Low-Carbon City Pilot Policy, Digitalization and Corporate Environmental and Social Responsibility Information Disclosure

Abstract

Low-carbon development is an important area that must be focused on in order to cope with climate change. Based on the institutional theory, this paper uses a sample of Chinese A-share listed firms from 2008 to 2021 and constructs a difference-in-differences model to examine the impact of low-carbon city pilot policy on corporate environmental and social responsibility information disclosure. The results show that the implementation of low-carbon city pilot policy in a region where companies are located significantly promotes corporate environmental and social responsibility information disclosure, and the degree of digital transformation of enterprises in the pilot region has a moderating effect on it. The mechanism analysis reveals that the policy promotes the corporate environmental and social responsibility information disclosure primarily by enhancing the environmental performance and increasing media attention, and heterogeneity analysis shows that when the enterprise has green investors or belongs to an industry with low carbon emissions, the policy has a more significant impact. Additionally, the study finds that the low-carbon city pilot policy has a positive impact on the quality of corporate environmental information disclosure. In terms of the goals of Carbon Peaking and Carbon Neutrality, this study provides new evidence on how low-carbon city pilot policy influences corporate environmental and social responsibility, offering valuable insights for advancing the country’s low-carbon development agenda.

1. Introduction

Low-carbon development forms part of the basic approach to China’s ecological civilization construction and is an important area of focus for guiding international cooperation in addressing climate change. In order to achieve China’s commitment to the 2030 carbon peak target and realize the comprehensive low-carbon transformation of the economy and society, the National Development and Reform Commission began to implement the low-carbon city pilot policy in 2010 and expanded the scope of the pilot in 2012 and 2017. Carrying out low-carbon city pilots is conducive to exploring practical and distinctive low-carbon development models in different regions, natural conditions and development bases in China, and promoting China’s economic low-carbon transformation and accelerating the conversion of new and old drivers. The low-carbon city pilot policy aims to achieve energy conservation and emission reduction by strengthening the concept of low-carbon development, exploring low-carbon development models, and accumulating experience for China’s low-carbon development. Examining the implementation effects of this pilot policy provides reference for China to give attention to the effectiveness of the policy and comprehensively promote low-carbon development.

Under the policy, the focus of all sectors of society will be on the performance of enterprises in fulfilling their environmental responsibilities. With the increasing number of low-carbon policies in China, more and more enterprises have begun to compile and publish social responsibility reports to disclose relevant information. If an enterprise chooses to publish a social responsibility report to disclose its own environmental information separately, it can show that the enterprise attaches more importance to its own low-carbon cause. As a comprehensive environmental regulation measure [1], the low-carbon city pilot policy plays a key role in promoting enterprises to achieve low-carbon emission reduction. Existing studies have found that whether the concept of low-carbon emission reduction can be effectively transformed into the optimization of corporate capital structure depends on the response strategy of the enterprise [2]. The implementation of the low-carbon city pilot policy will significantly promote the green innovation of enterprises [3], and will have a greater role in promoting green innovation for enterprises in cities with high carbon emission reduction potential [4]. Low-carbon city construction promotes the adjustment of the industrial structure and optimization of the total factor productivity of enterprises, improves carbon emission efficiency [1], and achieves energy conservation and emission reduction in a more environmentally friendly way of resource allocation. Scholars have explored the policy effects of low-carbon city pilot projects from different aspects, ecological performance [5], economic performance [6], and employment [7]. Research on corporate environmental and social responsibility information disclosure mainly focuses on corporate competitive advantage and financial performance [8], recognizing that corporate social responsibility information disclosure positively impacts factors such as corporate value [9], investment efficiency [10], and the quality of earnings management [11]. The literature related to environmental regulation and low-carbon city construction mainly focuses on macro-level research, and micro-evidence from enterprises with less frequency. Macro-level studies focus on the impact of low-carbon city construction on air quality, while micro-level studies focus on the relationship between low-carbon city pilot policies and green innovation, which is similar to the micro-level research on environmental regulation. Among the studies related to corporate social responsibility, few studies related to environmental social responsibility have explored the motivations of low-carbon policies and their internal impact paths.

The paper adopts the data of A-share listed companies in Shanghai and Shenzhen Stock Exchanges from 2008 to 2021, and constructs a difference-in-differences two-way fixed model based on institutional theory to empirically test the impact of low-carbon city pilot policy on corporate environmental and social responsibility information disclosure and verify its mechanism of action. It also examines the moderating effect of corporate digital transformation on this impact and analyzes the heterogeneity of impact of factors green investors and different carbon emission level industries.

The innovation of the paper is mainly reflected in three aspects. First, based on institutional theory, the environmental responsibility spillover effect of low-carbon city pilot policy in China is demonstrated from both theoretical and empirical aspects. The results show that the implementation of low-carbon city pilot policy has a positive impact on the disclosure of environmental social responsibility information of listed companies in China, which provides new evidence for a comprehensive evaluation of the environmental effects of low-carbon city pilot policy. Second, there are few studies on the impact of the low-carbon city pilot policy on corporate environmental and social responsibility information. This paper analyzes the impact of the pilot policy in China on the environmental and social responsibility information disclosure motivation of listed companies, and to a certain extent provides empirical evidence for the conclusion about corporate environmental and social responsibility information; this enriches the research perspective of the spillover effect of the low-carbon city pilot policy in China from the perspective of corporate digital transformation. Third, the existing literature on the transmission mechanism of the environmental policy spillover effect focuses on external channels represented by financing constraints [12], but the financing constraints faced by large-scale enterprises may not be obvious, and the environmental performance and media attention should also be considered as mechanisms. Based on theoretical analysis, the paper empirically verifies these two mechanisms and explores heterogeneous effects from the perspectives of corporate industry type and green investors, enriching the understanding of how the low-carbon city pilot policy influences corporate environmental and social responsibility information disclosure.

