Environmental Regulation, Government-Business Relations, and Corporate Green Innovation: Evidence from Low-Carbon City Pilot Policy

Abstract

Environmental degradation and economic development have long been seen as incompatible, posing a pressing challenge for society. Government–business collaboration stands as an effective avenue for addressing environmental issues. This paper, using the 2007–2021 Low-Carbon City Pilot (LCCP) program in China as a quasi-natural experiment, explores the impact of governmental environmental regulations on corporate green innovation and examines the moderating role of government–business collusion. The findings indicate that the LCCP policy has effectively promoted corporate green technological innovation, with results remaining robust across various sensitivity tests. Heterogeneity analysis further reveals that the policy’s impact is more pronounced in the eastern regions, state-owned enterprises, and low-tech industries. Moreover, government–business collusion significantly undermines the benefits of green innovation, though this effect is partially alleviated when local government officials are replaced. Under the influence of government environmental regulation, green innovation fosters the sustainable development of enterprises. However, the formation of collusion between government and business diminishes the incentive for companies to take on environmental and social responsibilities. The study enriches the existing literature on environmental policy factors and offers both theoretical and practical insights for the government in formulating relevant environmental policies and promoting corporate green innovation.

1. Introduction

In its 20th National Congress report, the Party explicitly outlined the strategic initiative to “actively and prudently promote carbon peaking and carbon neutrality”, accompanied by specific directives. Advancing the greening and decarbonization of the economy and society is essential to achieving high-quality development. The Central Economic Work Conference emphasized that fostering new industrial competitive advantages is critical to meeting the goals of carbon peaking and carbon neutrality. To realize the “dual carbon” goals, national policies that provide support and guidance are crucial, alongside the indispensable coordination and cooperation between regional governments, and the commitment to low-carbon development by enterprises. In the context of addressing climate change and promoting eco-city development, China has set ambitious carbon goals. To fulfill the objectives of energy conservation and emissions reduction, environmental regulatory measures have become increasingly stringent and comprehensive, while efforts to establish a green, low-carbon, safe, and efficient modern energy system are well underway. With advancements in energy technology innovation, environmental issues have shown gradual improvement [1]. However, inefficiencies in policy implementation and the inadequacy of corporate innovation in meeting emissions reduction requirements remain evident [2]. This highlights the immaturity of enterprises’ green innovation capabilities and the lack of effective incentives for green production, signaling an urgent need for policy intervention.

Government-issued environmental regulations serve both incentive and constraint functions. Numerous studies have investigated the connection between environmental regulation and the innovation efforts of corporations aimed at sustainability, indicating that environmental regulations impact corporate development [3,4,5]. However, there remains inconsistency in existing studies regarding whether environmental regulation promotes or inhibits industrial innovation. On one hand, research by Su and Zhou (2019) [6] demonstrates a “U-shaped” correlation between formal environmental regulation and corporate innovation, indicating a modest impact. Conversely, an “inverted U-shaped” relationship is observed between informal environmental regulation and corporate innovation. This phenomenon may stem from the lack of comprehensiveness in informal regulations, causing enterprises to struggle under the dual pressures of economic and green innovation, thereby hindering normal development. Yang and Zhao (2023) [7] found, through threshold effect analysis, that a specific intensity of environmental regulation presents a “U-shaped” curve relationship with corporate technological innovation. When environmental regulations are weak, enterprises often focus on low-cost pollution control to maximize profits, thereby reducing investment in technological innovation. As regulations strengthen or pollution control outcomes become unsatisfactory, firms may increase their technological investments, forming a “U-shaped” relationship. On the other hand, some literature suggests a significant positive relationship between government environmental regulation and corporate innovation upgrading [8,9,10]. At the urban level, incentive-based environmental regulations help enhance the rationality and advancement of industrial structures, facilitating corporate transformation and upgrading to achieve both economic growth and environmental emission reduction goals [11,12,13]. At the enterprise level, research by Khan and Liu (2023) [14] indicates that for knowledge-intensive firms, both incentive-based and mandatory environmental regulations create a conducive atmosphere for green innovation, motivating enterprises to actively engage in green innovation. This conclusion is also applicable to other industrial enterprises [15,16,17], although the impact of different types of environmental regulations on corporate behavior varies [18,19,20]. Furthermore, the same environmental regulation can have differing effects on various types of green innovation within enterprises [21,22,23].

The LCCP has been pivotal in reshaping conventional extensive development models and has emerged as a prominent area of research since its initiation, characterized by weak incentive and constraint functions within environmental regulatory policies. Existing literature investigates the influence of policy execution on urban pollution management and green innovation. Zheng et al. (2021) [24] utilized the PSM-DID approach to illustrate that the LCCP substantially accelerates industrial structural upgrades. Green innovation emerges as a crucial catalyst for economic, social, and ecological advancement, significantly contributing to the development of low-carbon cities. Liu et al. (2023) [25] explored the relationship between the LCCP Policy and corporate green technological innovation, confirming that the pilot initiative enhances firms’ green innovation capacities via a multi-period difference–in-differences model. As enterprises increasingly contribute to environmental governance and low-carbon objectives, research perspectives have transitioned from macro-level analyses to micro-level insights. Qu et al. (2023) [26] observed that the LCCP encourages corporate energy technology innovation, fostering advancements in clean energy technologies essential for sustainable development. Ma et al. (2021) [27], by analyzing data from green patent applications by listed companies in China, found that the implementation of low-carbon policies induces corporate technological innovation, particularly influencing high-carbon enterprises and non-state-owned firms significantly. Yang (2023) [28] further examined the environmental performance of A-share listed companies within the low-carbon pilot framework, demonstrating that these policies effectively incentivize enterprises to pursue low-carbon green innovation by alleviating financing constraints. However, the results revealed that state-owned enterprises are more significantly impacted by these policies than their non-state-owned counterparts. Consequently, the diverse effects of the LCCP Policy on corporate green innovation remain a subject of ongoing debate, necessitating further investigation. Moreover, the literature reflects discrepancies in the selection of proxy variables for measuring corporate environmental green innovation. Currently, two predominant methodologies exist: one develops a green technology innovation evaluation system encompassing R&D investment and researcher counts [29,30], while the other utilizes the number of corporate green invention applications as a proxy variable [31].

The LCCP policy plays a crucial role in promoting corporate sustainable development, encouraging companies to respond to environmental policies through green innovation, thereby enhancing their long-term market competitiveness and ecological adaptability. Green innovation is one of the core pathways for enterprises to achieve sustainable development strategies; it not only helps optimize resource allocation but also increases production efficiency, granting companies a significant advantage in market competition. Wang et al. (2022) [32] note that through green technological innovation and clean production, enterprises can substantially reduce resource waste and operational costs, thus enhancing their adaptability within a low-carbon economy and optimizing resource allocation, which contributes to the formation of more sustainable production models, particularly evident in energy-intensive industries [33]. Furthermore, with the growth in demand, green innovation not only assists enterprises in meeting this market trend but also opens new market opportunities. Green innovative products often exhibit a premium effect, allowing companies to achieve substantial financial returns while satisfying the green market demand. This market expansion and innovative pricing strategy positively influence the long-term financial performance of enterprises [34]. Concurrently, corporate green innovation demonstrates a proactive commitment to social responsibility in sustainable development, enabling firms to build a positive brand image and enhance societal trust [35]. This trust is particularly important in sustainable development strategies as it further fosters interactions between enterprises and stakeholders, establishing an eco-friendly social relationship network. Research by Xu et al. (2024) [36] indicates that a company’s reputation in the realm of social responsibility is closely linked to its green innovation efforts, with a positive social image leading to increased brand loyalty and customer satisfaction. In light of the global trend toward sustainable development, green innovation empowers companies to transition from passive compliance with regulations to proactive market leadership, creating a win-win scenario where environmental, economic, and social benefits coexist, thereby providing robust support for their sustainable development strategies.

