Yearbook, China Urban Statistical Yearbook, Provincial Statistical Yearbooks and Prefectural Statistical Yearbooks
	Digital NQPF	D_NewProd	The length of fiber-optic cable routes (kilometers), the number of internet broadband access ports (10,000 ports), software business revenues (CNY billion), employment in the information transmission, software, and information technology services sector (persons), and the number of 5G base stations (10,000 stations) are combined by the entropy weight method	China Statistical Yearbook, China Environmental Statistical Yearbook, China Urban Statistical Yearbook, Provincial Statistical Yearbooks and Prefectural Statistical Yearbooks

of Appendix A.

3.3.2. Explanatory Variables

The explanatory variable in this paper is environmental regulation (EnvReg), which is measured by the logarithm of energy-saving and environmental protection expenditures (CNY million). Additionally, for robustness testing, this study refers to [26,50,51], who use the logarithm of investments in urban environmental infrastructure construction and the ER method as alternative measures of environmental regulation to conduct further validation.

3.3.3. Intermediary Variables

The mediating variables in this paper are industrial structure upgrading, technological innovation, and GDP level. Following [52], technological innovation (TechInnov) is measured by the ratio of R&D expenditure to the gross regional product of large-scale industrial enterprises, multiplied by 100. Additionally, the measures of industrial structure upgrading (UpIndustr) are based on the approaches of [53,54].

3.3.4. Control Variables

The control variables in this paper include industrial structure, as referenced in [52]; the level of economic development, labor productivity, degree of government intervention, degree of foreign trade, and innovation support, following [55]; as well as financial development, based on [56]. Labor input is also included as a control variable to account for other factors that may influence NQPF. For the full list of variable definitions, calculation methods, and data sources, see

Table A2. The sign, calculation methods and data sources for all variables.

Type	Variable	Label	Calculation Method	Data Source
Explanatory variable	Environmental Regulation	EnvReg	Taking the logarithm of energy conservation and environmental protection expenditures (CNY million)	Statistical Yearbook for Prefectural Municipalities
Explained variable	New Quality Productive Forces	NewProd	Technology, green and digital NQPF. Referring to Table A1	China Statistical Yearbook, China Environmental Statistical Yearbook, China Urban Statistical Yearbook, Provincial Statistical Yearbooks, and Prefectural Statistical Yearbooks
Control Variables	Labor Productivity	LabProd	Share of gross regional product and total employment (million/million) in logarithmic scale	China Statistical Yearbook, China Environmental Statistical Yearbook, China Urban Statistical Yearbook, Provincial Statistical Yearbooks, and Prefectural Statistical Yearbooks
	Level of Financial Development	FinDev	Share of GDP in loan balances of financial institutions
	Financial Support	FinSup	Ratio of total deposits and loans of financial institutions to GDP
	Labor Input	LabInp	Logarithm for the average of the number of employed persons at the end of the previous year and at the end of the current year
	Industrial Structure	IndustrStru	Share of tertiary sector in GDP (%)
	Level of Economic Development	EconDev	Logarithm of GDP per capita
	Level of Government Intervention	GovLev	Local general public budget expenditures as a share of regional GDP
	Create Support Degree	CreateSup	Share of fiscal expenditure on science and technology in local general public budget expenditure
	External Trade Degree	ExterTrad	Ratio of total exports and imports to gross regional product
Intermediary Variables	Technological Innovation	TechInnov	R&D expenditure/GDP of industrial enterprises above designated size multiplied by 100	China Statistical Yearbook, China Environmental Statistical Yearbook, China Urban Statistical Yearbook, Provincial Statistical Yearbooks and Prefectural Statistical Yearbooks
	Upgrading of Industrial Structure	UpIndustr	Measured by the Thiel index and the ratio of the output value of the tertiary industry to the output value of the secondary industry Referring to [53]
	GDP Growth	Ln(GDP)	Logarithm of GDP

in Appendix A. Additionally, the analysis incorporates year- and city-fixed effects to control for unobserved heterogeneity.

4. Empirical Results

4.1. Descriptive Statistics

Table 1. Descriptive statistics.