2. Institutional Background and Theoretical Hypotheses

2.1. Policy Background

The low-carbon city pilot policy is an urban-level environmental regulation policy proposed to implement China’s climate action goals. The national level has not set specific goals for the low-carbon city pilot, such as the peak time of carbon emissions and emission standards for different industries. Instead, this power has been delegated to each pilot government, and each pilot can promote low-carbon work according to its own situation. In July 2010, China’s National Development and Reform Commission issued the “Notice on Carrying out Pilot Work in Low-Carbon Provinces and Low-Carbon Cities”, identifying five provinces including Guangdong Province and eight cities including Shenzhen as the first batch of pilot areas. The second and third batches of pilot areas were announced in November 2012 and January 2017, respectively. The second batch of pilot areas included 29 provinces and cities including Beijing, Shanghai, and Hainan, and the third batch of pilot areas included 41 cities including Nanjing and Hefei and four districts and counties. In the process of building low-carbon cities, the three batches of pilot cities generally attach great importance to carbon emissions and have formulated a series of policies to promote low-carbon development. Some studies have confirmed that the pilot policy can significantly improve the carbon emission efficiency [13] and reduce city-level carbon emissions [14] and corporate carbon reduction performance [15].

2.2. Theoretical Hypotheses

The low-carbon city pilot policy can encourage enterprises to proactively disclose environmental and social responsibility information. Based on institutional theory, organizations should focus on the institutional perspectives of stakeholders who have a critical impact on their reputation and viability and seek legitimacy by proactively engaging in legitimizing behaviors [16,17]. On the one hand, firms could demonstrate their compliance with institutional requirements and enhance their legitimacy in the implementation of the low-carbon city pilot policy through proactively disclosing environmental and social responsibility information. Although environmental regulation brings compliance costs and constrains firms’ profit-maximizing behaviors through emission reduction targets and market-based mechanisms, these institutional pressures also encourage enterprises to adjust their production and operation strategies. It means that disclosure becomes a strategic response for enterprises. By actively releasing social responsibility information, firms can signal that they are fulfilling environmental obligations, reduce external legitimacy risks, and secure a more favorable position in the regulatory environment. On the other hand, the low-carbon city pilot policy, as an environmental regulation from the government, can constrain and motivate enterprises to a certain extent. This pilot policy has a certain degree of mandatory and normative nature. Under the guidance of the government, in compliance with relevant systems, and under the mandatory constraints of relevant laws, enterprises will improve their environmental protection behavior. According to Klassen and Whybark (1999) [18], environmental protection laws and regulations can promote enterprises to invest in corresponding environmental protection technologies to improve their environmental protection performance. The government plays an important role in improving corporate environmental protection behavior [19]. From the perspective of institutional constraints, the low-carbon city pilot policy may enable enterprises to passively disclose social responsibility information and prove their fulfillment of environmental social responsibility.

In addition, the fulfillment of social responsibility can improve corporate reputation, and a good reputation can help enterprises obtain social capital by transmitting “signals”. According to stakeholder theory, the government is one of the stakeholders of enterprises. When the government proposes to strengthen environmental protection in the pilot areas through the low-carbon city pilot policy, enterprises can cater to the government’s preferences by fulfilling their social responsibilities for the environment, enhance the legitimacy of the enterprise, establish a good reputation for the enterprise, and send “signals” to the government to obtain government support and resources. The view of the institutional isomorphism school also suggests that organizations can obtain institutional legitimacy from various stakeholders through institutional isomorphism. On the contrary, if the government clearly conveys its preference for environmental protection through the implementation of the low-carbon city pilot policy, but the enterprise ignores the fulfillment of its environmental social responsibilities, and this non-cooperative behavior is discovered by stakeholders, stakeholders may “punish” it by stopping or reducing its support for the enterprise [20]. In this case, the implementation of the low-carbon city pilot policy can also improve the level of environmental social responsibility fulfillment of enterprises to a certain extent. This inference is not limited to the government as a stakeholder. When other stakeholders show preference and attention to corporate environmental protection, the above inference also applies. Therefore, the following hypotheses are proposed:

Hypothesis 1 (H1).

The implementation of the low-carbon city pilot policy will promote the disclosure of social responsibility information by listed companies in the pilot areas.

Digital technology plays an important role in corporate environmental governance [17], and government governance is inseparable from the degree of match between corporate innovation and the changing information environment [21]. Many studies have found that digital technology plays an important moderating role in related research on corporate low-carbon development [22]; the increase in the intensity of environmental regulation can positively regulate the impact of media attention on corporate environmental protection investment [23]. In companies with a higher degree of digital transformation, relevant laws and regulations are implemented more thoroughly, the information environment is better, the government obtains relevant information faster, and the transparency and efficient transmission of information are conducive to the government’s better supervision of various behaviors of enterprises. Therefore, for companies with a higher degree of digital transformation, the implementation of the low-carbon city pilot policy has a greater impact on corporate environmental and social responsibility information disclosure. Therefore, the following hypotheses are proposed:

Hypothesis 2 (H2).

The impact of the low-carbon city pilot policy on corporate environmental and social responsibility disclosure is moderated by the degree of digital transformation. And firms with a higher level of digital transformation exhibit a stronger policy-induced promotion effect.

Low-carbon city pilot policy has increased the environmental responsibility of enterprises in China to a certain extent in the short term which will promote information disclosure. The implementation of the pilot policy will change the behavioral choices of enterprises. The environmental behavior of enterprises is not only an economic choice based on theoretical analysis but also has a clear green orientation [24]. The environmental performance of enterprises is a signal of enhanced environmental responsibility. Enterprises will assume environmental responsibility due to the signal transmission effect of social responsibility behavior. Due to the obvious information asymmetry in the capital market, listed companies with higher value are more inclined to assume social responsibility and disclose relevant information, sending more positive signals to the capital market, thereby strengthening communication with investors and improving the value of the company [25]. Another more direct reason why the low-carbon city pilot policy affects the these information disclosure of enterprises is the impact of economic resources on environmental responsibility. If enterprises regard environmental responsibility behavior as an investment, then the decline in the availability of economic resources can increase the environmental responsibility of enterprises. When the government determines the pilot area for the low-carbon city pilot policy, the environmental protection quality of the pilot area may attract more attention from the outside media than other non-pilot areas [26]. However, if enterprises regard environmental responsibility as a kind of operating cost, there may not be an obvious causal relationship between resource constraints and environmental responsibility [27]. Therefore, the implementation of the low-carbon city pilot policy may attract more attention from the outside world, with more attention given to low-carbon issues, providing a greater focus on environmental governance, producing good environmental performance, and improving corporate environmental responsibility.