Government–business relations are a key factor in economic development, serving as an important means of government intervention in the micro-economy and directly influencing corporate behavior. Nie (2020) [37] categorizes government–business relations into four types: cooperation, separation, harm, and collusion. The current literature predominantly focuses on discussions of cooperation and collusion. Government–business cooperation facilitates firms’ transformation toward better practices and promotes effective policy implementation, while collusion represents a distorted relationship formed for short-term benefits, offering no advantages in other areas. Yu (2019) [38] found that frequent changes in government–business relations exhibit nonlinear effects on corporate innovation, with the relationship between official turnover and innovation following an “inverted U-shape”. Jiang et al. (2022) [39] indicate that excessive intimacy between enterprises and government reduces the flexibility of innovation decision-making. As environmental pollution issues intensify, scholars have begun to examine the role of government–business relations in pollution control. From a macro perspective, the dynamic game model constructed by Long and Hu (2014) [40] suggests that enterprises may influence government actions through bribery, leading to a relaxation of environmental regulations in pursuit of political and economic interests, thereby exacerbating environmental pollution. Research based on provincial panel data shows that government–business collusion intensifies regional environmental pollution and pollution transfer phenomena [41]. From a micro perspective, Meng et al. (2023) [42] point out that the interaction of collusion obstructs the achievement of ESG goals, increasing air pollution, with significant effects observed in heavily polluting enterprises. Clearly, government–business collusion not only exacerbates corporate environmental pollution but also hinders environmental protection and sustainable development. Nonetheless, there is a paucity of micro-level studies examining the influence of collusion on corporate environmental green innovation. Therefore, this paper aims to further investigate how government–business collusion impacts the effectiveness of government environmental regulation in fostering corporate green innovation.

Compared to previous studies, this paper’s marginal contributions are primarily reflected in several aspects. First, in terms of research content, this paper uses the LCCP Policy as a natural experiment to explore its impact on corporate environmental green innovation, while considering government–business collusion as a moderating variable to analyze its relationship with corporate green innovation. This approach helps clarify the importance of government–business relations in the process of corporate green innovation. Second, regarding research methodology, the use of difference-in-differences and triple-difference methods, along with moderation effects, enhances the accuracy and persuasiveness of the findings. Finally, in terms of research perspective, while existing studies on the Low-Carbon Pilot Policy predominantly focus on the macro level, there has been a partial shift toward the micro level, yet significant expansion opportunities remain. This paper aims to provide further insights at the micro level, intending to offer important theoretical support for promoting corporate green innovation.

The subsequent chapters of this paper are organized as follows: Section 2 presents the theoretical analysis and research hypotheses. Section 3 outlines the research design, including data sources, variables and their measurements, as well as the empirical model employed. Section 4 reveals the main results and robustness tests. Section 5 provides a detailed discussion of the heterogeneity analysis, examining the moderating effects of government–business collusion on corporate sustainable development. Finally, Section 6 concludes the paper and offers recommendations. The theoretical framework of the study is illustrated in the figure below (Figure 1).

2. Theoretical Analysis and Research Hypotheses

2.1. Theoretical Analysis

To effectively control greenhouse gas emissions and actively address climate change, in July 2010, the National Development and Reform Commission (NDRC) issued a notification regarding the initiation of low-carbon province and city pilot programs. This established the first batch of pilot cities, which included the provinces of Guangdong, Liaoning, Yunnan, Hubei, and Shanxi, and the municipalities of Tianjin, Chongqing, Shenzhen, Xiamen, Hangzhou, Nanchang, Guiyang, and Baoding. In 2013, the pilot program was expanded to cover one province and twenty-eight cities, followed by the implementation of a third batch of low-carbon city pilots in 2018. As of now, the LCCP policy has essentially achieved nationwide coverage. Compared to other policies, the LCCP policy is characterized by its weak binding nature, as it does not impose specific targets for the low-carbon cities; instead, each city is encouraged to advance the pilot work according to its own developmental status [43]. The primary tasks of the government in implementing low-carbon pilot policies can be categorized into two aspects: first, to formulate low-carbon city planning aimed at fostering urban low-carbon development, coupled with the promotion of the optimization of industrial and energy structures; second, to develop policy recommendations that align with the construction of low-carbon cities, facilitating the integration of ecological construction and environmental protection mechanisms during the development process. The establishment of low-carbon cities necessitates active cooperation from enterprises, where enhancing energy efficiency, substituting clean energy for traditional energy sources, and transitioning high-pollution enterprises to low-pollution ones are all measures that contribute to the low-carbon development of cities. In this context, corporate green innovation is indispensable, as the development of LCCP policies plays a pivotal role in promoting environmental technological innovation and corporate transformation.

Research theories regarding environmental regulation and corporate economic sustainability can be categorized into two distinct directions: one is the cost effect, and the other is the “Porter Hypothesis” [44]. The cost effect posits that enhanced environmental regulation increases production costs for enterprises, subsequently reducing their willingness to invest. Simultaneously, it may provoke companies to intensify their focus on environmental pollution control, potentially leading to liquidity risks that are detrimental to investments in green innovation. In contrast, the Porter Hypothesis suggests that appropriate environmental regulation can stimulate a corporate awareness of green innovation. The investments that companies make in green innovation can offset the costs incurred from environmental improvements, thereby contributing to enhanced economic development capabilities. The LCCP policy is distinct from other types of policies in that it exhibits weak incentive effects and low enforcement intensity. Therefore, during its implementation, local governments must proactively adopt measures to motivate cities and enterprises toward green innovation. Local governments can utilize mandatory regulatory tools to impose specific requirements on production processes, strictly regulating the concentration of pollutants emitted by enterprises and closely scrutinizing the development of heavily polluting firms. Such measures can heighten corporate attention to environmental green technological innovation, further advancing corporate environmental green initiatives while simultaneously establishing a positive corporate image to better secure government support. Additionally, the government can provide environmental subsidies to enterprises through mechanisms such as environmental grants and technical assistance. Carbon emission trading allows for the sale of carbon emissions as a production factor, which not only constrains the carbon emissions of enterprises but also promotes their environmental innovation from a financial perspective.

Regarding government–business relations, Chen et al. (2023) [45] assert that a robust relationship between the government and enterprises serves as a driving force for the rise of developing countries. The report from the 19th National Congress emphasized the importance of constructing an economic mechanism characterized by effective market mechanisms, vibrant micro-entities, and appropriate macro-control. Under the hierarchical administrative system, local governments in China occupy a pivotal position between the central government and enterprises, acting as the primary overseers of policy implementation while maintaining direct contact with businesses. The occurrence of government–business collusion arises from the temptations of mutual interests between local governments and enterprises. Typically, local government revenue sharing is derived from taxes paid by enterprises to the central government. To secure greater economic revenue, local governments may permit enterprises to circumvent regulations to save costs or enhance profits, such as overlooking excessive emissions of pollutants or tolerating tax evasion. The collusion between government and enterprises in the pursuit of high-quality economic development can lead to production safety accidents or social issues, with instances of land violations and substandard construction projects already evident. The level of corporate attention to environmental concerns will directly determine the success of low-carbon city initiatives, particularly as enterprises with relatively high pollution levels are a focal point for government oversight in the construction of LCCP. Enterprises that contribute significantly to pollution have a substantial impact on environmental degradation, and increasing investments in green technological innovation may raise production costs, potentially leading to short-term profit declines. Such companies, seeking to maximize profits and evade responsibility for pollution control, may resort to bribery to local governments, facilitating adjustments to their environmental data to align with urban development models while continuing to discharge pollutants. When local governments, in pursuit of short-term economic benefits or career advancement, turn a blind eye to or condone bribery from enterprises, this gives rise to government–business collusion during the implementation of low-carbon pilot policies [46].

