Corporate Social Responsibility and Investor Relations Management: Evidence from China

Abstract

The implementation of corporate social responsibility (CSR) in conjunction with proficient investor relations management (IRM) can enhance the reputation and appeal of enterprises, thereby fostering the sustainable development of enterprises. This paper examines the correlation between CSR and IRM by exploring the potential misinterpretation of socially responsible actions by listed companies as “hypocrisy”. We use the fixed effect model, moderating effect model and instrumental variable method to examine the correlation between CSR and IRM. The findings indicate that actively fulfilling corporate social responsibility can enhance interaction and communication between listed companies and investors in the capital market, thereby mitigating the risk of being perceived as “hypocrisy”. This positive effect is particularly pronounced when companies are experiencing poor operational performance. These conclusions remain robust even after conducting various tests to address endogeneity concerns. In terms of the underlying mechanisms, corporate social responsibility primarily enhances investor relations management through strengthening network communication and on-site interactions. Moreover, enterprises are more inclined to proactively interact with investors in the capital market when companies face severe financial difficulties, stringent financing constraints, or poor quality of information disclosure. Additionally, our study extends its analysis to elucidate how corporate social responsibility can mitigate the risk of stock price crashes from the perspective of investor relations management.

1. Introduction

Corporate social responsibility (CSR) refers to the fact that, in the course of its business operations, in addition to pursuing economic profits and satisfying the interests of shareholders, an enterprise also achieves a win–win situation for all its stakeholders (including employees, consumers, suppliers, communities, the environment, etc.) as well as the wider social and natural environment. This pattern has gradually become a consensus [1]. The concept emphasizes that while pursuing economic benefits, companies must balance social, environmental and ethical responsibilities to achieve sustainable development [2]. The discussion surrounding corporate social responsibility (CSR) has been had for a long time and has garnered significant attention from various sectors of society. In the early stages, research primarily focused on the question of whether companies should undertake social responsibility, with Friedman [3] being a prominent figure in this debate. Most middle managers at the time perceived CSR as an additional cost and burden, leading academics such as Milton Friedman to oppose the concept. Friedman argued that executives might exploit shareholder funds under the guise of CSR. Furthermore, some managers held the belief that CSR would damage the financial performance of the company, resulting in limited attention being given to CSR in its nascent stages [4].

In the 1960s, the stakeholder concept emerged in the field of management literature [5]. Since then, the discussion surrounding this concept has remained ongoing as it continues to gain prominence in enterprise theory. The introduction of stakeholder theory has established a clear framework for CSR [6]. Through active participation in governance, stakeholders can alleviate the agency problems within enterprises and motivate them to fulfill their social responsibilities more effectively [7]. Nevertheless, certain studies still argue that CSR can be a resource drain to some extent, as it consumes resources that could otherwise be allocated to core business activities, potentially impacting the economic returns of enterprises [8,9]. Additionally, concerns have been raised that managers may engage in excessive CSR investments for personal gain [10].

After nearly a century of development, the concept of CSR has evolved into a comprehensive and influential notion [11] (Corporate social responsibility and legal framework (information relevant to the situation in China): Environmental Protection Law of the People’s Republic of China (revised in 2014); Company Law of the People’s Republic of China (Fourth Amendment in 2023). For example, Article 5 of the Company Law clearly stipulates: “In conducting business activities, a company must abide by laws and administrative regulations, abide by social ethics and business ethics, be honest and trustworthy, accept the supervision of the government and the public, and assume social responsibilities.” This provision provides a legal basis for the fulfillment of corporate social responsibility). Existing studies have generally demonstrated that enterprises can significantly enhance their financial performance and capital market standing by fulfilling their social responsibilities [12,13,14]. From one perspective, engaging in socially beneficial actions can enhance an enterprise’s reputation and foster consumer loyalty [15,16]. On the other hand, companies have also succeeded in implementing a win–win approach known as “doing well by doing good thing”. However, within the field of CSR, scholars have continuously debated whether enterprises actively fulfilling social responsibilities are “genuinely altruistic” or merely engaging in “corporate hypocrisy”. Some studies have pointed out that enterprises’ purported fulfillment of social responsibilities may be a form of “saying one thing and doing another” [17,18]. It is suggested that such behavior might be used by professional managers to conceal poor performance or unethical conduct within the organization [19]. Thus, CSR is more likely to be perceived as a self-serving behavior on the part of managers [20]. It is precisely because CSR may be misunderstood by market investors as a means for companies to cover up bad truths that enterprises should pay more attention to IRM when performing CSR.