When the government determines the pilot areas for the low-carbon city pilot policy in China, the environmental protection quality of these areas may attract more attention from the outside world, such as the media, compared with other non-pilot areas. Media attention is the source of legitimacy for companies [28], and legitimacy can help companies gain support from stakeholders including the government. The relationship between media attention and corporate environmental protection investment found that media attention can increase corporate environmental protection investment [23]. Therefore, the implementation of the low-carbon city pilot policy may attract more attention from the outside world to the issue of social responsibility information disclosure and thus fulfills their social responsibility for the environment. Therefore, the following hypotheses are proposed:

Hypothesis 3 (H3).

The implementation of the low-carbon city pilot policy can improve corporate environmental performance and media attention, thereby promoting the disclosure of corporate environmental and social responsibility information.

3. Research Design and Data

3.1. Samples and Data

This paper selects A-share listed companies in the Shanghai and Shenzhen Stock Exchanges from 2008 to 2021 as research samples. The paper processes the initial samples in the following order: (1) Eliminate listed companies in the financial insurance sector; (2) Eliminate listed companies that have been treated as ST or *ST; “ST” and “*ST” usually refer to specially treated stocks in the Chinese stock market Special Treatment, which are handled specially by the exchange due to financial abnormalities or operational issues. “ST” indicates that the Shanghai Stock Exchange has implemented special treatment for listed companies with financial abnormalities, such as consecutive losses for two years) or operational issues. “*ST” typically refers to listed companies with negative net profits for two consecutive years or significant operational risks. Next, eliminate samples with missing observations; (3) Eliminate outlier samples, etc. After the above sorting, this study finally obtained 29,206 company-year observations. In order to reduce the impact of extreme values on the research results, the paper also performed Winsorize processing of 1% and 99% for all continuous variables involved in the study.

The data on the cities where the companies are located and the environmental and social responsibility of the companies used in the paper are all from the China Research Data Service (CNRDS). The data on financial and corporate governance at the enterprise level are all from the China Securities Market and Accounting Research (CSMAR) and the China Research Data Service (CNRDS). The data on economic development and industrial structure governance at the city level are from the China Urban Statistical Yearbook and China Statistical Yearbook. The low-carbon city pilots were launched in July 2010, November 2012, and January 2017 in China. Taking into account the implementation time and lag of the pilot policy, as well as the feasibility of this study, the paper sets the starting time of the three batches of low-carbon city pilot policies as 2011, 2013, and 2017. According to Song Hong et al. [29], the paper regards all cities in low-carbon pilot provinces as low-carbon pilot cities. If a city has multiple policy implementation times, it is defined according to the earliest time.

3.2. Variable Specification

Explanatory variables. Corporate environmental and social responsibility disclosure (ECSR), if a social responsibility report is published, it is marked as 1, otherwise it is 0.

Explanatory variables. Low-Carbon City Pilot (PilotPost), whether the low-carbon city pilot policy is implemented in the location of the listed company. If the city where the listed company is located is selected as a pilot city, it is assigned a value of 1 from the pilot year and in the sample period thereafter, otherwise it is 0.

Control variables. According to Sun Gang et al. [30], the study controls the influence of the following variables: (1) City level: environmental regulation intensity (Reg); (2) Enterprise level: company size (Size), debt-to-asset ratio (Lev), net profit margin of total assets (ROA), board size (Board), proportion of independent directors (Indep), Tobin Q value (TobinQ), shareholding ratio of the largest shareholder (Top1), enterprise age (Firmage), nature of ownership (SOE), and degree of equity checks and balances (Banlance).

Moderating variables. For the degree of digital transformation of listed companies (Dig), according to Chen and Srinivasan [31] and He et al. [32], the study constructs a dictionary of digital terms, revolving around seven topics—analytics, automation, artificial intelligence (AI), big data, cloud (computing), digitization, and machine learning—to count digital terms in the firms’ disclosures, and uses the total frequency of digitalization-related words in the annual reports of listed companies to measure it.

3.3. Model Specification

In the document on low-carbon cities issued by the National Development and Reform Commission, the Commission launched three batches of low-carbon city pilots since 2010. The main objectives of the three batches of pilot projects are basically the same, namely, controlling greenhouse gas emissions, exploring green and low-carbon development models, and leading and demonstrating low-carbon development across the country. Since the order in which cities enter the list of low-carbon city pilots is different, according to Baker et al. [33], the paper constructs a multi-period difference-in-differences model (DID), taking the cities where the enterprises are located that are not included in the pilot list as the control group and the cities that have entered the list as the treatment group, and its model is set as follows:

$${ECSR}_{it}={\alpha }_0+{\alpha }_1{Pilot}_i\times {Post}_t+{\alpha }_2{Controls}_t+u_i+{\delta }_t+{\epsilon }_{it}$$

(1)

in which i denotes the city, t denotes the year, ECSR_it is the corporate environmental and social responsibility information disclosure status, and Pilot_i is a dummy variable between groups; in all, these denote whether the city has carried out a low-carbon city pilot if it takes a value of 1 and the rest of the years take a value of 0. Similarly, the dummy variable Post_t takes a value of 1 in the year of the low-carbon city pilot and 0 otherwise. The dummy variable Pilot × Post takes a value of 1 if city i is listed as a low-carbon city pilot in year t, and 0 otherwise. Controls_t denotes a series of control variables, $u_i$ denotes firm fixed effect, ${\delta }_t$ denotes time fixed effect, and ${\epsilon }_{it}$denotes random interference;

Table 1. Variables and definition.

Variables	Variable Definition
ECSR	If a listed company publishes a social responsibility report, it is marked as 1; 0 otherwise
PilotPost	If the region where the listed company is located implements the low-carbon city pilot policy in the year, it is marked as 1; 0 otherwise
Dig	The degree of digital transformation of listed companies
lnpgdp	Logarithm of GDP per capita at the city level
Size	The natural logarithm of the company’s total assets at the end of the period
Lev	The company’s total debt-to-asset ratio at the end of the period
ROA	Net profit/total assets at the end of the period
Board	The natural logarithm of the total number of board members
Indep	The number of independent directors/total number of board members
TobinQ	The ratio of the company’s market value to the company’s total assets at the end of the period
Top1	The ratio of the number of shares held by the largest shareholder to the total share capital
Firmage	The natural logarithm of the number of years the company has been listed plus 1
SOE	When the company’s equity nature is a state-owned enterprise, the value is 1, 0 otherwise
Banlance	The ratio of the second largest shareholder’s shareholding ratio to the largest shareholder’s shareholding ratio
Year	Year dummy variable
Firm	Firm dummy variable

defines the variables.