Variable	(1)	(2)	(3)	(4)	(5)
	Observations	Mean	Sd	Min	Max
T_NewProd	688	0.050	0.090	0.003	0.729
G_NewProd	688	0.181	0.116	0.012	0.586
D_NewProd	688	0.043	0.081	0.001	0.936
NewProd	688	0.069	0.077	0.008	0.783
EnvReg	688	11.60	0.827	9.365	14.71
LabProd	688	11.61	0.743	9.941	13.85
FinDev	688	1.155	0.545	0.310	3.915
FinSup	688	2.689	0.992	0.910	6.887
LabInp	688	5.275	0.859	2.615	7.422
IndustrStru	688	0.460	0.083	0.242	0.775
EconDev	688	10.90	0.511	9.480	12.20
GovLev	688	0.208	0.085	0.080	0.572
CreateSup	688	0.004	0.004	0.000	0.062
ExterTrad	688	0.639	1.297	0.000	9.546
TechInnov	688	1.526	1.233	0.007	11.67
UpIndustr	688	2.336	0.122	2.052	2.739
Ln (GDP)	688	7.669	0.878	5.669	10.71

presents the descriptive statistics of the main variables. For the explained variable, the mean value of NQPF is 0.069, with a maximum of 0.783 and a minimum of 0.008, indicating substantial variation in NQPF levels across different prefectures. For the explanatory variable, the mean intensity of environmental regulation (EnvReg) is 11.60, ranging from 9.365 to 14.71, reflecting differences in environmental regulatory strictness among prefectures in the Yangtze River Basin. Among the control variables, the mean value of financial support (FinSup) is 2.689, suggesting a moderate level of financial backing across regions; the mean government innovation support (CreateSup) is 0.004, indicating limited innovation support at the prefecture level; the degree of external trade (ExterTrad) varies widely, with a maximum of 9.546 and a minimum of 0.000, highlighting significant disparities in foreign trade activity among the studied prefectures. For mediating variables, the degree of technological innovation (TechInnov) varies widely, with a maximum of 11.67 and a minimum of 0.007, indicating uneven innovation capacity across regions, which may affect the mechanism of environmental regulation.

4.2. Base Regression Analysis

Table 2. Benchmark regression results for the impact of environmental regulation on NQPF.

Variable	(1)	(2)	(3)	(4)
	T_NewProd	G_NewProd	D_NewProd	NewProd
EnvReg	0.002	0.004	0.008 *	0.005 *
	(0.002)	(0.008)	(0.004)	(0.003)
LabProd	0.017	−0.008	0.056 **	0.031
	(0.012)	(0.035)	(0.028)	(0.019)
FinDev	0.005	−0.014	−0.039	−0.019
	(0.023)	(0.042)	(0.052)	(0.035)
FinSup	−0.009	−0.008	0.015	0.002
	(0.015)	(0.025)	(0.033)	(0.022)
LabInp	0.040 ***	−0.000	0.116 ***	0.069 ***
	(0.015)	(0.044)	(0.036)	(0.025)
IndustrStru	0.033 **	0.062	0.052 *	0.047 **
	(0.015)	(0.067)	(0.028)	(0.021)
EconDev	−0.023 ***	−0.014	−0.053 ***	−0.036 ***
	(0.006)	(0.025)	(0.013)	(0.010)
GovLev	0.013	−0.134	−0.001	−0.019
	(0.018)	(0.098)	(0.039)	(0.027)
CreateSup	−0.110	0.933	−0.286	−0.010
	(0.234)	(1.061)	(0.528)	(0.358)
ExterTrad	0.007 ***	0.006 **	0.012 ***	0.009 ***
	(0.002)	(0.003)	(0.003)	(0.002)
Id FE	YES	YES	YES	YES
Year FE	YES	YES	YES	YES
Constant	0.294	0.643	−0.608	−0.0653
	(0.247)	(0.698)	(0.578)	(0.394)
Observations	688	688	688	688
R-squared	0.971	0.812	0.859	0.915

Notes: The values in parentheses indicate the robust standard errors. Significance levels are represented by ***, **, and *, corresponding to 1%, 5%, and 10%, respectively. The same applies hereafter.

presents the benchmark regression results examining the impact of environmental regulations on NQPF. Here, T_NewProd represents scientific and technological NQPF, G_NewProd represents green NQPF, D_NewProd represents digital NQPF, and NewProd represents overall NQPF. In Columns (1)–(4), after controlling for all relevant variables, the coefficients for environmental regulations are significantly positive in Columns (3) and (4), which shows 1% increase in EnvReg leads to a 0.008 gain in D_NewProd and a 0.005 gain in NewProd. This indicates that, after accounting for other potential influencing factors, environmental regulations have a significant positive effect on NQPF. Further analysis shows that environmental regulations primarily enhance digital NQPF, effectively driving the overall development of NQPF. This can be attributed to the fact that while environmental regulations impose external pressures on enterprises, they also incentivize innovation, encouraging increased investment in digital technology R&D and green production process upgrades, thereby fostering an overall improvement in NQPF [57]. In summary, Hypothesis 1 is supported: environmental regulations can effectively enhance the level of NQPF.

