Does ESG Information Disclosure Improve Green Innovation in Manufacturing Enterprises?

Abstract

Based on the data of Chinese A-share manufacturing firms from 2015 to 2021, this paper employs the multiple regression method to explore the relationship between ESG information disclosure and corporate green innovation, while also examining the impact of local government environmental attention on this relationship. The results indicate that: (1) ESG information disclosure significantly promotes corporate green innovation, particularly in the eastern and central regions of China, as well as areas with a strong Confucian cultural background; (2) Alleviating financing constraints and reducing agency costs are important channels through which ESG information disclosure influences corporate green innovation; and (3) Local government environmental attention exhibits an “inverted U-shaped” regulating effect on the relationship between ESG disclosure and corporate green innovation.

1. Introduction

Environmental protection and innovation are integral components of the 2030 Sustainable Development Goals (SDGs) and represent significant challenges for countries worldwide. The “Sustainable Development Goals Report 2022” indicates that environmental degradation and the climate crisis pose obstacles to achieving economic and social sustainability goals, shifting the focus from economic growth rates to health and resilience. To address this, the Chinese government has implemented policies like the “Regulations on Disclosing Environmental Information by Enterprises” and the “Abolition of Regulations Pertaining to the Import of Solid Waste”, signaling an escalating focus on environmental issues and tighter controls on emissions from enterprises [1]. In this context, companies will bear higher costs for environmental pollutant emissions, compelling them to allocate limited funds towards green innovation to address increasing government regulatory costs and enhance their green innovation capabilities.

ESG, a recent concept in the field of sustainable development, encompasses three main components: environmental factors, social considerations, and corporate governance practices. Currently, investigations into ESG information disclosure predominantly concentrateon its connection with capital market responses or corporate financial outcomes, and research on the influence of ESG on green innovation in the manufacturing industry is scarce. Current literature on the subject only investigates the link between ESG and green innovation from a unidirectional standpoint [2,3,4] and fails to delve into the moderating factors that influence their relationship [5]. Earlier research has indicated that green innovation in corporations is affected by institutional elements [6]. Governmental oversight and environmental focus can provide enterprises with business prospects and resources, enhancing their green R&D capabilities; however, overly stringent environmental regulations can adversely affect their green innovation efforts [7]. Nevertheless, researchers frequently neglect the crucial role of the government when investigating the link between ESG disclosure and green innovation in corporations. In the context of China, the government is a vital component in fostering the culture of corporate innovation. Existing studies have highlighted the positive influence of government focus on the environment in stimulating green innovation among enterprises [8]. Nonetheless, there is an urgent need to expand research on how government environmental focus affects the green innovation of manufacturing firms and to determine the circumstances under which this effect is more significant.

As a result, our study employs data from manufacturing companies listed between 2015 and 2021. Drawing on signaling and stakeholder theories, we pair company registration details with factors like environmental focus by prefecture-level municipal governments. The aim is to empirically assess how ESG influences green innovation in corporations and examine the moderating influence of governmental environmental attention. Our findings indicate that the revelation of ESG information fosters green innovation in manufacturing companies. Moreover, the extent of government emphasis on the environment exhibits an “inverted U-shaped” influence on the connection between ESG and green innovation. A comprehensive analysis reveals that ESG boosts green innovation by easing financial constraints and lowering agency costs. In contrast to earlier research, the distinct contribution of this study primarily rests on the following aspects: first, it enhances the comprehension of the direct elements that impact green innovation in enterprises. The research points out that diverse direct factors influence corporate green innovation [9]; however, there is a paucity of research on the association between ESG and green innovation in the manufacturing industry [4]. This study provides novel perspectives on the factors affecting green innovation in manufacturing firms from an ESG standpoint. Second, our study sheds new light on the interplay and moderating impacts between ESG disclosure and green innovation in corporations and enhances the body of research in this area. Lastly, valuable data were extracted from local government reports through text mining techniques, and a government environmental focus index was developed to empirically assess its moderation on the relationship between ESG and corporate green innovation. A link between macro-level markets and micro-level firms was established, and our findings offer crucial decision-making guidance and theoretical grounding for China’s pursuit of sustainable development objectives and the formulation of green innovation strategies.

2. Literature Review and Research Hypotheses

2.1. ESG and Corporate Green Innovation

With the growing popularity of sustainable development concepts, research on corporate ESG reporting has become a notable focus of academic attention. Presently, researchers are primarily exploring the factors that influence and the outcomes of ESG disclosure. Regarding the influencing factors, elements such as the senior management team [10], knowledge networks in green supply chains [11], Confucian principles [12], and digital transformation plans [13] can boost the extent of ESG in companies. The diverse skills of the senior management team enhance a company’s capacity to analyze information and make decisions, reinforce corporate governance practices, and promote the disclosure of ESG-related information [14]. In terms of the consequences of its impact, the majority of studies suggest that ESG information disclosure positively affects companies [15,16,17,18,19,20]. Firms with ESG information disclosure can improve investment effectiveness [21], raise environmental benchmarks [22], and stimulate green innovation by reducing debt levels [23] and financing costs [6,24,25]. For instance, research conducted by Li and Rasiah (2024) revealed that disclosing ESG fosters green innovation in corporations, a viewpoint echoed by Yuan et al. (2024) [6]. However, existing research mainly concentrates on the effect of ESG on green innovation within the entire industry, neglecting the distinct features of manufacturing companies. Additionally, as environmental regulations imposed by the government become stricter, the government’s focus on the environment is expected to affect the relationship between the two.