2.2. Research Hypotheses

Governmental environmental regulations can effectively promote corporate green R&D innovation. Research and development not only serves as a crucial pillar of a nation’s comprehensive strength, but also acts as a key driver of social and industrial transformation, and a manifestation of a company’s core competitiveness. With the intensification of environmental pollution and excessive energy consumption, green technological innovation has garnered increasing attention. As an integrated environmental regulatory policy, the LCCP initiative incentivizes enterprises to engage in green R&D by regulating their behavior [47]. The government provides support to green enterprises in terms of funding, technology, and talent, while the policy reduces the costs associated with green innovation, such as through special financial subsidies and tax reductions, thereby mitigating the uncertainties and risks faced by enterprises in the innovation process [25]. The LCCP policy offers multifaceted support to corporate green R&D, lowering innovation risks, while the outcomes of green technological innovation inject vitality into corporate growth and economic success. For the government, corporate green innovation helps alleviate environmental and energy-related challenges [48]. As enterprises transition towards cleaner and low-carbon production models, their level of green technological innovation rises, facilitating industrial upgrading and optimization. By evaluating the implementation of their own city policies and comparing them with other provinces and cities, local governments can refine and adjust the LCCP policy, thereby advancing sustainable development comprehensively [49].

H1. 

The implementation of the LCCP Policy can promote corporate green technological innovation.

The conflict of interest between the government and enterprises is a significant factor inhibiting corporate green performance. Firstly, as a public authority, the government typically enacts environmental regulations and standards to safeguard public interests, requiring enterprises to comply. This imposes additional costs and constraints on businesses, potentially hindering their development. In contrast, enterprises tend to prioritize economic gains, often intensifying pollution-generating activities such as emissions and resource waste to maximize profits. This conflict with government environmental mandates ultimately suppresses corporate green performance [50]. Secondly, improper relationships between government officials and corporate representatives further undermine green performance. Corruption, including bribery and quid pro quo exchanges, may lead to regulatory leniency or grant enterprises privileges and exemptions, allowing them to evade environmental responsibilities [51]. Such unethical practices distort market competition by skewing the balance between environmental costs and benefits, thereby further curtailing corporate green performance. Moreover, the lack of a cooperative, mutually beneficial mechanism between the government and enterprises limits green outcomes. Governments often rely on punitive measures to regulate corporate environmental behavior, while offering insufficient incentives or support, which dampens business enthusiasm for proactively adopting eco-friendly practices [52]. When opportunities for collaboration and communication between the government and enterprises are restricted, businesses struggle to fully understand environmental policies and measures, making it difficult to secure the necessary support and resources.

At the same time, the establishment of collusive relationships requires time and the maintenance of interests. When the status of the stakeholders diminishes or their rights change, the relationship between the government and enterprises may be disrupted. Li et al. (2024) [53] found that changes in officials can effectively govern the collusive relationships between government and enterprises. Newly appointed officials typically do not have the deep-rooted interests and long-term interactions that former officials established with local enterprises. The existing collusive relationships are often built through long-term exchanges of benefits, resource allocation, and informal networks. Upon taking office, new officials usually reassess policy objectives and resource distribution methods while scrutinizing current government–business relations in order to establish their own governance style and authority. Driven by performance evaluations, new officials can also bring about greater transparency in policies and strengthen accountability mechanisms, aiming to disrupt the existing patterns of interest. In this process, informal ties with enterprises are likely to diminish, leading to greater adherence to open and fair procedures. This shift can weaken the motivation of enterprises to rely on officials for privileges, compelling them to depend more on market competition for their development, thereby further reducing the likelihood of government–business collusion.

H2a. 

Government–business collusion inhibits corporate green benefits.

H2b. 

Changes in officials mitigate the suppressive effect of government-business collusion.

By adopting environmental protection actions and implementing sustainable development strategies, companies convey positive environmental signals to the outside world, thereby enhancing their financial performance. These environmental signals improve corporate reputation and image, attracting more investors and consumers, increasing sales revenue and market share, and ultimately improving financial status [54]. At the same time, environmental protection actions promote the efficient use of resources and energy conservation; by reducing waste and lowering costs, companies can enhance operational efficiency and improve profit margins [55]. Actively addressing environmental issues reflects a company’s sense of responsibility towards the environment and society, demonstrating a high level of corporate social responsibility. By focusing on environmental and social issues, companies can reduce pollutant emissions, utilize resources effectively, use clean energy, minimize negative environmental impacts, and improve relationships with stakeholders (employees, communities, governments, etc.), thereby increasing social recognition, enhancing brand reputation, and shaping a positive corporate image, which in turn, boosts productivity and performance.

However, the interest-driven relationship between the government and enterprises exerts a certain suppressive effect on sustainable development. While government–business collusion can enhance corporate financial performance by providing additional funding and resource support, aiding companies in scaling up and improving production capacity, it may also lower the level of corporate environmental and social responsibility, thereby inhibiting sustainable development [56]. Companies often prioritize short-term economic benefits over the long-term impacts of environmental protection and social responsibility, resulting in a diminished level of environmental and social accountability. Furthermore, government–business collusion can lead to issues of information asymmetry. Although cooperation involves the exchange of interests and information sharing, information asymmetry may prevent enterprises from fully understanding the actual conditions of environmental and social issues, hindering effective environmental regulation and the implementation of sustainable development [57]. The potential risks of corruption can result in the improper allocation of resources and inadequate environmental protection measures [58]. Corrupt practices may lead to the neglect or concealment of environmental problems, with enterprises colluding with government officials to obtain privileges, ultimately affecting environmental remediation and deviating from sustainable development goals.

H3a. 

Under the stimulation of the LCCP Policy, green innovation technologies effectively enhance corporate sustainable development.

H3b. 

The LCCP Policy, under the moderating effect of government–business collusion, suppresses corporate sustainable development.

3. Research Design

3.1. Data Sources

The green patent data for this paper are sourced from the China National Research Data Service (CNRDS), while corporate financial information is gathered from the CSMAR database and supplemented by the Wind database. Urban-level data are obtained from the EPS database and the National Bureau of Statistics. This paper employs the LCCP policy as a form of government environmental regulation, treating it as a quasi-natural experiment within the model. The specific pilot cities are identified from the National Development and Reform Commission’s work notification and organized into three batches to establish control and experimental groups. Since the policy is implemented in three phases, A-share listed companies from 2007 to 2021 are selected as the original research sample to ensure the validity of pre- and post-implementation comparisons. ST and PT companies, as well as those with significant missing financial data, are excluded. Furthermore, all data undergo winsorization at the 1% level to mitigate the impact of outliers on the regression results.

3.2. Variable Measurement

3.2.1. Dependent Variable

This paper defines corporate green innovation as the dependent variable. While existing literature often measures corporate innovation capabilities through R&D investment or patent counts, a lack of standardized metrics persists. Given that R&D investment primarily captures early-stage innovation, accurately assessing innovation output can be challenging. Thus, the inclusion of green patents offers a clearer representation of corporate innovation performance. This paper employs the count of granted green patents as a measure of a company’s innovation capability. Additionally, green patents can be classified into two categories: green invention patents and green utility model patents. Green invention patents signify substantial technological innovations, while green utility model patents represent incremental advancements. Therefore, this paper investigates the number of granted green invention patents (GIO) and green utility model patents (GUO) as dependent variables.