4. Analysis of Empirical Results

4.1. Descriptive Statistics

Table 2. Descriptive statistics of variables.

Variable	N	Mean	Sd	Min	Max
ECSR	29,206	0.274	0.446	0.0000	1.0000
Pilot × Post	29,206	0.562	0.496	0.0000	1.0000
Dig	29,206	9.390	5.814	−3.0000	22.0000
lnpgdp	29,206	11.55	0.552	−1.763	14.19
Size	29,206	22.14	1.305	19.41	26.45
Lev	29,206	0.418	0.209	0.0270	0.925
ROA	29,206	0.045	0.064	−0.398	0.255
Board	29,206	2.135	0.198	1.609	2.708
Indep	29,206	0.374	0.0530	0.250	0.600
Top1	29,206	0.352	0.150	0.0810	0.758
TobinQ	29,206	2.039	1.376	0.802	17.73
FirmAge	29,206	2.834	0.365	0.693	3.611
SOE	29,206	0.374	0.484	0.0000	1.0000
Balance	29,206	0.724	0.612	0.0160	2.961

reports the descriptive statistics of the main variables. The mean of Pilot × Post is 0.562, indicating that after the implementation of the low-carbon city pilot policy, listed companies in the pilot areas accounted for about 56.2% of the sample.

In addition, the results of calculating the variance inflation factor (VIF) of the main variables showed that the maximum VIF value was 3.91, which is lower than the classic threshold of 10. Therefore, it can be concluded that the model constructed in this paper does not have serious multicollinearity problems.

4.2. Main Results

The paper uses the econometric method of controlling enterprise fixed effects and year fixed effects, and clustering standard errors to the enterprise level to estimate the benchmark model (1). The regression results are shown in

Table 3. Estimated results for ECSR.

Variable	(1)	(2)	(3)
	ECSR	ECSR	ECSR
Pilot × Post	0.0209 ***	0.0215 ***	0.0145
	(0.0079)	(0.0078)	(0.0160)
Dig			−0.0010
			(0.0049)
PilotPost *Dig			0.0085 *
			(0.0049)
lnpgdp		−0.0153	−0.0185
		(0.0142)	(0.0171)
Size		0.0996 ***	0.0971 ***
		(0.0082)	(0.0091)
Lev		−0.1162 ***	−0.1270 ***
		(0.0248)	(0.0253)
ROA		−0.0207	−0.0073
		(0.0388)	(0.0392)
Board		−0.0214	−0.0045
		(0.0236)	(0.0250)
Indep		−0.0562	−0.0312
		(0.0654)	(0.0713)
Top1		−0.0430	−0.0381
		(0.0591)	(0.0630)
TobinQ		0.0107 ***	0.0118 ***
		(0.0022)	(0.0023)
FirmAge		−0.0155	0.0421
		(0.0418)	(0.0470)
SOE		−0.0043	0.0010
		(0.0186)	(0.0191)
Balance		−0.0215 **	−0.0010
		(0.0106)	(0.0049)
Constant	0.3775 ***	−1.3982 ***	−1.7140 ***
	(0.0155)	(0.2668)	(0.3127)
Firm FE/Year FE	Yes	Yes	Yes
Observations	29,206	29,206	23,996
R-squared	0.0465	0.0775	0.0683

Notes: t-values in parentheses, * p < 0.1, ** p < 0.05, *** p < 0.01.

. Among them, column (1) is the estimation result without considering the control variables. The coefficient estimate of the cross-product Pilot × Post is 0.0209, and it passes the significance test at the 1% level. Column (2) is the estimation result when the control variables are considered. The coefficient estimate of the cross-product Pilot × Post is 0.0215, and it passes the significance test at the 1% level, indicating that the implementation of the low-carbon city pilot policy has improved the disclosure of corporate environmental and social responsibility information reports of listed companies; that is, in order to encourage enterprises to actively respond to the increased risks brought about by the low-carbon city pilot policy, the enterprise will have more actions in terms of corporate environmental responsibility. This shows that after the implementation of the low-carbon city pilot policy, the corporate environmental and social responsibility information disclosure in the pilot area will increase, verifying Hypothesis 1. Column (3) is the estimated result of the moderating effect of the degree of enterprise digital transformation. The coefficient estimate of the cross-product Pilot × Post × Dig is 0.0085 and passes the significance test at the 10% level, indicating that enterprise digital transformation has a moderating effect on the low-carbon city pilot policy in promoting the environmental and social responsibility information of listed companies. That is, in order to motivate the company to actively respond to the increased risks brought about by the low-carbon city pilot policy, the company will have more actions in terms of corporate environmental responsibility. This shows that after the implementation of the low-carbon city pilot policy, the corporate environmental and social responsibility information disclosure of companies in the pilot area will increase, verifying Hypothesis 2.

4.3. Robust Test

4.3.1. Parallel Trend Test

The difference-in-differences (DID) method can better solve the endogeneity problem in policy evaluation, but it is based on a series of important assumptions. Among them, the parallel trend assumption is the most important premise for using the DID method, which requires that in the absence of external policy influences, the outcome variables of the experimental group and the control group should develop in the same trend, and the two groups of samples should be comparable. Therefore, in accordance with Baker et al. [33] and Song Hong et al. [26], the study uses the event study method to study the dynamic effects of the low-carbon city pilot policy on enterprises and tests the common trend of the difference-in-differences method; its model is set as follows:

$${ECSR}_{i,t}=a_0+{\sum }_{k=-2}^8{\beta }_k\times {Politpost}_{i,t+k}+a_1{Controls}_{i,t}+r_t+u_i+{\epsilon }_{i,t}$$