Green innovation involves incorporating environmental consciousness and accountability into a firm’s manufacturing and business processes. It involves improving existing product designs, processes, organizational management, and developing new products to achieve sustainable environmental and economic benefits [26]. According to stakeholder theory, businesses are not only driven by the maximization of shareholder interests but also rely on active engagement with stakeholders such as employees, consumers, and the government [27]. As environmental consciousness and concern for green practices rise, stakeholders increasingly focus on a company’s environmental actions. They demand that companies prioritize environmental protection in their daily operations to reduce pollution emissions, and green innovation is a proactive response to this issue. Furthermore, proactive ESG information disclosure by companies can reduce the costs for stakeholders to obtain information. It assists stakeholders in overseeing and constraining corporate management from a governance perspective. Consequently, this leads to a decrease in agency costs and influences green innovation within corporations. Disclosing ESG information aids companies in establishing a favorable social responsibility reputation, capturing the interest of stakeholders, and offering assistance to firms pursuing green innovation endeavors. According to signaling theory, ESG disclosure serves as a vital signaling mechanism that diminishes principal-agent issues stemming from information asymmetry [28]. It conveys positive messages about the company’s image to the market, eases stakeholders’ worries [29], earns the trust of investors, and ultimately lessens corporate financing limitations, thereby promoting green growth [30]. Simultaneously, as a sustainable development evaluation system [31], ESG contributes to optimizing market investment structures [32], guiding social capital towards green, low-carbon, and ecological sectors, thereby promoting corporate green innovation activities. Given this background, the study proposes the following hypothesis.

H1: 

Disclosing ESG promotes green innovation in manufacturing companies.

2.2. Government Environmental Attention and Corporate Green Innovation

In China, the government is an important part of maintaining the spirit of corporate innovation. The structure and behavior of enterprise organizations are closely related to the institutional environment of the government [32]. Adeptly adapting to shifts in the institutional environment enables enterprises to acquire external resources necessary for their survival and growth. The characteristics of the institutional framework can shape the strategic choices of businesses [33]. Studies have shown that environmental or social issues are better handled by the government than by businesses, given the government’s proficiency in addressing such matters [34]. The influence of ESG on corporate green innovation may vary based on government policies, regulations, and official statements [35]. Government policies may set the benchmark for corporate development [36] and shape strategic decisions related to green innovation. Considering the government environment attention and sustainable development goals, companies may lower their financing costs by obtaining political resources in order to comply with the government’s “intervention hand” and thus have an impact on their green innovation. This study includes the government’s emphasis on environmental issues as a moderating variable to investigate the relationship between ESG reporting and green innovation in manufacturing companies.

When the government’s attention on environmental issues is limited, the positive effect of ESG reporting on green innovation in manufacturing companies becomes more evident. At first, business leaders give precedence to how ESG disclosure affects their reputation and brand identity, leading them to favor a long-term approach to green innovation. By developing new products, equipment, etc., they can stimulate green innovation behavior in enterprises, thereby reducing environmental protection costs and improving production efficiency and competitiveness [37]. With the expansion of government power, there is an increased emphasis on environmental concerns, resulting in higher investments in green innovation projects. Consequently, this amplifies the effect of ESG in promoting green innovation within corporations [5,35]. Furthermore, to meet market demands, gain consumer confidence, and boost brand reputation, companies will become more aggressive in highlighting their green innovation accomplishments through ESG disclosure. This voluntary disclosure of information not only helps companies attract green investors and consumers, but may also promote differentiation advantages in competition, further driving companies to increase investment in green innovation and form a virtuous cycle.

Nevertheless, as government focus on the environment intensifies, businesses will encounter mounting pressure from rising environmental expenses. As a result, the positive impact of ESG on green innovation in manufacturing companies may decrease. This occurs because, given constraints like technological capabilities and resource allocation within firms, heightened government environmental scrutiny can elevate operational costs, hindering R&D innovation [7]. Alternatively, stringent environmental regulations imposed by the government may compel firms to allocate more effort and funds towards pollution and emission management, instead of directly funding R&D for green products and services. Shifting to a compliance-focused strategy could divert resources away from green technology innovation and R&D, resulting in a deceleration of green innovation progress within firms. As a result, the efficacy of ESG reporting in fostering green innovation may be reduced. Given these assumptions, this research proposes the following hypothesis.

H2: 

Government’s attention on environment has an “inverted U-shaped” regulating effect on the relationship between ESG information disclosure and corporate green innovation.