3.2.2. Independent Variable

This paper uses the LCCP Policy as the policy intervention, which was implemented in three batches in 2010, 2013, and 2017. Accordingly, the paper measures the treatment group in batches based on the pilot cities designated by the policy. Specifically, enterprises located in pilot cities and operating in the year of policy implementation and thereafter are considered the treatment group (LCC = 1), while those that do not meet this criterion are regarded as the control group (LCC = 0).

3.2.3. Moderating Variable

Due to the limitations of precision in proxy variables and data availability, the relevant literature often starts from the provincial level, commonly using indicators such as the number of cases related to corruption, bribery, and misconduct [59]. Some studies have also explored the implied transfer of pollution, suggesting that higher levels of government–business collusion correspond to greater pollution transfer at the provincial level. This paper’s empirical model is based on micro-level enterprise data and adopts the method proposed by Zhang and Liu (2021) [60], using the actual corporate income tax paid to inversely measure the extent of government–business collusion. Specifically, a higher effective corporate income tax rate indicates lower levels of collusion. Given the relatively high value-added tax rate and the stringent invoice management and accounting checks during tax collection, the cost of evading value-added tax for enterprises is significantly high. In contrast, the collection and regulation of corporate income tax are mainly managed by local governments, making it difficult for the central government to enforce comprehensive and strict oversight, thus providing some leeway for tax evasion. When close ties exist between enterprises and local governments, the latter may adopt a lenient attitude towards tax evasion, resulting in a situation where the income tax owed by the enterprise exceeds the actual tax paid. Therefore, this paper calculates the average effective income tax rate by industry and year. If a company’s effective income tax rate for the year is below this average, it is considered to have a serious government–business collusion relationship (ITR = 1); otherwise, it is deemed not to have a serious collusion relationship (ITR = 0).

3.2.4. Control Variables

This paper identifies control variables that may impact the development of corporate green innovation at both the enterprise and provincial levels. The age of the enterprise is measured by the logarithm of the difference between the current operating year and the establishment year, calculated as ln(sample observation year − establishment year). The debt-to-asset ratio is determined by the total liabilities divided by the total assets. Enterprise size is assessed using the natural logarithm of total assets, expressed as ln(total assets). Capital intensity is calculated as the ratio of total assets to operating income. Tobin’s Q is computed by dividing the market value of the enterprise by total assets. Board composition is measured by the natural logarithm of the number of board members, or ln(number of board members). The economic scale is represented by the total economic output of each province, measured as ln(total economic output). The industrial structure, which may influence results due to varying provincial industrial focuses, is quantified by the ratio of the secondary industry to the tertiary industry. Urbanization is assessed through the ratio of the urban population to the total provincial population. Foreign direct investment is calculated as the ratio of actual foreign direct investment to total economic output.

3.3. Empirical Model

To validate the hypotheses, this paper utilizes the Difference-in-Differences (DID) method to assess the impact of the low-carbon city policy on corporate green innovation. The specific model is expressed in Equation (1) as follows:

$${\mathrm{Ginnovation}}_{itd}={\alpha }_0+{\alpha }_1{\mathrm{LCC}}_{itd}+{\alpha }_2{\mathrm{Controls}}_{it}+u_i+u_t+u_d+{\epsilon }_{itd}$$

(1)

$${\mathrm{LCC}}_{itd}={\mathrm{Treat}}_{id}\ast {\mathrm{Post}}_t$$

(2)

In this equation, ${Ginnovation}_{itd}$ represents the green innovation capability of enterprises in year t, measured by green invention patent grants (GIO) and green utility model patent grants (GUO). ${LCC}_{itd}$ denotes the policy effect of this paper, where ${Treat}_{id}$ indicates whether the city where the enterprise is located is a pilot city, and ${Post}_t$ indicates whether the city where the enterprise is located is a pilot city, and ${Controls}_{it}$ signifies the control variables; $u_i,u_t$ and $u_d$ represent enterprise fixed effects, time fixed effects, and industry fixed effects, respectively, while ${\epsilon }_{itd}$ is the error term.

To examine the suppressive effect of government–enterprise collusion on corporate green innovation capability, this paper constructs a Triple Difference (DDD) model. The specific model is outlined in Equation (2) as follows:

$$\begin{array}{cc}\hfill Gimnovatio{n}_{itd}=& {\beta }_0+{\beta }_{1}LC{C}_{itd}+{\beta }_{2}LC{C}_{itd}\ast IT{R}_{itd}+{\beta }_{3}IT{R}_{itd}+{\beta }_{4}Trea{t}_i+\hfill \\ \hfill & {\beta }_{5}Pos{t}_t+{\beta }_{6}Control{s}_{it}+u_i+u_t+u_d+{\epsilon }_{itd}\hfill \end{array}$$

(3)

In this equation, $IT{R}_{itd}$ represents the degree of government–enterprise collusion, Here, ${Treat}_{id}\ast {Post}_t$ represents the moderating effect used to assess how government–enterprise collusion suppresses the influence of the low-carbon policy on corporate innovation. The definitions of all other variables remain aligned with those established in the baseline regression.

4. Empirical Analysis

4.1. Descriptive Statistics

Descriptive statistics reveal that the treatment group constitutes 46.41% of the sample data, as shown in

Table 1. Descriptive statistics.

Statistic	N	Mean	SD	Min	Max
did	49,754	0.4641	0.4987	0	1
GIO	39,250	0.7228	3.2201	0.0000	33.0000
GUO	39,250	1.9950	6.5225	0.0000	59.0000
flever	44,758	1.3445	1.0821	0.2760	10.9719
tobinQ	38,660	2.0560	1.4500	0.8270	11.7247
roa	49,477	0.0538	0.0827	−0.4348	0.3488
size	40,387	22.1266	1.5037	18.7833	28.3562
lnage	39,674	2.0257	0.9372	0.0000	3.3322
Cpatial	40,347	3.1874	5.3767	0.3124	47.1867
lnGDP	49,727	10.4251	0.8072	7.3454	11.7338
Ur	49,754	0.6551	0.1378	0.3162	0.8957
FDI	45,126	0.0257	0.0148	0.0007	0.1048
Indus	49,754	0.8522	0.3169	0.1907	1.6677

. Excluding the impact of pilot provinces, only 75 cities were designated as pilot cities, representing about one-quarter of all cities. The third batch of pilot projects commenced in 2017, resulting in a relatively larger number of treatment cities with shorter treatment durations, thus encompassing a significant portion of listed companies. This underscores the relevance of studying the effects of low-carbon city policies on corporate green innovation. Regarding green patent performance, the mean for green invention innovation is lower than that for green utility model innovation. This aligns with findings from the European Patent Office (EPO), which indicates that the number of green utility model patents in Europe exceeds that of green invention patents by more than double.

4.2. Return to Baseline

Table 2. Benchmark regression results.

	(1)	(2)	(3)	(4)
	GIO	GUO	GIO	GUO
LCC	0.8772 **	0.7043 **	0.8917 **	0.6995 ***
	(0.3451)	(0.2877)	(0.3736)	(0.2537)
cons	0.6198 ***	1.9874 ***	−7.5000	−17.0016 ***
	(0.1549)	(0.1292)	(8.3936)	(5.5764)
control variable	NO	NO	YES	YES
fixed effect	YES	YES	YES	YES
N	40,292	40,292	40,292	40,292
R2	0.596	0.604	0.596	0.604

Note: *, **, *** denote at 10%, 5%, 1% significance levels, respectively; robust t-values in parentheses.

illustrates the baseline regression results regarding the impact of LCCP policies on corporate green innovation. Columns (1) and (2) present regression outcomes without control variables, indicating a significantly positive regression coefficient for the LCCP policy. This finding implies an enhancement in companies’ capabilities for green technology research and development, subsequently bolstering their green innovation capacity. Columns (3) and (4) display regression results after the inclusion of control variables, which continue to be significant at the 1% level and are consistent with prior findings. Notably, all models incorporate fixed effects at the enterprise, year, and industry levels, with robust standard errors clustered at the enterprise level.