4.3. Robustness Analysis

4.3.1. Exclusion of Pilot Policy Effects

To avoid potential interference from other policies during the sample period that could bias the benchmark regression results [58], cities designated as National Supply Chain Innovation and Application Demonstration Cities and Demonstration Enterprises—jointly issued by the Ministry of Commerce and eight other agencies—as well as cities included in the First Batch of Pilot Cities for the Collaborative Development of Smart City Infrastructure and Intelligent Networked Vehicles, identified by the Ministry of Housing and Urban-Rural Development and two other departments, are excluded from the benchmark regression model. By controlling for the impact of these two policies, the analysis ensures a more rigorous and reliable verification of the causality in the benchmark regression. Columns (3) and (4) of

Table 3. Robustness of regression results for the effect of environmental regulation on NQPF (excluding the effect of pilot policies).

Variable	(1)	(2)	(3)	(4)
	T_NewProd	G_NewProd	D_NewProd	NewProd
EnvReg	0.002 **	0.008	0.007 ***	0.005 ***
	(0.001)	(0.008)	(0.002)	(0.002)
Controls	YES	YES	YES	YES
Id FE	YES	YES	YES	YES
Year FE	YES	YES	YES	YES
Constant	0.065	1.091	−0.052	0.189
	(0.121)	(0.678)	(0.251)	(0.200)
Observations	688	688	688	688
R-squared	0.953	0.797	0.775	0.862

Note: *** and ** indicate statistical significance at the 1% and 5% levels, respectively.

show that, after excluding the influence of these policies, the regression results for digital and overall NQPF remain significant at the 1% level. An increase of 1% in EnvReg leads to a 0.002 gain in T_NewProd, a 0.007 gain in D_NewProd and a 0.005 gain in NewProd. This confirms the robustness of the benchmark regression results.

4.3.2. Exclusion of Provincial Capital Cities

Provincial capital cities typically experience faster development, and their unique characteristics may affect the accuracy of the study’s findings. To ensure the rigor of the analysis, provincial capital cities are excluded from the benchmark regression model. As shown in Columns (3) and (4) of

Table 4. Robustness for the impact of environmental regulation on NQPF (deleting capital cities) regression results.

Variable	(1)	(2)	(3)	(4)
	T_NewProd	G_NewProd	D_NewProd	NewProd
EnvReg	0.001 *	0.010	0.005 ***	0.004 **
	(0.001)	(0.008)	(0.002)	(0.002)
Controls	YES	YES	YES	YES
Id FE	YES	YES	YES	YES
Year FE	YES	YES	YES	YES
Constant	−0.093 *	0.593	−0.413 *	−0.122
	(0.055)	(0.657)	(0.218)	(0.168)
Observations	688	688	688	688
R-squared	0.954	0.795	0.781	0.801

Note: ***, **, and * indicate statistical significance at the 1%, 5%, and 10% levels, respectively.

, Each 1% rise in EnvReg results in 0.001-unit higher T_NewProd, 0.005-unit higher D_NewProd, and 0.004-unit higher NewProd. The regression results for digital, and overall NQPF remain significant after excluding these cities, further confirming the robustness of the benchmark regression results.

4.3.3. Double-Sided Trimming

A 1% double-sided trimming is applied to all variables to mitigate the influence of extreme data values on the regression results. As shown in Columns (3) and (4) of

Table 5. Robustness (reduced-tailed 1% treatment) regression results for the effect of environmental regulation on NQPF.

Variable	(1)	(2)	(3)	(4)
	T_NewProd	G_NewProd	D_NewProd	NewProd
EnvReg	0.002	0.003	0.006 **	0.004 *
	(0.001)	(0.008)	(0.003)	(0.002)
Controls	YES	YES	YES	YES
Id FE	YES	YES	YES	YES
Year FE	YES	YES	YES	YES
Constant	0.275 **	0.763	−0.102	0.185
	(0.120)	(0.685)	(0.337)	(0.234)
Observations	688	688	688	688
R-squared	0.986	0.813	0.909	0.942

Note: ** and * indicate statistical significance at the 5% and 10% levels, respectively.

, after this adjustment, 1% increase in EnvReg leads to a 0.006 gain in D_NewProd and a 0.004 gain in NewProd, and the regression results for numerical and overall NQPF remain significant at the 5% and 10% levels, respectively, demonstrating the robustness of the baseline regression findings.