(2)

in which the ${PilotPost}_{i,t+k}$ is a dummy variable before and after the low-carbon city pilot, k denotes the date, takes k years before the low-carbon city pilot, and a positive number indicates k years after the low-carbon city pilot. Since the period after the low-carbon city pilot in the sample is relatively long, the paper selects the 2 years before implementation and the 7 years after implementation. The coefficient estimate of ${\beta }_k$ is shown in Figure 1. It can be seen that before the implementation of the low-carbon city pilot policy, the coefficient of ${\beta }_k$ was not significantly different from 0, indicating that the parallel trend hypothesis was met. In the third year after the implementation of the low-carbon pilot policy, the coefficient of ${\beta }_k$ was significantly greater than 0 and had an increasing trend, and was significant in most years after the policy began. This may be because in the early stages of the implementation of the low-carbon pilot policy, enterprises were not sensitive to the increased risks brought about by the low-carbon pilot policy. With the progressive implementation of the policy and stricter law enforcement, firms face greater low-carbon risks, which in turn motivate them to assume social responsibilities and disclose related information. With the lagged effect, the low-carbon city pilot policy impact on corporate environmental and social responsibility information disclosure becomes significant only three years after implementation, as observed in the empirical results. We think that it can be explained by both institutional and corporate factors. On the institutional side, the transmission of the low-carbon city pilot policy from central design to local enforcement requires a period of adaptation in China. The government needs time to establish regulatory frameworks, detailed guidelines, and monitoring mechanisms of the low-carbon city pilot policy. On the corporate side, firms typically operate under annual decision-making and reporting cycles, and adjustments to corporate environmental and social responsibility information disclosure with some works. This includes strategic deliberation, cross-departmental coordination, and the establishment of data collection and reporting systems. Moreover, some firms often adopt a waiting strategy in the early stage, responding only after policy signals become clear and peer practices accumulate. Therefore, these institutional delays and corporate adjustment processes contribute to the lagged manifestation of the policy effect in corporate environmental and social responsibility information disclosure.

4.3.2. Placebo Test

A placebo test is performed by replacing the treatment group in order to avoid the influence of unobservable omitted variables on the benchmark regression results. The paper randomly selects 121 cities from the sample cities as false treatment group cities, and the remaining cities as false control group cities, so as to obtain the coefficient estimate of the impact of the low-carbon city pilot policy of implementing city placebo on corporate environmental responsibility. The above process is repeated 500 times to avoid the interference of other small probability factors on the estimation results. Figure 2 is a kernel density distribution of the estimated coefficients of the randomly generated treatment group. It is found that the estimated coefficients have a mean of zero and are normally distributed. The vast majority of regression results are not significant, indicating that the estimation results of the paper have a high degree of credibility.

4.4. Robustness Test

4.4.1. Sample Data Screening

To avoid the impact of extreme values on the benchmark regression results, the research sample was truncated by 1% and 5% according to the variable (ECSR), and then the model (1) was regressed. The estimation results show that after removing the extreme values, the coefficient estimates of Pilot × Post all passed the significance test at the 1% level. Columns (1) and (2) in

Table 4. Robustness test results.

Variables	1% Reduction in Tail	5% Reduction in Tail	Other Policies	psm1:3	psm1:5
	(1) ECSR	(2) ECSR	(3) ECSR	(4) ECSR	(5) ECSR
Pilot × Post	0.0215 ***	0.2532 ***	0.0215 ***	0.0232 ***	0.0201 **
	(0.0078)	(0.0811)	(0.0078)	(0.0086)	(0.0082)
Newpost			−0.1386 ***
			(0.0404)
lnpgdp	−0.0153	0.0215 ***	−0.0153	−0.0103	−0.0051
	(0.0142)	(0.0078)	(0.0142)	(0.0171)	(0.0163)
Size	0.0996 ***	−0.0153	0.0996 ***	0.1014 ***	0.1028 ***
	(0.0082)	(0.0142)	(0.0082)	(0.0100)	(0.0096)
Lev	−0.1162 ***	0.0996 ***	−0.1162 ***	−0.1089 ***	−0.1022 ***
	(0.0248)	(0.0082)	(0.0248)	(0.0301)	(0.0283)
ROA	−0.0207	−0.1162 ***	−0.0207	−0.0487	−0.0248
	(0.0388)	(0.0248)	(0.0388)	(0.0508)	(0.0489)
Board	−0.0214	−0.0207	−0.0214	−0.0109	−0.0099
	(0.0236)	(0.0388)	(0.0236)	(0.0281)	(0.0275)
Indep	−0.0562	−0.0214	−0.0562	−0.0674	−0.0474
	(0.0654)	(0.0236)	(0.0654)	(0.0793)	(0.0775)
Top1	−0.0430	−0.0562	−0.0430	−0.0390	−0.0631
	(0.0591)	(0.0654)	(0.0591)	(0.0697)	(0.0654)
TobinQ	0.0107 ***	−0.0430	0.0107 ***	0.0069 **	0.0074 **
	(0.0022)	(0.0591)	(0.0022)	(0.0029)	(0.0029)
FirmAge	−0.0155	0.0107 ***	−0.0155	−0.0309	−0.0208
	(0.0418)	(0.0022)	(0.0418)	(0.0485)	(0.0460)
SOE	−0.0043	−0.0155	−0.0043	−0.0028	0.0006
	(0.0186)	(0.0418)	(0.0186)	(0.0220)	(0.0213)
Balance	−0.0215 **	−0.0043	−0.0215 **	−0.0240 *	−0.0276 **
	(0.0106)	(0.0186)	(0.0106)	(0.0129)	(0.0123)
Constant	−1.4002 ***	−1.4002 ***	−1.3982 ***	−1.4422 ***	−1.5620 ***
	(0.2666)	(0.2666)	(0.2668)	(0.3186)	(0.3037)
Firm FE/Year FE	Yes	Yes	Yes	Yes	Yes
Observations	29,206	29,206	28,816	19,617	21,029
R-squared	0.0775	0.0775	0.0485	0.0465	0.0426

Notes: t-values in parentheses, * p < 0.1, ** p < 0.05, *** p < 0.01.

show the regression results, and the results still support Hypothesis 1.