From the regression results, the implementation of the LCCP policy shows a greater enhancement effect on green invention innovation. This is because green invention innovation typically involves improvements or innovations to existing technologies, possessing higher levels of novelty and practicality. In contrast, the green utility model patents often represent modifications or incremental innovations to existing technologies, which are generally easier to apply for and do not require thorough justification of the novelty of the innovation, though the actual effects may not be ideal. The LCCP policy creates an innovation-friendly environment through a systematic framework, complex technological requirements, experimentation, and flexibility, encouraging companies and research institutions to pursue bold green invention innovations. Moreover, long-term policy guarantees, market incentives, collaboration between public and private sectors, and the establishment of regional innovation ecosystems reduce the risks and increase the returns associated with invention innovations, making them more effective and impactful compared to innovations in green utility models.

4.3. Parallel Trend Test

As this paper employs the difference-in-differences (DID) method to evaluate the specific effects of LCCP policies, it is crucial to satisfy the assumptions necessary for DID application. In particular, the trends in green innovation for both the treatment and control groups should be consistent prior to the policy intervention. To examine this parallel trend assumption, the paper constructs a time dummy variable based on whether the cities of the enterprises are designated as policy pilot cities and the years subsequent to the pilot’s implementation. The annual effects of the LCCP policy are illustrated in Figure 2. Before the policy’s implementation, the interaction term between the year dummy variable and the treatment group dummy variable was found to be insignificant, suggesting that both groups exhibited comparable trends prior to the intervention.

As shown in Figure 2, the LCCP policy has a lagged effect on green invention innovation. Firstly, the technologies involved in green invention innovation are often in emerging fields or require interdisciplinary integration, meaning that the application and maintenance of patents necessitate high levels of expertise in both technology and law. These technologies typically involve high levels of technical difficulty and cost, requiring substantial investments in human, material, and financial resources during the research, development, and experimentation phases. In contrast to green invention patents, which tend to increase with the implementation of the policy over the years, the trend for green utility innovation is not significant. This is due to its lower technological content, which fails to meet the effects expected from long-term innovation strategies. Moreover, low-tech R&D innovations often trigger imitation by other companies, undermining a firm’s ability to maintain a monopoly on product innovation, thus reducing the motivation for companies to engage in green utility R&D innovation.

4.4. Bacon Decomposition

To obtain an unbiased average treatment effect from a two-way fixed effects regression, it is necessary to satisfy the parallel trends assumption and the assumption of no heterogeneity in treatment effects across groups and different time points. The parallel trends hypothesis has been previously established. Given that this paper involves three batches of treatment trials, there is interference from bad control groups and negative weights while conducting weighted averages [61]. Therefore, this paper employs the Bacon decomposition to identify the degree of bias. Since Bacon decomposition requires strong panel data, this paper selects enterprises with complete data from 2007 to 2021 as the research sample. The results are shown in

Table 3. Goodman-Bacon decomposition results.

	Estimates		Weight
	GIO	GUO
“Not yet treated group” as control group	0.0217	0.0222	13.75%
“Early treatment group” as control group	0.0232	0.0225	20.75%
“Never treated group” as control group	0.4749	0.6906	65.5%

, where bad control groups account for only 20.75%, and the proportion of negative weights is relatively small. Thus, it is concluded that this will not significantly interfere with the regression results of this paper.

4.5. Robustness Check

4.5.1. PSM-DID

By comparing the kernel density plots before and after matching, the study found no significant differences after matching, indicating that the PSM-DID method effectively eliminated the influence of heterogeneous covariates, thereby reducing the interference of sample differences on the regression results. The regression results after PSM-DID treatment are shown in

Table 4. Regression results for PSM-DID and replacement of explanatory variables.

	(1)	(2)	(3)	(4)	(5)
	GIO	GUO	GII	GUI	GUI
LCC	0.3075 ***	0.5478 ***	0.2770 *	0.0766
	(0.0894)	(0.1967)	(0.1653)	(0.1045)
LLCC					0.2892 ***
					(0.0850)
cons	−5.4062 **	−21.2356 ***	−27.7067 ***	−1.8513 ***	−5.0530 **
	(2.3636)	(4.5656)	(4.9067)	(2.8076)	(2.3558)
control variable	YES	YES	YES	YES	YES
fixed effect	YES	YES	YES	YES	YES
N	30,068	30,068	30,068	30,068	28,859
R2	0.636	0.660	0.514	0.492	0.652

Note: *, **, *** denote at 10%, 5%, 1% significance level, respectively; robust t-values in parentheses.

. Columns (1) and (2) exclude the impact of sample selection bias on the regression results, which are consistent with the baseline regression results. The LCCP policy significantly positively influences both green invention innovation and green utility model innovation at the 1% significance level, demonstrating that the LCCP policy promotes corporate green innovation.

4.5.2. Placebo Testing

To eliminate the influence of other factors on the regression results, this study used random sampling to select the treatment group and randomly selected policy implementation time points from 2007 to 2021, conducting 500 regression analyses. Kernel density plots were drawn based on the regression coefficients, as shown in Figure 3. Since the false policy times and false treatment groups were randomly generated, it can be concluded that the differences in the LCCP policy’s impact on enterprises are not significant, meaning that the regression coefficients for the false treatment variables are close to zero. Conversely, if the regression coefficients deviate from zero, it suggests a bias in the model. The majority of regression coefficients cluster around zero, indicating that the LCCP policy’s effect on corporate green innovation is not coincidental, demonstrating a real effect.

4.5.3. Replacement of Explanatory Variables

To validate the robustness of the results, this paper replaces the variable for green technological innovation with the count of green patent applications rather than granted patents. The findings are presented in Table 4, particularly in columns (3) and (4). In column (3), it is evident that the implementation of the LCCP policy has led to an increase in the number of applications for green invention patents. In contrast, the regression results in column (4) are not statistically significant, suggesting that the policy does not have a substantial impact on applications for green utility model patents. Two explanations are offered for this observation. First, environmental regulations tend to emphasize the development of new technologies rather than the improvement of existing ones. Consequently, under government signals, there has been a shift in the internal structure of green technological innovation within enterprises, which are inclined to invest more in new technology development in anticipation of monopolizing future green technology markets. Second, this paper posits that government environmental regulations exhibit a lag effect, necessitating time for implementation at the micro-enterprise level. To address this issue, the paper includes a lagged variable for the number of applications for green utility model patents in the original model. The results displayed in column (5) indicate a significant positive effect, suggesting that previous applications have a substantial influence on the current count of applications for green utility model patents.

4.5.4. Controlling for Fixed Effects

Building on the original model, this paper incorporates fixed effects to further validate the findings. Given that various industries experience differing levels of impact from environmental policies over time, the introduction of interaction fixed effects between industry and time serves to mitigate potential biases arising from these intersecting factors. Columns (1) and (2) in

Table 5. Other robustness tests.