3. Research Design

3.1. Model Design

To assess the influence of ESG on green innovation within manufacturing firms, this research develops the following baseline model:

$${Patent}_{i,t}={\alpha }_0+{\alpha }_1{ESG}_{i,t}+{\alpha }_2{Controls}_{i,t}+{Ind}_i+{Year}_t+Province+{\epsilon }_{i,t}$$

(1)

where ${Patent}_{i,t}$ represents the green innovation of enterprise $i$ in year $t$; ${ESG}_{i,t}$ represents the ESG information disclosure of enterprise $i$ in year $t$; ${Controls}_{i,t}$ represents a series of control variables, including firm-level variables such as company size ($Size$), leverage ($Lev$), profitability ($ROA$), ownership concentration ($Herf$), cash holdings ($Cash$), asset structure ($Asset$), and the proportion of independent directors ($Independent$), as well as macro-level control variables such as government regulatory capacity ($Tech$) and human capital level ($Human$); and ${\epsilon }_{i,t}$ represents the error term. This study controls for industry fixed effects ${Ind}_i$, time fixed effects ${Year}_t$, and regional fixed effects $Province$ to alleviate to some extent the endogeneity problems arising from omitted variables, etc.

Based on this foundation, the study further explores the “inverted U-shaped” moderating effect of government’s environmental focus on the link between ESG and green innovation in manufacturing firms. Referring to the work of Haans et al. (2016) and Luo et al. (2022), the subsequent model is established [38,39]:

$$\begin{gathered} {Patent}_{i,t}={\beta }_0+{\beta }_1{ESG}_{i,t}+{\beta }_2{Concern}_{i,t}+{\beta }_3{Concern}_{i,t}^2 \\ +{{\beta }_4{ESG}_{i,t} \ast Concern}_{i,t}+{\beta }_5{ESG}_{i,t} \ast {Concern}_{i,t}^2 \\ +{\beta }_6{Controls}_{i,t}+{Ind}_i+{Year}_t+Province+{\epsilon }_{i,t} \end{gathered}$$

(2)

In model (2), when the positive impact of ESG on Patent is significant, ESG*Concern is significant and the sign is positive, and ESG*Concern2 is significant and the sign is negative, government environmental attention has a significant “inverted U-shaped” regulatory effect on the relationship between ESG information disclosure and corporate green innovation.

3.2. Variable Measures

(1)

Explained variables

Corporate green innovation ($Patent$). This study adopts the methodology utilized by Jin et al. (2022), utilizing the natural logarithm of the count of green invention patent applications, incremented by one, as a measure of green innovation in companies [40].

(2)

Explanatory variables

ESG information disclosure (ESG). Starting in 2015, SynTao Green Finance initiated ESG ratings for the Shanghai and Shenzhen 300 Index stocks and published corresponding ESG evaluation indices. Over time, more companies have been incorporated into the SynTao Green Finance ESG rating framework. Hence, the release of rating data for company i in year t by SynTao Green Finance is denoted as 1; otherwise, it is noted as 0.

(3)

Regulating variable

Government environmental attention ($Concern$). The process of constructing the government environmental attention index is as follows: In the first step, government work reports for each province from 2015 to 2021 were collected. In the second step, Python software 3.9.0 was used to convert the government work reports of 31 provinces and cities into text documents and conduct text segmentation. In the third phase, drawing on the study by Chen and Chen (2018) and integrating it with the local government’s work report, this paper compiles and organizes a keyword list pertinent to government environmental focus. Subsequently, the word2vec model is employed to augment the vocabulary related to the environment [41]. A total of 95 environment-related words were extracted. The specifics encompass: safeguarding the environment, reducing emissions, ecological preservation, sustainability, low-carbon practices, ecological balance, preventing pollution, managing pollution, enhancing greenery, fostering green development, wastewater treatment, environmental impact analysis, overseeing environmental standards, ensuring safe disposal of household waste, ambient environmental quality, and atmospheric quality, among others. Ultimately, tally the occurrences of words related to the environment and apply the natural logarithm to the count plus one to represent the government’s level of environmental concern.

(4)

Control variables

Based on the study by Zheng et al. (2023), this study controls for the following variables [2]. Enterprise operational characteristics directly impact the level of corporate green innovation. The selected control variables at the enterprise characteristic level include company size ($Size$), leverage ($Lev$), profitability ($ROA$), ownership concentration ($Herf$), cash holdings ($Cash$), asset structure ($Asset$), and the proportion of independent directors ($Independent$). Additionally, this study introduces government regulatory capacity ($Tech$) and human capital level ($Human$) as macro-level control variables. The descriptions of the main control variables are presented in

Table 1. Description of main control variables.

	Variable Name	Variable Symbol	Variable Definition
Enterprise level Control variable	company size	$Size$	Natural Logarithm of Revenue
	leverage	$Lev$	Proportion of total obligations to overall assets
	profitability	$ROA$	Net earnings to total assets ratio
	ownership concentration	$Herf$	Sum of Holdings Ratio by Top Five Shareholders
	cash holdings	$Cash$	Cash and cash equivalents as a percentage of total assets
	asset structure	$Asset$	Fixed assets to total assets proportion
	the proportion of independent directors	$Independent$	Proportion of Independent Directors
Macro level Control variable	government regulatory capacity	$Tech$	Natural Logarithm of Scientific and Technological Expenditure in Each Region
	human capital level	$Human$	Natural Logarithm of the Average Wage of Urban Employees in Each Region

.