4.3.4. Replacement of Explanatory Variables

By replacing different explanatory variables in the test, if the results are not affected by specific measures, it indicates that the benchmark regression results are robust. The environmental regulation measured by energy conservation and environmental protection expenditures is re-estimated by replacing the environmental regulation measured by the logarithm of the investment in urban environmental infrastructure construction and the environmental regulation calculated by the ER method, respectively. The results in Columns (3) and (4) of

Table 6. Robustness (replacement of explanatory variables) regression results for the impact of environmental regulation on NQPF.

Variable	(1)	(2)	(3)	(4)
	T_NewProd	G_NewProd	D_NewProd	NewProd
EnvReg	0.003	0.046 ***	0.011 *	0.014 ***
	(0.003)	(0.010)	(0.006)	(0.005)
Controls	YES	YES	YES	YES
Id FE	YES	YES	YES	YES
Year FE	YES	YES	YES	YES
Constant	0.297	0.599	−0.597	−0.067
	(0.251)	(0.688)	(0.586)	(0.398)
Observations	688	688	688	688
R-squared	0.971	0.818	0.859	0.916

Note: *** and * indicate statistical significance at the 1% and 10% levels, respectively.

and

Table 7. Robustness (another replacement of explanatory variables) regression results for the impact of environmental regulation on NQPF.

Variable	(1)	(2)	(3)	(4)
	T_NewProd	G_NewProd	D_NewProd	NewProd
EnvReg	0.044 **	0.103	0.106 **	0.083 **
	(0.021)	(0.079)	(0.048)	(0.040)
Controls	YES	YES	YES	YES
Id FE	YES	YES	YES	YES
Year FE	YES	YES	YES	YES
Constant	0.255	0.546	−0.695	−0.137
	(0.258)	(0.712)	(0.604)	(0.410)
Observations	688	688	688	688
R-squared	0.971	0.812	0.859	0.915

Note: ** indicates statistical significance at the 5% levels.

indicate that the results of the benchmark regressions for digital, and overall NQPF are still robust. As shown in Table 6, a 1% increase in EnvReg is associated with a 0.046-unit rise in G_NewProd, a 0.011-unit increase in D_NewProd, and a 0.014-unit in NewProd. Meanwhile, Table 7 indicates that a 1% increase in EnvReg corresponds to a 0.044-unit growth in T_NewProd, a 0.106-unit rise in D_NewProd, and a 0.083-unit increase in NewProd.

4.3.5. Time-Fixed Effects Only

To assess the model’s sensitivity to time-fixed effects, the benchmark regression is run with only time-fixed effects, omitting the city-fixed effects used in the original time-city two-way fixed effects model. As shown in Columns (3) and (4) of

Table 8. Robustness (fixed time only) regression results for the effect of environmental regulation on NQPF.

Variable	(1)	(2)	(3)	(4)
	T_NewProd	G_NewProd	D_NewProd	NewProd
EnvReg	0.012 ***	0.013	0.011 **	0.012 ***
	(0.005)	(0.009)	(0.005)	(0.004)
Controls	YES	YES	YES	YES
Id FE	NO	NO	NO	NO
Year FE	YES	YES	YES	YES
Constant	−1.653 ***	−1.074 ***	−1.257 ***	−1.368 ***
	(0.125)	(0.195)	(0.137)	(0.116)
Observations	688	688	688	688
R-squared	0.771	0.317	0.691	0.706

Note: *** and ** indicate statistical significance at the 1% and 5% levels, respectively.

, under this specification, a 1% increase in EnvReg is associated with a 0.012-unit rise in T_NewProd, a 0.011-unit increase in D_NewProd, and a 0.012 gain in NewProd, and numerical NQPF remains significant at the 5% level, while the regression results for overall NQPF remain significant at the 1% level, indicating that the baseline regression results are robust.

4.4. Mechanism Analysis

To examine the mechanism by which environmental regulation influences the improvement of NQPF, the results are presented in

Table 9. Regression results of the mechanism for environmental regulation on NQPF.

Variable	(1)	(2)	(3)
	UpIndustr	TechInnov	Ln (GDP)
EnvReg	0.006 **	0.268 ***	0.013 ***
	(0.003)	(0.096)	(0.004)
Controls	YES	YES	YES
Id FE	YES	YES	YES
Year FE	YES	YES	YES
Constant	0.717 *	14.61 *	−4.118 ***
	(0.419)	(7.811)	(0.884)
Observations	688	688	688
R-squared	0.982	0.792	0.999

Note: ***, **, and * indicate statistical significance at the 1%, 5%, and 10% levels, respectively.

.