	(1)	(2)	(3)	(4)
	GIO	GUO	GIO	GUO
LCC	0.6766 **	0.6907 ***	0.5468 *	0.4514 *
	(0.2685)	(0.2632)	(0.3039)	(0.2539)
cons	−7.0863	−15.0882 **	−16.5663 **	−24.9221 ***
	(6.9378)	(5.9207)	(7.9565)	(5.3909)
control variable	YES	YES	YES	YES
fixed effect	YES	YES	YES	YES
Time × industry fixed effects	YES	YES	NO	NO
N	30,133	30,133	23,256	23,256
R2	0.696	0.698	0.473	0.539

Note: *, **, *** denote at 10%, 5%, 1% significance level, respectively; robust t-values in parentheses.

display the results with time-industry fixed effects controlled. The outcomes remain consistent with the baseline regression, indicating that the implementation of the LCCP policy continues to enhance corporate green innovation.

4.5.5. Exclusion of Other Policy Interferences

In October 2007, the State Council officially released the “Development Plan for the Energy Conservation and Environmental Protection Industry”, outlining the Chinese government’s strategy for the development of energy conservation and environmental protection sectors. This plan established the overall investment scale for these industries, detailing specific investment allocations across various fields and the construction of an energy conservation and environmental protection industrial system. The year 2008 marked a period of significant events, including the financial crisis and the Olympic Games. The financial crisis led to reduced market demand while tightening financial markets and credit contractions severely restricted corporate funding and human resources, subsequently diminishing market demand for green initiatives and corporate capacities for green development. To eliminate potential confounding effects from concurrent policies on regression results, this paper modifies the sample period by excluding data from 2007 to 2008. The regression results, presented in columns (3) and (4) of Table 5, remain significant even after accounting for these policy influences, confirming the substantial positive impact of the LCCP policy on corporate green innovation.

5. Further Analysis

5.1. Heterogeneous Effects Analysis

5.1.1. Regional Heterogeneity

Given the disparities in economic development, resource endowments, and industrial structures among regions, the efficacy of government environmental regulations varies, leading to differing impacts on levels of green innovation. This paper categorizes provinces into Eastern and Central/Western regions. As illustrated in columns (1) to (4) of

Table 6. Regression results for regional heterogeneity.

	(1)	(2)	(3)	(4)
	GIO	GIO	GUO	GUO
	Eastern Region	Central and Western Region	Eastern Region	Central and Western Region
LCC	0.9446 *	0.2139	0.2185	0.5138
	(0.5023)	(0.3207)	(0.2733)	(0.3987)
cons	−12.1444	−11.8871 ***	−7.6990	−28.4639 ***
	(21.9339)	(4.1235)	(10.8242)	(7.0556)
control variable	YES	YES	YES	YES
fixed effect	YES	YES	YES	YES
N	21,163	9,046	21,163	9,046
R2	0.629	0.422	0.677	0.555

Note: *, **, *** denote at 10%, 5%, 1% significance level, respectively; robust t-values in parentheses.

, the policy effects markedly bolster green innovation in the Eastern region, a trend not observed in non-Eastern areas. Specifically, columns (1) and (2) reveal that the LCCP policy exerts a positive influence on green invention innovation in the Eastern region; however, this effect lacks statistical significance in the Central/Western regions. Additionally, columns (3) and (4) indicate that the regression results for green utility model innovation in non-Eastern regions also show no significant outcomes. The Eastern region, being more developed, boasts stronger financial reserves and technical capabilities, which afford it certain advantages in environmental governance and research innovation. In contrast, while non-Eastern regions are endowed with abundant natural resources, their relatively lower levels of environmental pollution lead local governments to prioritize economic development over environmental protection. Moreover, due to insufficient technological advancement and governance capacity, non-Eastern regions may incur greater costs and time in adapting to government environmental regulations, resulting in less pronounced initial effects of policy implementation.

5.1.2. Nature of Business

The ownership structure of enterprises may significantly influence the progression of green innovation. To investigate this, the sample is categorized into state-owned enterprises (SOEs) and non-state-owned enterprises (NSOEs). The regression results displayed in

Table 7. Regression results for heterogeneity in nature of firms.

	(1)	(2)	(3)	(4)
	GIO	GIO	GUO	GUO
	State-Owned Enterprises	Non-State Enterprise	State-Owned Enterprises	Non-State Enterprise
LCC	1.4792 **	0.1353	0.9874 **	0.1974
	(0.7235)	(0.1310)	(0.4810)	(0.2314)
cons	−18.6830	−7.5953 **	−18.9864 *	−26.2800 ***
	(19.5749)	(2.9954)	(10.9121)	(5.7297)
control variable	YES	YES	YES	YES
fixed effect	YES	YES	YES	YES
N	13,044	16,421	13,044	16,421
R2	0.619	0.562	0.687	0.600

Note: *, **, *** denote at 10%, 5%, 1% significance level, respectively; robust t-values in parentheses.

indicate that the LCCP policy substantially enhances green invention innovation in SOEs, whereas its impact on NSOEs lacks statistical significance. A similar pattern emerges in the realm of green utility model innovation, where significant effects are observed for SOEs but not for NSOEs. These results imply that the policy effectively strengthens the green innovation capabilities of enterprises, particularly within the state-owned sector. From an internal governance perspective, the management of SOEs, appointed by the state, allows for more adaptive adjustments to investment strategies in response to policy support. This enables swift resource allocation favoring green innovation. Furthermore, as “pioneers” in the development of a green economy, SOEs set a demonstrative example in green R&D, characterized by more transparent investment behaviors that attract scrutiny and oversight from various stakeholders. In contrast, the regulatory pressure and social responsibility of NSOEs are comparatively weaker. In their pursuit of profit maximization, these enterprises often struggle to balance environmental protection, leading to slower advancements in green technology. Although NSOEs possess significant potential for improvement, their lack of a robust regulatory framework and supportive policies hampers their ability to markedly enhance green innovation capabilities in the short term.

5.1.3. Industry Technological Content

The differences in technological content between industries can have a significant impact on the energy utilization efficiency of enterprises. In this study, we follow the approach of Li et al. (2016) [62] and classify industries into high-tech and low-tech sectors based on the OECD (2003) classification standards. From the regression results presented in

Table 8. Regression results of industry technological heterogeneity.

	(1)	(2)	(3)	(4)
	GIO	GIO	GUO	GUO
	High-Tech Industry	Low-Tech Industry	High-Tech Industry	Low-Tech Industry
LCC	0.6984 **	1.0382 *	0.6995	0.5906 **
	(0.3339)	(0.5426)	(0.4277)	(0.2836)
cons	−13.9005 **	−4.1171	−18.7678 ***	−16.0285 *
	(5.4750)	(13.2505)	(6.2814)	(8.1799)
control variable	YES	YES	YES	YES
fixed effect	YES	YES	YES	YES
N	10,404	19,754	10,404	19,754
R2	0.479	0.655	0.810	0.543

Note: *, **, *** denote at 10%, 5%, 1% significance level, respectively; robust t-values in parentheses.

, it is evident that the LCCP policy positively influences green invention innovation in both high-tech and low-tech industries. While the policy has a positive impact on green practical innovation in low-tech sectors, its effect on green practical innovation in high-tech industries is not significant. Regarding the results of green invention innovation, both high-tech and low-tech enterprises experience positive effects; however, the benefits vary in degree, with a more pronounced effect observed in low-tech industries. This disparity may stem from a higher dependency on policy in aspects such as technological innovation, environmental investment, financing constraints, and market demand within these industries. Low-tech sectors typically exert a greater environmental impact and are relatively lagging in technological innovation. Consequently, the implementation of the LCCP policy provides them with opportunities for transformation and upgrading, encouraging them to reduce environmental impacts through green innovation while enhancing energy efficiency and resource utilization. Moreover, these sectors may be more susceptible to fluctuations in market demand, and the policy’s guidance on the demand side could have a more direct influence on green innovation within these industries. Additionally, the policy indirectly promotes innovation by increasing environmental investments, which is particularly crucial for low-tech sectors that often face resource and technological shortages. The augmented environmental investments can provide essential financial support for these enterprises to conduct research and development and apply green technologies. Furthermore, the effects of green practical innovation are significant in low-tech industries but not in high-tech ones. Given that the application and authorization thresholds for green practical patents are relatively low, low-tech industries, which typically lack innovative capacity, are less willing to assume high risks and prefer cost-effective and quick-return green practical patents to enhance their technological content and improve their green competitiveness as swiftly as possible.