3.3. Data Sources and Descriptive Statistics

This research employs diverse econometric techniques to investigate the influence of ESG disclosure on green innovation among A-share listed manufacturing firms spanning from 2015 to 2021. We followed the approach of Xu et al. (2024) and selected Chinese manufacturing listed companies as the research sample [42]. Additionally, the initial sample underwent the following screening: (1) exclusion of samples that were listed as ST or *ST during the study period; (2) exclusion of samples of companies that delisted during the research period; (3) samples with noticeable irregularities or absent data in financial metrics were omitted. Furthermore, to mitigate the effects of outliers on the regression outcomes, all continuous variables underwent a trimming process of 1% at both the upper and lower ends. The original data for government environmental attention are sourced from government work reports of various provinces, ESG data are sourced from Sino-Securities ESG rating data, the number of green invention patent applications is sourced from the China Research Database Services (CNRDS) database, and other data are sourced from the China Stock Market and Accounting Research Database (CSMAR) database.

The findings shown in

Table 2. Descriptive Statistics.

Variables	Observations	Mean	SD	Min	Max
$Patent$	10,967	0.459	0.823	0	3.638
$ESG$	10,967	0.158	0.365	0	1
$Concern$	10,967	0.017	0.003	0.010	0.027
$Size$	10,967	21.64	1.390	18.42	25.40
$Lev$	10,967	0.411	0.193	0.062	0.929
$ROA$	10,967	0.033	0.072	−0.323	0.200
$Herf$	10,967	0.504	0.143	0.198	0.844
$Cash$	10,967	0.163	0.106	0.019	0.556
$Asset$	10,967	0.227	0.136	0.015	0.627
$Independent$	10,967	3.581	0.521	0	4.151
$Tech$	10,967	14.68	1.004	11.76	16.27
$Human$	10,967	11.25	0.409	10.35	12.06

reveal that the mean natural logarithm of green invention patents stands at 0.459, spanning from 0 to 3.638, with a standard deviation of 0.823, highlighting considerable disparities in the count of green invention patents across various manufacturing firms. Regarding the control variables, the average value of company size is 21.64, the standard deviation is 1.390, the minimum value is 18.42, and the maximum value is 25.40, indicating significant differences in company size among different manufacturing enterprises. In addition, different manufacturing companies also exhibit noticeable differences in factors such as company size ($Size$), leverage ($Lev$), profitability ($ROA$), ownership concentration ($Herf$), cash holdings ($Cash$), asset structure ($Asset$), the proportion of independent directors ($Independent$), government regulatory capacity ($Tech$) and human capital level ($Human$), which will not be further elaborated.

4. Empirical Results

4.1. Baseline Regression

Table 3. Baseline regression results.

Variables	(1)	(2)	(3)	(4)	(5)
	Patent	Patent	Patent	Patent	Patent
$ESG$	0.484 ***	0.220 ***	0.205 ***	0.220 ***	0.722 **
	(23.176)	(9.161)	(8.336)	(9.170)	(2.032)
$Concern$					0.757
					(0.906)
${Concern}^2$					−0.082
					(−0.918)
$ESG \ast Concern$					0.233 *
					(1.932)
${ESG \ast Concern}^2$					−0.077 ***
					(−2.820)
$Size$		0.155 ***	0.137 ***	0.155 ***	0.155 ***
		(19.867)	(18.134)	(19.867)	(19.852)
$Lev$		0.223 ***	0.412 ***	0.222 ***	0.222 ***
		(4.640)	(8.346)	(4.622)	(4.616)
$ROA$		−0.163	−0.118	−0.165	−0.162
		(−1.429)	(−0.981)	(−1.442)	(−1.417)
$Herf$		−0.233 ***	−0.064	−0.233 ***	−0.233 ***
		(−4.213)	(−1.197)	(−4.213)	(−4.207)
$Cash$		−0.027	0.131 *	−0.028	−0.025
		(−0.368)	(1.719)	(−0.377)	(−0.349)
$Asset$		−0.132 **	−0.303 ***	−0.132 **	−0.130 **
		(−2.089)	(−5.165)	(−2.091)	(−2.059)
$Independent$		0.015	0.009	0.015	0.015
		(1.055)	(0.612)	(1.048)	(1.082)
$Tech$			0.120 ***	0.020	0.027
			(11.872)	(0.382)	(0.505)
$Human$			−0.192 ***	−0.076	−0.084
			(−7.689)	(−0.684)	(−0.750)
$Constant$	0.383 ***	−2.912 ***	−2.261 ***	−2.345 *	−4.098 *
	(50.143)	(−17.693)	(−8.672)	(−1.882)	(−1.708)
Time, Industry, and Regional fixed	Yes	Yes	No	Yes	Yes
$N$	10,967	10,967	10,967	10,967	10,967
$R^2$	0.249	0.293	0.124	0.293	0.294

Note: t-statistics are reported in parenthesis; ***, **, and * indicate significance levels 1%, 5%, and 10%, respectively.

showcases the results of the linear regression analysis (H1), examining the effect of ESG on green innovation within companies. Throughout columns (1) through (4) in Table 3, it is apparent that ESG is significantly positive at the 1% threshold. This indicates that ESG reporting has a considerable positive influence on green innovation in companies and enhances green patent output in the manufacturing industry. With column (4) serving as a benchmark, a one-unit increase in ESG disclosure leads to a 22% increase in the level of green innovation among firms. Therefore, these results support research Hypothesis 1.