(1)

Promoting Industrial Structure Upgrading. Column (1) of Table 9 shows that the coefficient for the impact of industrial structure upgrading on NQPF is significantly positive. A 1% increase in EnvReg leads to a 0.006-unit growth in UpIndustr because the implementation of environmental regulation policies encourages highly polluting and inefficient traditional industries to upgrade or exit the market, thereby freeing up resources and creating development opportunities for emerging sectors such as high-tech and modern service industries. As the industrial structure continuously optimizes, the overall economic development model transforms, gradually shifting towards being knowledge- and technology-intensive and innovation-driven, which in turn enhances NQPF. Specifically, emerging industries typically rely on advanced technologies, so upgrading the industrial structure directly and significantly boosts technology-driven NQPF. Moreover, following industrial optimization, the growth of green industries further promotes improvements in green NQPF. Additionally, the widespread application of digital technologies within emerging industries enhances digital NQPF, collectively driving an overall increase in NQPF.

(2)

Promoting Technological Innovation. Column (2) of Table 9 shows that the coefficient measuring the impact of technological innovation on NQPF is significantly positive. A 1% increase in EnvReg leads to a 0.268-unit growth in TechInnov because, following the implementation of stringent environmental regulations, enterprises are required to reduce pollution emissions. To comply, they must increase investment in environmental protection technology research and the adoption of cleaner production technologies. Throughout this process, enterprises actively pursue technological innovation. These innovations not only help firms meet regulatory requirements but also improve production efficiency and product quality, thereby enhancing their market competitiveness and driving continuous improvements in NQPF. Specifically, technological innovation has the most direct and significant effect on enhancing technology-driven NQPF. Additionally, many environmental technology innovations are closely linked to green NQPF, fostering advancements in green production technologies. Furthermore, digital technologies play a crucial role in technological innovation; through big data, artificial intelligence, and related technologies, enterprises can optimize production and management processes, thereby boosting digital NQPF and ultimately driving overall growth in NQPF.

(3)

Enhancement of GDP Level. Column (3) of Table 9 shows that the impact coefficient of Ln (GDP) on NQPF is significantly positive. A 1% increase in EnvReg leads to a 0.013-unit growth in Ln (GDP) because environmental regulation policies help redirect social resources toward more efficient and environmentally friendly industries and enterprises, thereby promoting sustainable economic development. As GDP levels rise, enterprises have greater capital and capacity to undertake technological transformation and innovation, while the government can allocate more resources to education, scientific research, and infrastructure development—factors that collectively provide a strong foundation for improving NQPF. On the one hand, economic prosperity encourages increased enterprise investment in R&D, accelerating technological progress and the widespread adoption of digital technologies, which significantly drive the growth of technology-based NQPF. On the other hand, with enhanced economic development, public environmental awareness rises and the concept of green development takes deeper root, prompting both enterprises and society to prioritize the development and application of green technologies, thereby substantially boosting green NQPF. Additionally, economic growth leads to increased market demand and consumption upgrades, encouraging enterprises to pursue further technological innovation and industrial upgrading to meet the demand for high-quality, green, and digital products and services. Ultimately, through the combined effects of these factors, environmental regulation policies not only directly contribute to GDP growth but also indirectly establish a solid economic foundation for productivity reforms, thereby promoting the overall enhancement of NQPF at the macro level.

In summary, Hypothesis 2 is confirmed: environmental regulation influences NQPF by promoting industrial structure upgrading, driving technological innovation, and enhancing GDP levels.

4.5. Heterogeneity Analysis

Due to variations in economic development, industrial structure, environmental policies, and other factors across regions, the relationship between environmental regulation and NQPF exhibits distinct characteristics [59,60]. Therefore, to clarify regional heterogeneity, this study conducted heterogeneity tests for the Chengdu–Chongqing City Cluster and the Wuhan “1 + 8” City Circle. The results are presented in

Table 10. Heteroscedasticity regression results for the effect of environmental regulation on NQPF.

Variable	(1)	(2)
	Chengdu–Chongqing City Cluster	Wuhan “1 + 8” City Circle
EnvReg	0.010 *	0.024
	(0.006)	(0.014)
Controls	YES	YES
Id FE	YES	YES
Year FE	YES	YES
Constant	1.820 **	−3.020
	(0.778)	(2.093)
Observations	152	48
R-squared	0.927	0.982

Note: ** and * indicate statistical significance at the 5% and 10% levels, respectively.

.