4.4.2. Eliminate Other Policy Interference

To avoid other policies that may affect corporate employment during the sample period and cause bias in the benchmark estimation results, this study collected and sorted out documents and found that during the sample period of this article, there may be a policy that may affect corporate environmental responsibility during the sample period, namely the new “Environmental Protection Law of the People’s Republic of China” that came into effect on 1 January 2015. In order to avoid interference from other pilot policies during the implementation of the low-carbon city pilot policy, according to Tang G. P. et al. [34], the paper constructs a dummy variable (Newpost) to represent the implementation of the new environmental protection law. The dummy variable of the new environmental protection law is added to the baseline regression. The regression results are shown in column (3) of Table 4. It can be seen that after controlling for interfering policies, the results still support Hypothesis 1.

4.4.3. Propensity Score Matching (PSM)

In order to alleviate the interference of sample selection bias on the conclusion, the paper adopts a year-by-year matching method to match the policy experimental group enterprises in each year with the control group enterprises. The enterprise characteristic variables (control variables in model (1)) in the year before the pilot policy time node are used as covariates, and whether it is the experimental group affected by the policy is used as the dependent variable. The nearest neighbor matching of “one-to-many, with replacement” is performed to find the control group that is most comparable to the experimental group in the year before the pilot policy time node is adopted in the paper. In addition, in order to ensure the effect of PSM, this study conducted a balance test and a common support test for each PSM. The results showed that there was no significant difference in the covariates after PSM, while the difference in the corporate environmental responsibility variables was still significant. The overlap between the experimental group and the control group after matching was higher, which met the common support hypothesis. Columns (4) and (5) in Table 4 lists the regression results of PSM-DID after 1:3 and 1:5 matching, among which the regression coefficients of Pilot × Post were significantly positive at the 1% and 5% level, respectively. The results still support Hypothesis 1.

5. Results of Further Analysis

5.1. Mechanism Analysis

5.1.1. Environment Performance

The theoretical analysis above shows that the implementation of the low-carbon city pilot policy can enhance corporate environmental performance and thus improve environmental and social responsibility information disclosure; that is, environmental performance plays a mediating role between policy implementation and environmental and social responsibility information disclosure. The implementation of the low-carbon city pilot policy has improved corporate environmental performance to a certain extent [13,14,15].

In order to verify the mediating mechanism of corporate environmental performance in the impact of the implementation of the low-carbon city pilot policy on corporate environmental and social responsibility information disclosure, its models are set as follows:

$$E_{it}/{Media}_{it}={\alpha }_0+{\alpha }_1{Pilot}_i\times {Post}_t+{\alpha }_2{Controls}_t+u_i+{\delta }_t+{\epsilon }_{it}$$

(3)

$$E{CSR}_{it}={\alpha }_0+{\alpha }_1{E}_{it}/{Media}_{it}+{\alpha }_2{Pilot}_i\times {Post}_t+{\alpha }_3{Controls}_t+u_i+{\delta }_t+{\epsilon }_{it}$$

(4)

in which i denotes the firm, t denotes the year, and $E_{it}$ denotes the environmental performance of firm i in year t. According to Khurram et al. [35], the data of ESG scores are from Huazheng International Consulting Services, and $E_{it}$ is the E score of ESG score. The definitions of other variables are the same as those in model (1). According to Wen Zhonglin [36], the study tests the mediating effect of the low-carbon city pilot policy on the disclosure of corporate environmental and social responsibility information.

The results of environmental performance playing a mediating role in the impact of the implementation of the low-carbon city pilot policy on the disclosure of corporate environmental and social responsibility information are shown in

Table 5. Mechanism test results.

Variables	(1)	(2)	(3)	(4)
	E	ECSR	Media	ECSR
E		0.0718 ***
		(0.0036)
Media				0.0000 **
				(0.0000)
Pilot × Post	0.0530 **	0.0177 **	48.6211 ***	0.0210 ***
	(0.0269)	(0.0072)	(15.8198)	(0.0078)
lnpgdp	0.0420	−0.0203 *	−13.4883	−0.0153
	(0.0551)	(0.0123)	(15.2874)	(0.0143)
Size	0.1794 ***	0.0878 ***	67.1765 ***	0.0984 ***
	(0.0220)	(0.0076)	(16.4493)	(0.0083)
Lev	−0.5568 ***	−0.0771 ***	50.2947	−0.1169 ***
	(0.0814)	(0.0226)	(56.8584)	(0.0249)
ROA	0.7344 ***	−0.0752 **	48.0675	−0.0196
	(0.1479)	(0.0362)	(55.4165)	(0.0390)
Board	−0.0885	−0.0144	−25.7034	−0.0209
	(0.0827)	(0.0221)	(44.5726)	(0.0237)
Indep	−0.4387 *	−0.0189	134.7706	−0.0576
	(0.2326)	(0.0628)	(137.8074)	(0.0657)
Top1	0.3411 **	−0.0679	32.2960	−0.0464
	(0.1656)	(0.0544)	(145.4460)	(0.0592)
TobinQ	−0.0001	0.0111 ***	25.0555 ***	0.0098 ***
	(0.0069)	(0.0021)	(3.5619)	(0.0023)
FirmAge	−0.1056	−0.0006	−91.1909 *	−0.0150
	(0.1285)	(0.0398)	(51.9086)	(0.0418)
SOE	0.0019	−0.0030	−5.7286	−0.0042
	(0.0557)	(0.0173)	(34.0529)	(0.0186)
Balance	−0.0091	0.0718 ***	41.3094	−0.0222 **
	(0.0319)	(0.0036)	(52.5212)	(0.0106)
Constant	2.5664 ***	−1.8119 ***	−1285.7468 ***	−1.3667 ***
	(0.8114)	(0.2477)	(292.0556)	(0.2674)
Firm FE/Year FE	Yes	Yes	Yes	Yes
Observations	28,068	28,068	28,386	28,386
R-squared	0.0483	0.1391	0.0640	0.0780

Notes: t-values in parentheses, * p < 0.1, ** p < 0.05, *** p < 0.01.

. From the column (1) of Table 5, it can be seen that the coefficient of corporate environmental performance is positive and significant at the 5% significance level, indicating that pilot policy implementation can indeed significantly promote the environmental performance of corporate. The column (2) of Table 5 shows the regression results of environmental performance, pilot policy implementation and corporate environmental and social responsibility information disclosure, where the regression coefficient of corporate environmental performance is positive and significant at the 5% significance level, indicating that the pilot policy implementation can indeed significantly promote corporate environmental and social responsibility information disclosure by improving environmental performance, verifying Hypothesis 3.