A potential explanation is that ESG information disclosure enhances the openness and reliability of corporate data, allowing firms to showcase their environmental, social, and governance achievements more thoroughly. This heightened transparency not only aligns with the sustainability expectations of stakeholders like investors and customers, but also boosts the market competitiveness and appeal of businesses. As a result, manufacturing firms gain more financial backing and market demand. Another factor is that ESG disclosure prompts companies to focus more on their internal green innovation practices, optimize resource allocation, boost R&D funding, and strengthen their green innovation abilities. Furthermore, governments may offer additional policy assistance and market rewards to firms with ESG information disclosure, thereby lowering their innovation expenses and enhancing innovation effectiveness.

4.2. Moderation Effect Test of Government Environmental Attention

This research explores the influence of governmental focus on the environment on the link between ESG disclosure and green innovation in firms from the standpoint of environmental attention by the government. The regression outcomes are presented in column (5) of Table 3. $ESG \ast {Concern}^2$ is notably negative, suggesting an “inverted U-shaped” moderation effect of government environmental attention on the association between ESG disclosure and green innovation in manufacturing firms. This result offers evidence in support of Hypothesis 2 of the study.

A possible explanation could be that government focus on the environment and businesses’ green innovation efforts leads to both “innovative impacts” and “cost benefits” [43], with moderate government environmental attention promoting a “cost effect” smaller than the “innovation effect”. Enterprises actively disclose ESG information, respond to government environmental protection policies, and obtain political resources to help reduce financing costs, thereby helping enterprises develop better and faster and promoting green innovation. At the same time, as the government’s attention to the environment continues to increase, when it exceeds a certain limit, the “cost effect” outweighs the “innovation effect”, which will hinder green innovation of enterprises. For example, in the process of obtaining political resources, enterprises will incur rent-seeking costs and may encroach on their green innovation research and development resources. In addition, when the government pays significant attention to the environment, although enterprises have obtained political resources, they may also distort their normal investment and business behavior in order to cater to the government’s “intervention hand” [44], which is also not conducive to green innovation. It is noteworthy that while government emphasis on the environment may diminish the beneficial influence of ESG on businesses’ green innovation to a certain degree, as long as this focus stays within a moderate level, where the “innovative benefits” surpass the “cost implications”, it can still foster corporate green innovation.

5. Robustness Tests

5.1. Substitution of Explained Variables

To assess the robustness of the findings, this research employs a method of substituting the dependent variable. In particular, the study utilizes the count of green patents secured independently ($Patent1$) and collaboratively ($Patent2$) by enterprises as alternative measures for green innovation in manufacturing firms. The regression outcomes presented in columns (1) and (2) of

Table 4. Robustness test results.

Variables	$\mathit{P}\mathit{a}\mathit{t}\mathit{e}\mathit{n}\mathit{t}1$	$\mathit{P}\mathit{a}\mathit{t}\mathit{e}\mathit{n}\mathit{t}2$	$\mathit{F}2\mathit{P}\mathit{a}\mathit{t}\mathit{e}\mathit{n}\mathit{t}$	$\mathit{P}\mathit{a}\mathit{t}\mathit{e}\mathit{n}\mathit{t}$
	(1)	(2)	(3)	(4)
$ESG$	0.176 ***	0.139 ***	0.267 ***	0.126 ***
	(7.565)	(9.427)	(8.919)	(2.614)
Controls	Yes	Yes	Yes	Yes
$Constant$	−1.798	−0.962	−1.333	3.838
	(−1.486)	(−1.254)	(−0.901)	(1.643)
Time-fixed Time, Industry, and Regional fixed	Yes	Yes	Yes	Yes
$N$	10,967	10,967	7808	2061
$R^2$	0.260	0.173	0.300	0.343

Note: t-statistics are reported in parenthesis; *** indicate significance levels 1%.

reveal that the initial regression results maintain their robustness.

5.2. Analyzing Corporate Green Innovation Data for the Next Two Periods

Due to the lengthy process of green innovation in firms, this study employs the total number of green invention patents secured by listed manufacturing enterprises over the following two time frames as the dependent variable for subsequent analysis. The results are shown in column 3 of Table 4. In this context, F2Patent denotes the total count of green invention patents obtained by manufacturing enterprises in the subsequent two periods. The findings indicate a positive ESG coefficient, which supports the reliability of the initial regression outcomes.

5.3. Adjusting the Regression Sample

Given the escalating concern over environmental pollution, green innovation has become particularly noticeable and exemplary in manufacturing enterprises that are significant polluters. As a result, this research concentrated on these companies by consulting the “Classification Directory for Environmental Protection Verification Industries of Listed Firms” (Circular Office (2008) No. 373) issued by the Ministry of Environmental Protection. We narrowed down our research sample to listed companies within this high-pollution manufacturing sector and carried out a reassessment. The findings, shown in column 4 of Table 4, unequivocally indicate that the findings for heavily polluting manufacturing enterprises align with the previous test outcomes.