In the Chengdu–Chongqing urban agglomeration, located in the upper reaches of the Yangtze River, environmental regulation has a significant positive impact on NQPF. This can be attributed to the region’s rapid economic development and accelerating urbanization. Under the influence of market mechanisms, enterprises are more inclined to respond to the innovation pressures imposed by environmental regulations by increasing investments in new technology, green industries, and digital technologies, thereby enhancing NQPF. Additionally, guided by industrial policies, the Chengdu–Chongqing region is actively developing high-tech and modern service industries. Environmental regulations play a more pronounced role in promoting industrial upgrading, which further drives growth in NQPF.

In the middle reaches of the Yangtze River, the impact of environmental regulations on new productivity is not significant, mainly because the region’s economic development is relatively stable, industrial restructuring progresses slowly, and traditional industries still constitute a substantial portion of the economy. These traditional industries tend to be less responsive to environmental regulations and have limited incentives for technological innovation and digital transformation, making it challenging for environmental regulations to effectively drive new productivity. Moreover, the implementation of environmental policies and supporting measures in the midstream region is less comprehensive than in the upstream area, reducing the effectiveness of regulations as innovation incentives for enterprises. Additionally, the collaborative innovation mechanisms among cities in the midstream region are underdeveloped, hindering the formation of effective industrial clusters and a robust innovation ecosystem. Together, these factors constrain the ability of environmental regulations to promote new productivity in the region.

4.6. Further Analysis

4.6.1. Threshold Effect

This study employs a dual fixed-effects model, controlling for prefecture-level city and time effects, across all models to ensure the accuracy of the regression. To test Hypothesis 3, which investigates whether the impact of environmental regulations (EnvReg) on new-type productivity (NewProd) and its decomposition terms exhibits nonlinearity with changes in log GDP (Ln (GDP)), the log of GDP is used as a threshold variable. Prior to estimating the panel threshold model, it is necessary to test for the existence of a threshold effect and determine the number of thresholds. This study utilizes a triple panel threshold model to initiate the estimation process and constructs corresponding p-values based on simulation results. The test results, presented in

Table 11. Single-threshold estimator.

Model	Threshold	Lower	Upper
Th-1	9.7733	9.6404	9.8993

and

Table 12. Single-threshold effect test.

Threshold	RSS	MSE	Fstat	Prob	Crit10	Crit5	Crit1
Single(NewProd)	0.3025	0.0004	93.02	0.0067	34.3896	43.9600	81.5636
Single(D_NewProd)	0.5640	0.0008	89.73	0.0233	43.3721	61.2591	118.4295

, show that the F-statistic for the single-threshold effect on NewProd and its decomposition term D_NewProd is significant at the 10% level, while all tests for double and triple thresholds are not significant. These findings suggest that EnvReg exhibits a single-threshold effect on changes in NewProd and D_NewProd with respect to Ln (GDP), indicating that the direction and magnitude of EnvReg’s influence vary depending on the level of Ln (GDP).

Table 13. Single-threshold effect regression results.

Variable	(1)	(2)
	NewProd	D_NewProd
${\mathrm{E}\mathrm{n}\mathrm{v}\mathrm{R}\mathrm{e}\mathrm{g}}_{it}\times I\left({th}_{it}<\theta \right)$	0.005 *	0.008 *
	(0.003)	(0.004)
${\mathrm{E}\mathrm{n}\mathrm{v}\mathrm{R}\mathrm{e}\mathrm{g}}_{it}\times I\left({th}_{it}\ge \theta \right)$	0.012 ***	0.017 ***
	(0.004)	(0.005)
Controls	YES	YES
Id FE	YES	YES
Year FE	YES	YES
Constant	−0.177	−0.546 *
	(0.226)	(0.293)
Observations	688	688
R-squared	0.436	0.455

Note: *** and * indicate statistical significance at the 1% and 10% levels, respectively.

presents the comprehensive estimation results of the threshold effect. To investigate the threshold effect of environmental regulations on new productive capacity, Columns (1) and (2) display the regression results including control variables. Specifically, Column (1) highlights the single-threshold effect of EnvReg on NewProd in relation to changes in Ln (GDP), with the coefficient for the Ln (GDP) interval being significant at the 10% level. When Ln (GDP) is below 9.7733, the estimated coefficient for NewProd is 0.005, indicating that environmental regulations significantly promote improvements in NewProd as Ln (GDP) changes. This effect may be attributed to the relatively low sensitivity of enterprises in less developed regions to environmental regulations; at lower stages of economic development, firms primarily focus on scale expansion and economic growth. Although environmental regulations impose certain requirements, enterprises at this stage generally have limited technological innovation capabilities and resource investments. Consequently, they tend to adopt marginal improvement strategies, such as simply enhancing environmental protection measures within production processes. While this approach has a modest effect on advancing NQPF, it nonetheless encourages firms to improve environmental performance, driving gradual progress in overall NQPF. When Ln (GDP) exceeds 9.7733, the coefficient for NewProd remains positive and increases to 0.012, indicating that the positive impact of EnvReg on NewProd strengthens significantly in this range. This can be explained by the fact that enterprises in more economically developed regions possess stronger risk-bearing capacities and greater willingness to invest in technological innovation, enabling them to respond more effectively to stringent environmental regulations. Developed regions typically feature more advanced financial markets and richer educational resources, making it easier for firms to secure funding for green technological innovation and attract high-quality talent to drive transformation and upgrading. Through technological and managerial innovation, enterprises enhance NQPF, better adapt to rigorous environmental regulations, and achieve coordinated economic and environmental development.