5.2. Government-Business Relations

Government–business collusion refers to a relationship in which the government and enterprises conspire to achieve mutual benefits. When the collaboration between government and business becomes overly close, it can have a suppressive effect on corporate green technological innovation. Such collusion may lead to excessive government intervention in the market, negatively impacting market competition. Overly close ties between the government and certain enterprises or industries result in preferential treatment, rather than equitable support for all businesses. Additionally, government–business collusion may compromise the safety and sustainability of green technological innovations [46,63]. A close relationship between the government and enterprises can lead to inadequate regulation of corporate green innovation, weakening oversight on the safety and sustainability of these innovations.

However, changes in local officials can disrupt the long-standing collusive relationships between government and enterprises. New officials, motivated by their performance assessments and the legitimacy of local governance, typically scrutinize and readjust their interactions with enterprises more rigorously. In this context, companies can no longer rely on the privileged resources gained through relationships established with their predecessors, which exposes them to a more equitable competitive environment in terms of policies and markets, thus suppressing the incentivizing effect of such collusion on corporate green innovation. This study manually collects data on changes in city party secretaries to examine the impact of official transitions on corporate green innovation.

To address this issue, the paper constructs a triple-difference model to explore the suppressive effects of government–business collusion on corporate green innovation. The results, as shown in

Table 9. Government–business collusion and changes in officials.

	(1)	(2)	(3)	(4)
	GIO	GUO	GIO	GUO
LCC*ITR	−0.4384 **	−0.5983 **
	(0.1903)	(0.2750)
LCC*ITR*Mpc			−0.1464	−0.0418
			(0.2790)	(0.2049)
Mpc			0.1727 **	0.2632 ***
			(0.0701)	(0.0906)
LCC	1.1567 ***	1.0614 ***	0.9349 **	0.7537 ***
	(0.4012)	(0.3233)	(0.3889)	(0.2588)
ITR	0.0069	0.0055	−0.1681	−0.2515 *
	(0.1276)	(0.1444)	(0.1806)	(0.1386)
cons	−7.5618	−17.0815 ***	−7.0449	−16.7626 ***
	(8.3295)	(5.5861)	(8.2662)	(5.5768)
control variable	YES	YES	YES	YES
fixed effect	YES	YES	YES	YES
N	30,222	30,222	30,186	30,186
R2	0.613	0.663	0.613	0.663

Note: *, **, *** denote at 10%, 5%, 1% significance level, respectively; robust t-values in parentheses.

, indicate a significant inhibitory effect of collusion on green innovation in columns (1)–(2). This suppression may stem from the fact that collusion between government and enterprises often results in governmental market intervention and distortion, thereby weakening market competition and stifling green innovation. While such collaboration may provide resources like funding, technical consulting, and policy support to promote green practical innovations, it can also lead the government to prioritize these benefits for colluding firms over others. Consequently, this preferential treatment creates increased competitive pressure on non-colluding enterprises, adversely impacting their capacity for green innovation. Columns (3) to (4) report the impact of changes in local officials on the innovation suppression effect of government–business collusion. The results indicate that changes in local officials can reduce the influence of government–business collusion on corporate innovation and, to some extent, promote green innovation among enterprises.

5.3. Sustainable Development

Under the impetus of government environmental regulations, corporate green innovation contributes to pollution reduction, emissions control, and the promotion of industrial green transformation, ultimately facilitating sustainable development. As enterprises undergo green transitions, their financial performance and development quality also improve. First, driven by technological advancements, firms experience a reduction in energy consumption costs, optimize resource allocation, and enhance the returns generated from lower unit costs [64]. Enterprises participating in the LCCP program are typically subject to higher environmental standards, making it imperative for them to emphasize the sustainability and environmental friendliness of their supply chains. This leads to the selection of upstream and downstream partners that prioritize green materials and products, thereby enhancing corporate image, attracting environmentally conscious consumers, and improving financial performance. Secondly, the government stimulates corporate environmental social responsibility (ESR) performance by sending positive environmental signals. Under stringent environmental regulations and policies, as well as clearly defined environmental protection requirements and standards, enterprises are encouraged to proactively fulfill their environmental responsibilities and engage in environmental protection initiatives [30,65]. On the one hand, financial and tax incentives are offered to boost ESR performance, such as tax deductions or subsidies for environmental investments, encouraging enterprises to invest in eco-friendly technologies and equipment upgrades. On the other hand, the government strengthens transparency and regulation of environmental information, increasing the visibility of corporate environmental performance. By disclosing environmental data and emissions information, the government can guide businesses to take greater responsibility, foster self-regulation, and encourage continuous improvement. Accordingly, this paper constructs the following model:

$$S{d}_{itd}={\alpha }_0+{\alpha }_{1}LC{C}_{itd}+{\alpha }_2{\mathrm{Controls}}_{it}+u_i+u_t+u_d+{\epsilon }_{itd}$$

(4)

$$\begin{array}{cc}\hfill S{d}_{itd}=& {\beta }_0+{\beta }_{1}LC{C}_{itd}+{\beta }_{2}LC{C}_{itd}\ast IT{R}_{itd}+{\beta }_{3}IT{R}_{itd}+\hfill \\ \hfill & {\beta }_{4}Control{s}_{it}+u_i+u_t+u_d+{\epsilon }_{itd}\hfill \end{array}$$

(5)

In this context, $S{d}_{itd}$ represents the sustainability indicators, which include financial performance (Fp) and environmental social responsibility performance (EGF). The formula employs a difference-in-differences (DID) model to examine the impact of policy effects on corporate sustainability. Additionally, an interaction term $LC{C}_{itd}\ast IT{R}_{itd}$ is introduced to identify the influence of government–enterprise collusion on sustainability outcomes.

This paper utilizes the indicators developed by Xie and Zhu (2021) [66], employing financial performance and environmental social responsibility performance as metrics to evaluate corporate sustainability. Financial performance is assessed using the return on total assets, calculated as (Interest Expense + Total Profit)/ Total Assets. For environmental social responsibility performance, the relevant word frequency from listed companies’ annual reports is extracted and subsequently analyzed using the entropy method. Specifically, positive indicators related to corporate environmental social responsibility are selected, and their weighted sum is calculated using the entropy method. The regression results displayed in columns (1) and (2) of

Table 10. Sustainable development.

	(1)	(2)	(3)	(4)
	Fp	EGF	Fp	EGF
LCC*ITR			0.0180 ***	−0.0304 ***
			(0.0016)	(0.0029)
LCC	0.0094 ***	0.1150 ***	−0.0015	0.1335 ***
	(0.0009)	(0.0045)	(0.0010)	(0.0045)
ITR			−0.0038 ***	0.0008
			(0.0003)	(0.0020)
cons	0.0407	0.0402	0.0457	−0.0436
	(0.0290)	(0.0637)	(0.0290)	(0.0628)
control variable	YES	YES	YES	YES
fixed effect	YES	YES	YES	YES
N	13,711	29,391	13,711	29,391
R2	0.867	0.641	0.869	0.643

Note: *, **, *** denote at 10%, 5%, 1% significance level, respectively; robust t-values in parentheses.

reveal that the regression coefficients are both positive and statistically significant at the 1% level, demonstrating that the LCCP policy effectively promotes corporate sustainability.