5.4. Endogeneity Tests

In our baseline regression analysis, we have meticulously controlled for a range of factors that may influence corporate green innovation, including the incorporation of year, industry, and province fixed effects. This approach has significantly reduced the potential for endogeneity in our study. To further address any remaining endogeneity concerns, we employ an instrumental variable (IV) strategy. Drawing on the approach of Zhu et al. (2025), we utilize the number of shares held by “ESG investment funds” (fundnumber) as an instrument for corporate ESG advantage [45]. The rationale for this choice is grounded in the evidence that ESG funds influence corporate behavior through mechanisms such as “voting with their feet” and active monitoring [46], thereby transmitting their investment philosophies to the companies in which they hold stakes [47,48]. Importantly, ESG funds typically do not engage in the day-to-day operations of the firms they invest in, ensuring that their influence on corporate green innovation is indirect rather than direct. Specifically, we use fundnumber as an instrument for ESG information disclosure.

Table 5. Endogeneity test results.

Variables	$\mathit{E}\mathit{S}\mathit{G}$	$\mathit{P}\mathit{a}\mathit{t}\mathit{e}\mathit{n}\mathit{t}$
	(1)	(2)
$ESG$		0.980 ***
		(9.32)
fundnumber	0.009 ***
	(25.52)
Controls	Yes	Yes
Kleibergen–Paap rk LM statistic	625.86 (P: 0.000)
Kleibergen–Paap Wald rk F	651.50
Time, Industry, and Regional fixed	Yes
$N$	10,462
$R^2$	0.016

Note: t-statistics are reported in parenthesis; *** indicate significance levels 1%.

presents the results of the two-stage least squares (2SLS) regression using fundnumber as the instrumental variable. Column (1) reports the first-stage regression results, where the coefficient on fundnumber is positive and statistically significant at the 1% level. This confirms a strong correlation between the number of shares held by ESG investment funds and corporate ESG disclosure, consistent with our theoretical expectations. To assess the validity of our instrument, we conducted several diagnostic tests: First, the first-stage F-statistic is 651.50, well above the conventional threshold of 10, which rules out concerns of a weak instrument. Second, the Kleibergen–Paap Wald rk F-statistic, at 651.50, exceeds the 10% maximal bias critical value of 16.38 from the Stock–Yogo weak ID test, allowing us to reject the null hypothesis of weak identification with high confidence. Third, the Kleibergen–Paap rk LM statistic is significant at the 1% level, rejecting the null hypothesis of underidentification. Together, these tests provide robust evidence that our instrument is both valid and strong. Column 2 presents the second-stage regression results, where the estimated coefficient for ESG disclosure is positive and statistically significant. This finding indicates that, even after addressing potential endogeneity through the instrumental variable approach, our baseline results remain robust. Specifically, the results confirm that corporate ESG disclosure has a positive and significant impact on green innovation, thereby providing further support for Hypothesis H1.

6. Heterogeneity Analysis

6.1. Regional Heterogeneity Analysis

New institutional economics points out that the institutional environment not only provides background conditions for the development of enterprises but also affects their strategic decision-making [49]. In addition, due to the different stages of market economy development, formal systems, and resource endowments, there are significant heterogeneity characteristics in the regional distribution of ESG and corporate green innovation. Based on this, our study divides the entire dataset into three separate regions, Eastern, Central, and Western China, in accordance with the geographical distribution of the country and utilizing the approach described by Yuan et al. (2024) [6]. The objective is to perform a regional heterogeneity examination to explore the impact of ESG information disclosure on green innovation among corporations in different regional settings. The regression results are shown in columns (1) to (3) of

Table 6. Heterogeneity analysis results.

Variables	Eastern	Central	Western	The “Strong” Confucian Culture Group	The “Weak” Confucian Culture Group
	(1)	(2)	(3)	(4)	(5)
$ESG$	0.250 ***	0.162 ***	0.084	0.249 ***	0.163 ***
	(8.167)	(3.079)	(1.501)	(8.255)	(4.106)
Controls	Yes	Yes	Yes	Yes	Yes
$Constant$	−3.299 *	−3.022	−0.276	0.300	−4.034 ***
	(−1.714)	(−1.423)	(−0.102)	(0.152)	(−2.611)
Time, Industry, and Regional fixed	Yes	Yes	Yes	Yes	Yes
Formation coefficient difference test p-value				0.070
$N$	7360	2021	1586	7216	3751
$R^2$	0.302	0.410	0.447	0.306	0.381

Note: t-statistics are reported in parenthesis; *** and * indicate significance levels 1% and 10%, respectively.

.

Considering regional differences, ESG exerts a notably stronger positive effect on green innovation in corporations located in the eastern and central areas. This disparity arises primarily because these regions boast a more advanced economic development and well-established formal frameworks compared to the western region. As a result, people in these regions demonstrate a higher inclination to prioritize corporate social responsibility and ESG disclosure, which in turn promotes green innovation within companies. Conversely, the influence of ESG disclosure on green innovation in the western region is comparatively restricted. This is attributable to the slower uptake of ESG disclosure practices by western Chinese companies in contrast to their counterparts in the eastern and central areas. Furthermore, there is a lower level of awareness and acceptance of ESG principles in the west, resulting in the advantages of ESG disclosure not being fully harnessed.