5.1.2. Media Attention

The theoretical analysis above shows that the implementation of the low-carbon city pilot policy can increase the media attention of enterprises, thereby improving the disclosure of corporate environmental and social responsibility information; that is, media attention plays a mediating role between policy implementation and the disclosure of corporate environmental and social responsibility information. When the government determines the pilot area for the low-carbon city pilot policy, enterprises in the pilot area will receive more attention from the outside world, which may prompt the holding companies to pay more attention to low-carbon issues, strengthen corporate environmental responsibility, and thus improve environmental and social responsibility information disclosure.

As an important information carrier in the securities market, the media can also reflect the degree of attention of investors to a certain stock to a certain extent. Media reports have obvious emotional characteristics; the optimistic words used in the reports can attract investors’ attention, incite their investment sentiment, and thus affect asset prices [37]. To verify the mediating role of media attention, according to Xu et al. [38], the study determines the number of reports by the number of times the company name appears in newspapers and online financial news titles. Based on the previous analysis, the paper uses Media to measure the media attention of companies. The variable ${Media}_{it}$ denotes the media attention of company i in year t. The paper defines the variable Media as the natural logarithm of one plus the total number. The data comes from the CNRDS database. The variables are brought into model (3) and model (4) for regression.

The regression results are shown in Table 5. Among them, the coefficient of column (3) is significantly positive, which indicates that the implementation of the low-carbon city pilot policy can indeed significantly increase the media attentions of companies, while the coefficient of column (4) is significantly positive, and the results are significant at the 1% and 5% significance level, respectively, which indicates that the implementation of the low-carbon city pilot policy promotes the disclosure of environmental and social responsibility information through media attention, verifying Hypothesis 3.

5.2. Heterogeneity Analysis

5.2.1. Green Investors

Institutional investors tend to invest in companies with good credit and a high sense of social responsibility, as well as in the green bond market [39], which helps to enhance corporate environmental governance enthusiasm and social responsibility, curb corporate tax evasion, and thus enhance corporate value. As the government actively advocates green and sustainable development, institutional investors have gradually avoided or reduced investment in polluting companies, “forcing” polluting companies to transform towards a cleaner direction [40]. As a type of institutional investors, green investors can promote companies to increase green spending, implement green actions, and improve green governance performance. Both green actions and green governance performance help improve corporate operating performance [41]. While green investors focus on environmental responsibility goals, they will also encourage companies to pay attention to environmental and social responsibilities.

According to Jiang et al. [41] and Chi et al. [42], the paper selected the original data of green investors comes from the Guotai An Database (CSMAR). It is matched with the “Fund Entity Information Table” and “Stock Investment Details Table” in the listed company fund market series to obtain the fund details table of investment in listed companies. Then, the “investment objectives” and “investment scope” of each fund are manually queried to see whether they contain words related to the environment, such as “environmental protection”, “ecology”, “green”, “new energy development”, “clean energy”, “low carbon”, “sustainable”, “energy saving”, etc. If they exist, the company is considered to have green investors; otherwise, they do not. On this basis, the number of green investors included in the company in that year is counted; companies with green investors are defined as the (GI = 1) group, and companies without green investors are defined as the (GI = 0) group. The relevant test results are shown in columns (1) and (2) of

Table 6. Heterogeneity and impact on the quality of information disclosure test results.

Variables	(1)	(2)	(3)	(4)	(5)
	ECSR (GI = 1)	ECSR (GI = 0)	ECSR (Indu = 1)	ECSR (Indu = 0)	Equal
ECSR					0.9862 ***
					(0.0317)
Pilot × Post	0.0119	0.0273 ***	0.0123	0.0249 *	−0.0515 **
	(0.0137)	(0.0089)	(0.0088)	(0.0148)	(0.0229)
lnpgdp	−0.0098	−0.0147	−0.0173	0.0011	−0.0891 **
	(0.0260)	(0.0171)	(0.0159)	(0.0266)	(0.0346)
Size	0.1343 ***	0.0727 ***	0.0954 ***	0.0935 ***	0.0721 ***
	(0.0136)	(0.0094)	(0.0104)	(0.0138)	(0.0177)
Lev	−0.1024 **	−0.1115 ***	−0.1534 ***	−0.0679	0.0531
	(0.0446)	(0.0289)	(0.0273)	(0.0466)	(0.0655)
ROA	−0.1249	−0.0136	−0.0311	−0.0789	0.0303
	(0.0814)	(0.0366)	(0.0428)	(0.0728)	(0.1001)
Board	−0.0065	−0.0405	−0.0152	−0.0208	−0.0347
	(0.0373)	(0.0291)	(0.0257)	(0.0475)	(0.0648)
Indep	−0.1026	−0.0431	−0.0508	−0.0786	−0.0780
	(0.1081)	(0.0850)	(0.0686)	(0.1210)	(0.1811)
Top1	−0.0476	−0.0108	−0.0549	−0.0700	−0.2244 *
	(0.0959)	(0.0673)	(0.0717)	(0.0969)	(0.1318)
TobinQ	0.0158 ***	0.0025	0.0132 ***	0.0033	−0.0058
	(0.0037)	(0.0024)	(0.0025)	(0.0042)	(0.0055)
FirmAge	0.0276	−0.0770	−0.0027	0.0346	0.2731 ***
	(0.0672)	(0.0484)	(0.0474)	(0.0897)	(0.0909)
SOE	−0.0003	0.0038	−0.0060	−0.0204	0.0004
	(0.0391)	(0.0190)	(0.0237)	(0.0293)	(0.0433)
Balance	−0.0283 *	−0.0161	−0.0265 **	−0.0054	−0.0550 **
	(0.0164)	(0.0127)	(0.0113)	(0.0227)	(0.0244)
Constant	−2.3244 ***	−0.6712 **	−1.3283 ***	−1.5599 ***	0.0893
	(0.4373)	(0.3135)	(0.3157)	(0.4939)	(0.5823)
Firm FE/Year FE	Yes	Yes	Yes	Yes	Yes
Observations	11,834	15,704	20,249	8496	29,206
R-squared	0.0995	0.0634	0.0427	0.0425	0.6829

Notes: t-values in parentheses, * p < 0.1, ** p < 0.05, *** p < 0.01.