Building on this foundation, the paper further introduces the moderating effect of government–business collusion. The regression results, as shown in columns (3) and (4) of Table 10, indicate that, under the influence of collusion, corporate financial performance significantly improves, while environmental responsibility performance significantly declines. In this context, the prioritization of economic interests leads to an emphasis on cooperation for economic development and profit maximization. Consequently, firms may focus more on financial performance, resulting in reduced investment and commitment to environmental responsibilities, which may be neglected or sacrificed. Additionally, the dynamics of interest conflict and power imbalance come into play; if a firm possesses greater power and influence within the collusion, it might manipulate policies or circumvent regulations to pursue its own interests, thereby neglecting its environmental responsibilities. Moreover, the presence of regulatory oversight failures or protective measures further exacerbates the motivation for firms to disregard their environmental responsibility performance.

6. Conclusions and Policy Recommendations

6.1. Conclusions

As global attention to low-carbon economies and green transitions continues to intensify, the importance of optimizing policy support and enhancing corporate environmental awareness has become increasingly evident. However, there remains room for improvement in the implementation of current low-carbon policies, particularly regarding enterprises’ deficiencies in technological innovation, which limits their capacity for structural optimization and emission reduction. In some regions, local governments, while pursuing the dual goals of economic growth and carbon reduction, may introduce projects or enterprises that superficially comply with low-carbon requirements but fail to meet actual environmental protection standards. This approach results in negligible reductions in pollutant emissions and limited improvements in regional environmental quality. To promote genuine low-carbon development, this study explores the impact of environmental policies on corporate green innovation from both theoretical and empirical perspectives, aiming to provide policymakers with a more scientifically grounded basis for decision-making.

This study employs sample data from China’s A-share listed companies from 2007 to 2021, utilizing the LCCP policy as a quasi-natural experiment to empirically investigate the impact of government environmental regulation on corporate green innovation. The LCCP initiative has achieved positive effects in enhancing the efficiency of the green economy, contributing not only to the realization of energy-saving and emission reduction targets but also to the sustainable development of cities. First, the research finds that the LCCP policy effectively stimulates enterprises to engage in green innovation and research and development. This is evidenced by an increase in the number of authorized green invention patents and green utility model patents, a result that remains significant after a series of robustness checks. Second, heterogeneity analysis indicates that, overall, the effects of corporate green innovation are more pronounced in the eastern regions and among state-owned enterprises, while low-tech industries exhibit more notable performance in green innovation due to their developmental flexibility. The results of the moderating effect analysis show that government–business collusion suppresses enterprises’ efforts in green invention and utility model innovation. However, when local officials change, the collusive relationships between government and enterprises are disrupted, thereby enhancing the impact of the policy on corporate green innovation. Further research indicates that stimulated by this policy, corporate green innovation activities significantly improve enterprises’ sustainable development. Under the moderating effect of government–business collusion, the financial performance of enterprises increases significantly, while their environmental social responsibility is suppressed.

Although environmental policies can enhance corporate levels of green innovation to a certain extent, integrating environmental governance into long-term development strategies and thus promoting sustainable development, existing policies still exhibit shortcomings in their incentive mechanisms. These deficiencies may lead to inadequate investment by enterprises in environmental governance, hindering a comprehensive implementation of green transformation. Furthermore, the profit-driven interactive relationships between government and enterprises can also influence the actual effectiveness of policy execution, resulting in behaviors that deviate from the original intentions. For instance, some companies may adopt superficial environmental measures solely under policy pressure, without genuinely internalizing environmental objectives as core aspects of their management. Therefore, during the implementation of these policies, it is essential to establish a sound, rational, and equitable incentive and constraint mechanism to effectively mitigate the occurrence of opportunistic behavior and ensure that the objectives of environmental policies are genuinely achieved.

6.2. Policy Recommendations

(1) Enhance the guiding role of low-carbon city policies and comprehensively advance the implementation of LCCP initiatives. The low-carbon city policy, formulated by the central government and executed at the local level, is well-suited to China’s unique national conditions. The state strongly advocates for industrial structure reform, and the LCCP policy serves as a vital support mechanism for industrial growth, facilitating the optimization and upgrading of the industrial structure. Cities that have not effectively implemented the policy should carefully analyze the challenges faced and enhance oversight and enforcement measures to ensure compliance and effectiveness. As a vital component of policy execution, supervision must be reinforced to address issues promptly. Furthermore, expanding the promotion of low-carbon city policies and using comparative analyses among different cities can aid others in exploring low-carbon transformation strategies, ensuring that the policy has a nationwide impact.

(2) Develop differentiated policies. Given the diverse circumstances of different cities and enterprises, policies should be tailored to the specific characteristics of each locale and business. For smaller cities, which may have lagging development and weaker economic foundations, targeted support should be provided to encourage green technology innovation among enterprises. In contrast, larger cities could benefit from a shift in traditional policy frameworks, focusing on stimulating corporate innovation. Attention should be directed toward high-pollution, high-energy-consumption enterprises, ensuring their energy-saving and emission reduction targets meet regulatory standards through on-site assessments and appropriate guidance to facilitate industrial structure upgrades. Significant disparities are evident between state-owned and non-state-owned enterprises, particularly concerning resource allocation and policy support. Therefore, the government must incentivize non-state-owned enterprises to align their development pace with that of state-owned enterprises, ensuring equitable resource distribution while emphasizing the coordination of human and natural resources.

(3) Improve government–enterprise relations to achieve win-win development. In the implementation of LCCP policies, it is essential to leverage the regulatory and execution roles of local governments fully. Timely oversight of policy implementation should involve refining industrial policies and tailoring specific measures for different enterprises to encourage increased investment in green innovation. At the same time, businesses should adeptly utilize available policies and government subsidies to generate effective outputs. Any issues arising during policy implementation must be addressed promptly through open communication, ensuring that adjustments to government policies are swiftly applied at the enterprise level to mitigate the interference caused by information asymmetry. Furthermore, it is crucial to establish a government–enterprise collaboration model for green technology innovation that is driven by market demand. When executing policies, flexibility is vital, allowing for the adaptation of measures based on the specific realities and characteristics of the relevant enterprises.

(4) Fully utilize internal resources to achieve green innovation and promote sustainable development. Under the framework of LCCP policies, enterprises must first concentrate resources to overcome core green technologies, aligning their efforts with a primary focus on green and low-carbon initiatives. This involves nurturing enterprises that specialize in green technological innovation, thereby enhancing their capacity to originate such innovations. Concurrently, companies should expand the production and promotion of green and low-carbon products, leveraging the strategy of increasing domestic demand to actively guide and cultivate diverse forms of green consumer demand, which in turn, stimulates the emergence of innovative green technologies. Moreover, enterprises need to strengthen the integration of industrial chains, innovation chains, funding chains, and talent chains, fostering collaborative interactions among green technology innovation firms, educational institutions, research organizations, intermediary bodies, and financial institutions. This cooperation will facilitate innovation across the entire value chain of the green technology market. By implementing these critical steps, enterprises can not only respond to policy calls but also gain a competitive edge in the market, ultimately achieving high-quality development.

6.3. Limitations and Future Research

Due to limitations in data accessibility, this study finds it challenging to intricately depict the specific relational networks among enterprises. Furthermore, the focus of this research does not extend to the rent-seeking behaviors between other entities beyond the government and their impact on corporate green innovation. Future research endeavors will strive to broaden the analytical perspective, aiming to investigate the potential influences of various stakeholders, such as banks and the general public, on corporate green innovation. This exploration seeks to establish a more comprehensive analytical framework.