6.2. Analysis Based on Confucian Culture

The ancients said, “If you can improve yourself in a day, do so; if you can improve yourself daily, do so daily; continue to improve yourself every day” and “Though Zhou is an old state, it is still capable of new endeavors” (from “The Book of Songs”). Such innovative thoughts and the spirit of continuous renewal inherent in the excellent traditional Chinese culture propelled China to lead in technological innovation from the Han to the Ming dynasty [50]. Since ancient times, China has highly valued its excellent traditional culture. Confucian culture, as the essence of this traditional culture, contains rich economic management theories. For example, in the current context of weak intellectual property protection and incomplete legal systems in China, the core ethical thoughts of Confucianism, such as “benevolence, righteousness, propriety, wisdom, and trust” not only help regulate the behavior of corporate competitors, reduce the risk of research and innovation achievements being plagiarized by competitors, and stimulate corporate innovation enthusiasm, but also impact corporate ESG information disclosure. Moreover, Confucianism is deeply ingrained in the values of Chinese entrepreneurs and influences their strategic decisions [51]. The innovative spirit advocated by Confucianism, “renewing oneself and breaking new ground”, aligns closely with modern scientific and technological innovation and will inspire corporate innovative behavior. The level of Confucian cultural impact in a company’s registration region mirrors the extent to which the company is shaped by Confucian values, consequently influencing its decisions related to green innovation. Following the approach of Huang et al. (2024) [12], the number of Confucian academies, schools, and Confucian temples was measured by adding 1 to the logarithm, and the overall sample was divided based on the median number (2.8). Divide samples above the median into strong Confucian culture and samples below the median into weak Confucian culture to explore the impact of ESG on corporate green innovation. These data are sourced from the “Unified Annals of the Qing Dynasty”,“Unified Annals of the Ming Dynasty”, and local chronicles of the Ming and Qing dynasties, and are manually collated. The regression outcomes are presented in Table 6, specifically in columns 4 and 5. Column 4 corresponds to the group with a “high” level of Confucian culture (Confucian culture ≥ 2.8), while column 5 represents the group with a “low” level of Confucian culture (Confucian culture < 2.8).

The findings indicated a notably positive effect of ESG information disclosure on green innovation in companies for both groups with “low” and “high” levels of Confucian culture. Nevertheless, the regression coefficient for the group with a “strong” Confucian culture (0.249) was markedly higher compared to the group with a “weak” Confucian culture (0.163), and this difference was confirmed by the inter-group coefficient comparison test. This indicates that regions with a strong Confucian cultural background experience a more significant positive impact of corporate ESG on green innovation, suggesting that Confucian culture may play a role in shaping corporate green innovation practices.

7. Further Research and Discussion

7.1. Financing Constraints

Based on the signaling theory, firms convey favorable indicators of their robust performance through the disclosure of ESG information to the marketplace. This helps to mitigate information asymmetry between firms and their stakeholders, boosts operational transparency, lowers investors’ risk perceptions, draws in more potential investors, expands financing avenues, and furnishes financial backing for firms to pursue green innovation [2]. Over the long haul, given the substantial upfront investment costs, prolonged payback periods, and considerable uncertainty linked to corporate innovation, firms with constrained funds take into account financing limitations when making decisions about engaging in green innovation endeavors [52]. Raimo et al. (2021) discovered that firms with robust ESG information disclosure are more prone to secure external resources, including third-party financing. Moreover, they face lower risks in terms of regulation, litigation, and technology [53,54], making it easier for them to secure government support, attract media attention, and gain favor from stakeholders. This helps alleviate financing constraints [55], thereby assisting companies in achieving green innovation. Consequently, drawing on the research by Qian et al. (2017), this study utilizes the WW index as a surrogate measure for financing constraints ($FC$). It delves deeper into the way ESG information disclosure, by easing financing constraints, influences the pathway of green innovation within firms. A larger WW index indicates more substantial financing constraints in companies [56].

The effects of ESG on financing constraints are presented in columns 1 and 2 of

Table 7. Mechanism test result.

Variables	Financing Constraints		Agency costs
	(1)	(2)	(3)	(4)
$ESG$	−0.198 ***	−0.160 ***	−0.027 ***	−0.011 ***
	(−19.071)	(−15.619)	(−15.958)	(−6.514)
Controls	No	Yes	No	Yes
$Constant$	−0.804 ***	−1.742 ***	0.092 ***	0.326 ***
	(−212.135)	(−2.914)	(151.639)	(3.446)
Time, Industry, and Regional fixed	Yes	Yes	Yes	Yes
$N$	10,967	10,967	10,967	10,967
$R^2$	0.219	0.303	0.273	0.371

Note: t-statistics are reported in parenthesis; *** indicate significance levels 1%.

. The findings suggest that, irrespective of the inclusion of control variables, ESG information disclosure notably decreases financing constraints. Essentially, ESG mitigates information asymmetry, boosts stakeholder engagement, and assists firms in securing more external financing, ultimately offering financial backing for green innovation.

7.2. Agency Costs

Initiating green innovation activities requires significant resource investment, yet it comes with a long payback period and uncertainty, making it easy for corporate management to abandon green innovation decisions in favor of pursuing short-term operational performance and economic cash flows, thereby hindering the enhancement of a company’s green innovation level [57]. From the perspective of corporate development and expansion, external financing is crucial for reducing agency costs stemming from information asymmetry, and it is vital for the company’s growth [58]. Additionally, behaviors such as excessive pollution emissions, excessive environmental investments, and resource consumption in companies will lead to principal-agent problems [59], which are also unfavorable for companies to make green innovation decisions.