Column (2) highlights the single-threshold effect of EnvReg on D_NewProd as Ln (GDP) changes, with the coefficient in the Ln (GDP) interval significant at the 10% level. When Ln (GDP) is below 9.7733, the estimated coefficient for D_NewProd is 0.008, indicating that environmental regulations significantly promote improvements in D_NewProd as Ln (GDP) changes. This may be attributed to the relatively simple industrial structure and weak foundation for digital transformation in underdeveloped regions; at early stages of economic development, regional industries are dominated by traditional labor-intensive sectors, with insufficient digital infrastructure and limited application scenarios for data elements. However, the implementation of environmental regulations has encouraged enterprises to explore new production models. In this process, firms have gradually recognized the potential of digital technologies to enhance production efficiency and environmental performance, thereby driving the initial development of digital new productive forces. For example, some enterprises have attempted to introduce basic digital environmental monitoring systems. Although their impact on accelerating digital transformation is limited, these efforts lay the groundwork for subsequent digital development. When Ln (GDP) exceeds 9.7733, the coefficient for D_NewProd remains positive and increases to 0.017, indicating that within this range, the positive effect of EnvReg on D_NewProd strengthens significantly. This suggests that stricter environmental regulations enable economically developed regions to better align with regulatory requirements, leveraging their favorable industrial ecosystems and policy support frameworks. Developed regions typically have well-established digital economy ecosystems, including abundant digital technology R&D institutions, mature digital service markets, and supportive policies for digital industry growth. Under such conditions, enterprises can fully leverage external resources to accelerate digital transformation. By deeply integrating cutting-edge technologies such as big data, artificial intelligence, and the Internet of Things with traditional industries, they foster new business models and industry formats, significantly enhancing digital NQPF capacity and ultimately driving substantial improvements in overall NQPF. In summary, both statistical and economic evidence support research Hypothesis 3, according to which the impact of environmental regulations on NQPF exhibits a threshold effect as Ln (GDP) changes.

4.6.2. Spatial Effect

This study estimates the spatial impact of environmental regulation (EnvReg) on NQPF using a spatial Durbin model with panel data from 86 prefecture-level cities in the Yangtze River Basin, China, spanning 2015 to 2022 [61].

Table 14. Spatial Durbin regression results.

Variable	(1) D_NewProd		(2) NewProd
	Main	WX	Main	WX
EnvReg	0.009 *	0.254 ***	0.006 *	0.158 **
	(1.89)	(2.59)	(1.88)	(2.22)
Controls	YES	YES	YES	YES
Id FE	YES	YES	YES	YES
Year FE	YES	YES	YES	YES
Observations	688	688	688	688
R-squared	0.331	0.331	0.234	0.234

Note: ***, **, and * indicate statistical significance at the 1%, 5%, and 10% levels, respectively.

presents the corresponding regression results. The sub-indicators of NQPF analyzed here primarily focus on digital NQPF (D_NewProd), alongside overall NQPF (NewProd). Columns (1) and (2), respectively, illustrate the spatial effects of environmental regulation on D_NewProd and NewProd.

As shown in Columns (1) and (2) of Table 14, environmental regulation has a significant positive effect on both digital NQPF (D_NewProd) and overall NQPF (NewProd), passing the 10% significance level test. This indicates that environmental regulation promotes NQPF, with every 1% increase in environmental regulation associated with a 0.009-unit increase in D_NewProd and a 0.006-unit increase in NewProd. This effect may be driven by policy guidance and incentives that encourage the development of both digital and overall NQPF. Furthermore, the spatial spillover effects of environmental regulation are also significant and positive. The estimated spatial spillover coefficients reveal that environmental regulation in neighboring prefectures positively influences their digital NQPF (D_NewProd) and overall NQPF (NewProd), demonstrating a beneficial spatial spillover effect across regions.