. Among them, in the group with green investors, the coefficient of Pilot × Post is significantly positive at the 1% level, while the coefficient in the group without green investors is not significant, indicating that after the implementation of the low-carbon city pilot policy, enterprises with green investors in the pilot area are more likely to disclose corporate environmental and social responsibility information.

5.2.2. Carbon Emissions

Taking into account the impact of carbon emissions in different industries, the paper divides the sample into a group with higher carbon emissions and a group with lower carbon emissions. According to Hu et al. [43], if a listed company belongs to the eight major industries, such as petrochemicals, chemicals, building materials, steel, nonferrous metals, papermaking, electricity and aviation, it belongs to the group with higher carbon emissions, and Indu takes a value of 1. Others belong to the group with lower carbon emissions, and Indu takes a value of 0. The relevant test results are shown in columns (3) and (4) in Table 6. Among them, the Pilot × Post coefficient in the group with lower carbon emissions is significantly positive at the 10% level, but not significant in the group with higher carbon emissions. This indicates that after the implementation of the low-carbon city pilot policy, enterprises in the pilot area belonging to industries with lower carbon emissions are more likely to disclose corporate environmental and social responsibility information than enterprises in industries with higher carbon emissions.

5.3. Impact on the Quality of Information Disclosure

In order to verify whether the impact of the low-carbon city pilot policy on corporate environmental and social responsibility information disclosure will have a significant impact on the quality of environmental information disclosure, its models are set as follows:

$$E{qual}_{it}={\alpha }_0+{\alpha }_1{ECSR}_{it}+{\alpha }_2{Pilot}_i\times {Post}_t+{\alpha }_3{Controls}_t+u_i+{\delta }_t+{\epsilon }_{it}$$

(5)

in which i denotes the firm, t denotes the year, and ${Equal}_{it}$ denotes the Information quality of environmental information disclosure of firm i in year t. According to Kong et al. [44], this paper assigns values to corporate disclosure information. If environmental management disclosure, certification disclosure, and environmental information disclosure carrier items are disclosed, 2 points are given; otherwise, 0 points are given. If environmental liability disclosure, environmental performance and governance disclosure monetizable items have quantitative and qualitative descriptions in environmental information disclosure, 2 points are given, if they only have qualitative descriptions, 1 point is given, and if they are not disclosed, 0 points are given. The maximum score is no more than 5 points, and the minimum score is 0 points. The total score is added by 1 and the natural logarithm is taken to obtain the corporate environmental information disclosure quality index. The definitions of other variables are the same as those in model (1).

The results are listed in column (5) of Table 6. It can be seen that the coefficient of corporate environmental and social responsibility information disclosure is positive and significant at the 1% significance level, indicating that the positive impact of the implementation of the low-carbon city pilot policy on corporate environmental and social responsibility information disclosure has a positive impact on the quality of disclosure.

6. Discussion

This study integrates theoretical analysis with empirical evidence to examine the impact of the low-carbon city pilot policy on corporate environmental and social responsibility information disclosure. Using a fixed effects difference-in-differences model with Chinese A-share listed firms, the analysis yields four key findings. First, the low-carbon city pilot policy significantly enhances corporate environmental and social responsibility information disclosure among firms in pilot regions, highlighting the effectiveness of place-based environmental regulation in promoting corporate transparency. Second, corporate digital transformation moderates this positive impact, with the mechanism operating primarily through improved corporate environmental performance and increased media attention. Third, the policy’s effect is heterogeneous, and it is more pronounced for firms with green investors and those operating in low-carbon-emission industries. Fourth, the study finds that the low-carbon city pilot policy has a positive impact on the quality of corporate environmental information disclosure.

This study provides novel evidence on the governance effects of the low-carbon city pilot policy and underscores the importance of institutional and technological factors in advancing corporate environmental and social responsibility. This study contributes to the literature on environmental regulation and corporate disclosure, while offering practical insights for achieving the goals of the Carbon Peaking and Carbon Neutrality.

Based on the above conclusions, the paper obtains the following policy implications: First, the low-carbon city pilot policy can strengthen the disclosure of corporate environmental and social responsibility information in the pilot area. Although the pilot policy is in the process of continuous advancement, there are still many cities that have not joined the construction of low-carbon cities. Policymakers can try to promote the low-carbon city pilot nationwide, and provide reference for the implementation of China’s “carbon peak in 2030, carbon neutrality in 2060” climate action from the city level. Policymakers can expand the pilot scope of the low-carbon city pilot policy. In addition, the government departments should strengthen the construction of the institutional environment. The study demonstrates that the institutional environment is a critical mechanism through which the low-carbon city pilot policy enhances corporate environmental and social responsibility disclosure. Therefore, alongside implementing targeted policies, policymakers must strengthen the institutional framework itself. This entails enacting more comprehensive legislation and enforcing stricter laws to enhance regulatory oversight and deterrence, thereby compelling enterprises to prioritize environmental and social responsibility performance. To sum up, in the process of promoting the low-carbon city pilot policy, enterprises should give full attention to the subjective initiative of environmental governance subjects when dealing with the risk factors brought about by environmental regulations. The industry attributes of listed companies and the attitudes of institutional investors may bring greater risks and uncertainties to enterprises. Therefore, enterprises should adhere to the concept of low-carbon development in the future, actively assume environmental responsibilities, and take appropriate corporate governance measures to improve their ability to resist risks when facing external environmental risk shocks.

This study proposes environmental performance as an important pathway for the policy’s influence, using the E score in the ESG rating to test its mediating role. However, the ESG ratings is are comprehensive indicators, which may introduce measurement bias. Future research would benefit from employing more nuanced variables capture corporate environmental performance. Moreover, the study relied on the manual screening of keywords in fund prospectuses to identify green investors in heterogeneity analysis. While intuitive, this method has limitations. If a comprehensive and authoritative green fund database becomes available, future studies could cross-validate our findings against it to enhance the robustness of the identification. We contend that these potential limitations do not undermine the main conclusions of this study but rather indicate productive avenues for future refinement.