However, most studies suggest that companies can reduce corporate agency costs by fulfilling their social responsibilities. This is mainly based on two factors: first, from the perspective of signaling theory, actively fulfilling social responsibilities can enhance a company’s reputation and image. By disclosing corporate operating information through media reports and information disclosure, it helps stakeholders monitor the company’s daily business activities and investments, thereby reducing monitoring costs. It also promotes communication between stakeholders and shareholders, to some extent, reducing agency costs. Second, ESG by companies covers relevant information such as environmental protection investments and business development. This helps expand the knowledge rights of the public, investors, and other stakeholders regarding corporate information, encouraging greater transparency in the decision-making and investment activities of corporate management. Therefore, ESG information disclosure can to some extent constrain the short-sightedness and self-interest of management, further reducing corporate agency costs and reducing the likelihood of corporate management abandoning green innovation decisions due to the uncertainty of green innovation risks. To validate the connection between ESG, reduced agency costs, and the promotion of corporate green innovation, this study refers to the work of Li et al. (2022) and employs the administrative cost ratio as a measure of agency costs ($AC$). A higher administrative cost ratio indicates greater agency costs. The administrative cost ratio is determined as the proportion of management expenses to operating revenue [60].

Columns 3 and 4 of Table 7 present the impact of ESG on agency costs. The results indicate that ESG can affect corporate green innovation by reducing agency costs.

8. Conclusions and Implications

8.1. Conclusions

This study conducts an empirical investigation into the influence and moderating influence of ESG on green innovation in corporations, utilizing data from publicly listed manufacturing firms. The findings suggest that ESG is advantageous for enhancing green innovation among manufacturing companies. The government’s emphasis on environmental issues demonstrates a significant “inverted U-shaped” regulatory influence on the link between ESG disclosure and corporate green innovation. A moderate level of government attention to the environment enhances green innovation in manufacturing firms through ESG disclosure, whereas excessive attention diminishes the positive relationship between the two. The heterogeneity analysis shows that ESG exerts a more pronounced positive impact on green innovation in firms located in the eastern and central regions, as well as in areas with a strong Confucian cultural background. Further findings suggest that ESG information disclosure can ease financing constraints and lower agency costs, thereby fostering green innovation in businesses.

8.2. Implications

The findings of this study carry significant policy recommendations. Initially, corporations ought to embed the principle that “pristine environments are priceless resources” into their organizational culture, elevate their ESG information disclosure standards, and align ESG with their green innovation strategies. Enhancing the standard and frequency of ESG information disclosure will further strengthen the benefits that such disclosure brings to boosting corporate green innovation capabilities.

Second, we consider the “inverted U-shaped” effect of government emphasis on the environment on the relationship between ESG and green innovation in manufacturing companies. This suggests that local governments ought to adopt tailored and flexible environmental preservation approaches, keep governmental environmental attention at a moderate level, fully utilize the beneficial impact of ESG on corporate green innovation practices, and aim for a mutually beneficial outcome among ecological preservation, ESG, and corporate green innovation advancement. Once again, enterprises should implement precise policies and design targeted strategies based on regional heterogeneity and the characteristics of Confucian culture. In western regions and areas with weak Confucian culture, it is necessary to actively develop infrastructure construction, strengthen cultural confidence, fully leverage cultural soft power, enable enterprises to benefit from ESG, and promote green innovation.

Furthermore, further research indicates that companies should focus on ESG information disclosure, incorporate ESG into their daily business operations, fully leverage the role of ESG in alleviating financing constraints and reducing agency costs, and seek support and resources from government and other stakeholders. Fully unleash the contribution of ESG to the development of corporate green innovation.

Finally, our research findings offer insights for developing countries. For instance, nations increasingly prioritizing environmental issues and pursuing sustainable development should emphasize ESG information disclosure to harness its positive effects on manufacturing firms’ green innovation. For developing countries actively promoting green transition and strengthening environmental regulation, a moderate level of governmental environmental attention can provide clear policy guidance and incentives, fostering a virtuous cycle between ESG disclosures and green innovation. Excessive attention, however, may lead to an overwhelming policy burden, dampening corporate innovation incentives. Additionally, given the significant divergence in ESG ratings for the same corporation across different rating agencies, the impact of ESG rating discrepancies on manufacturing firms’ green innovation warrants further exploration in future research.

8.3. Limitations

One notable limitation of the current research lies in the inherent shortcomings of existing ESG (Environmental, Social, and Governance) scoring systems. Specifically, the lack of standardized criteria across different rating agencies has resulted in inconsistent and divergent ESG scores, which can lead to confusion and controversy in both academic and practical contexts. This inconsistency undermines the reliability and comparability of ESG data, posing challenges for researchers and practitioners seeking to draw meaningful conclusions or make informed decisions based on these metrics. Consequently, the development of a unified, transparent, and credible ESG scoring framework and algorithm emerges as a compelling and significant area for future academic inquiry. Addressing this issue would not only enhance the robustness of ESG-related research but also contribute to the broader adoption and effectiveness of ESG practices in promoting sustainable development.