5. Conclusions and Recommendations

5.1. Conclusions

Based on panel data from 86 prefecture-level cities in the Yangtze River Basin spanning 2015 to 2022, this study investigates the impact of environmental regulation on NQPF and draws the following conclusions.

(1)

Environmental regulation significantly promotes the development of digital and overall NQPF.

(2)

Environmental regulation drives advancement in digital NQPF and overall NQPF by facilitating industrial structure upgrading, encouraging technological innovation, and increasing GDP levels.

(3)

The effect of environmental regulation on digital NQPF and overall NQPF is nonlinear and exhibits a threshold effect. In less developed regions, environmental regulation has a moderate promotional impact, while in more developed regions, once environmental regulation intensity surpasses a certain threshold, its positive effect on digital NQPF and overall NQPF increases substantially.

(4)

Environmental regulation exerts a significant spatial spillover effect on digital NQPF and overall NQPF. That is, environmental regulation in one region not only enhances local digital NQPF and overall NQPF but also positively affects neighboring regions through spatial correlations, indicating a synergistic promotional effect across regions.

5.2. Suggestions

(1)

In light of the significant contribution of environmental regulation to new productivity, the government should continue to strengthen the formulation and implementation of environmental regulation policies, particularly in the upstream and downstream areas of the Yangtze River Basin. However, it is essential to ensure policy precision by designing differentiated environmental regulations tailored to each region’s level of economic development, industrial structure characteristics, and environmental carrying capacity, thereby avoiding a “one-size-fits-all” approach. For regions with relatively homogeneous industrial structures, an appropriate transition period should be provided while guiding industries toward diversification and high-tech development to foster a positive interaction between environmental regulation and economic growth. Furthermore, the government should fully consider the spatial spillover effects of environmental regulations on NQPF by enhancing inter-regional cooperation and coordination. Promoting experience exchange and resource sharing among prefectures and cities within the Yangtze River Basin will maximize the spatial synergy of environmental regulation, thereby driving the collective improvement of NQPF across the entire region.

(2)

Considering the threshold effect between environmental regulation and NQPF, policymakers should closely monitor changes in the intensity of environmental regulations and establish a robust mechanism for evaluating and dynamically adjusting policy impacts. The intensity threshold of environmental regulation should be reasonably determined based on the economic development level of different regions and the capacity of enterprises to bear the costs. For less developed regions, targeted policy support and guidance are necessary to help enterprises enhance their technological innovation capabilities and environmental management standards, enabling them to better meet regulatory requirements. In more developed regions, when the intensity of environmental regulation approaches the threshold, policy measures should be timely adjusted to avoid dampening enterprises’ innovation incentives due to excessive regulatory pressure, while simultaneously strengthening support for technological innovation and industrial upgrading. Additionally, efforts to promote technological innovation and industrial upgrading should be reinforced to ensure the continuous and stable improvement of NQPF. The government can also alleviate the cost burden faced by enterprises through financial subsidies, tax incentives, and support for research and development, thereby improving enterprises’ acceptance and implementation of environmental regulations. These measures will maximize the role of environmental regulation in promoting NQPF and facilitate the green transformation and high-quality development of the Yangtze River Basin economy.

5.3. Limitation and Future Research Directions

Like any empirical study, this research has certain limitations that offer opportunities for future exploration.

(1)

The environmental regulation measured in this paper is a primarily mandatory regulation proxied by energy conservation and environmental protection expenditures. This approach captures only one dimension of environmental governance, as environmental regulation systems also include market-incentivized tools like carbon emission trading, pollution rights trading and voluntary actions like corporate environmental information disclosure, green certification. These different types of regulation may affect NQPF through distinct mechanisms. Future research could construct a multi-dimensional index of environmental regulation, incorporating market and voluntary instruments, to compare their heterogeneous impacts on NQPF and verify whether the conclusions of this study hold under diverse regulatory frameworks.

(2)

This study examines NQPF at the prefecture-level city scale without differentiating between industries. However, NQPF exhibits significant industrial heterogeneity; for example, agriculture, manufacturing, and the power sector differ markedly in production modes, pollution intensity, and dependence on digital technologies. Environmental regulation may thus exert varying effects on their NQPF. For example, manufacturing might be more sensitive to regulation-driven digital transformation like intelligent emission monitoring, while agriculture could be more affected by green technology adoption like low-carbon breeding. Future research could disaggregate NQPF by industry, exploring how environmental regulation influences their respective NQPF. Such analyses could further compare whether different types of environmental regulation yield consistent conclusions across industries and whether their mechanisms vary by sector.