Therefore, when an enterprise actively performs its social responsibilities or achieves a high score in the third-party CSR evaluation system, it runs the risk of being misinterpreted by investors in the capital market as a self-serving behavior by corporate managers or as an attempt to conceal poor performance [19]. If this misunderstanding persists over time, genuinely virtuous enterprises may be unjustly labeled as hypocritical due to concerns about their exceptional performance in social responsibility, potentially diminishing their motivation to fulfill their social responsibilities. Apart from facing public and media condemnation, there appear to be limited effective measures to address the issue of CSR hypocrisy [21]. Consequently, hypocritical companies may engage in improved social responsibility performance driven by motives such as “cover-up” or “self-interest”. In the long run, this may lead enterprises into a detrimental cycle in which bad money drives out good. Therefore, fostering effective communication between enterprises and existing or potential investors in the capital market becomes crucial in order to prevent misunderstandings and ensure accurate perceptions of CSR efforts.

As we all know, information asymmetry is a common problem in the capital market. By strengthening interactive communication, enterprises can reduce information asymmetry between investors and enable investors to make investment decisions based on more comprehensive information, thus protecting the legitimate rights and interests of investors. Furthermore, effective investor relations management (IRM) is the basic manifestation and external extension of good corporate governance. IRM encompasses the communication and engagement between enterprises and investors, with the objective of cultivating transparent and steadfast investor relationships, attracting investment, and augmenting corporate value and reputation. The proactive discharge of social responsibilities by enterprises can bolster investors’ confidence in the businesses and contribute to the stabilization of investor relations. Existing studies have found that through interactive communication with investors, enterprises can establish a closer investor relationship network, enhance investors’ sense of identity and belonging to enterprises [22], reduce the volatility and uncertainty of the stock market, and provide investors with a more stable and reliable investment environment [23]. Previous authors believed that CSR was a way to cover up bad behaviors in enterprises, which would harm the interests of investors. As a result, the more companies focus on CSR, the worse the investor market may react [24,25]. Therefore, while performing CSR, enterprises should do better in the dissemination of true information and IRM. From this perspective, if appropriate and necessary investor relations management (IRM) is not implemented, companies that seriously implement CSR and promote sustainable development may be mistaken as “hypocritical” enterprises, which is not conducive to the long-term sustainable development of enterprises. Moreover, the fulfillment of corporate social responsibility and the reinforcement of investor relations management jointly promote the sustainable development of enterprises. Enterprises enhance their competitiveness and sustainable development ability by fulfilling social responsibility, while investors realize their investment objectives and value pursuit by paying attention to and supporting enterprises with social responsibility. This virtuous circle helps to promote the sustainable development of the whole society.

Then, how does CSR affect the IRM of listed companies? How does the IRM of listed companies limit the positive transmission mechanism of CSR? How will the capital market react to CSR? This paper focuses on non-financial listed companies in the Shanghai and Shenzhen A-share markets from 2011 to 2020, and empirically assesses the influence of CSR ratings on the level of IRM in listed companies and stock price crash risk. This article makes three main marginal contributions. First of all, most of the previous literature starts with the listed companies themselves, regarding how they fulfill their social responsibilities, paying attention to the “true kindness” or “hypocrisy” motivation of listed companies to live up to their social obligations [17,18,26], and seldom focusing on investors’ views on their performances. This paper innovatively researches the process of enterprises’ social responsibilities from the perspective of investor relations management. How to prevent being misunderstood by investors as a “hypocrisy” behavior is assessed by the relevant research on corporate social responsibility. The findings suggest that companies better at CSR can protect themselves from being misunderstood by strengthening their IRM. Secondly, previous studies on investor relationship management focused more on the specific practices of investor relationship management, and less on investor relationship management strategies under the scenario that enterprises actively fulfill their social responsibilities. This study complements relevant studies on investor relationship management. The active and effective internal management of listed companies promotes the positive transmission mechanism of corporate social responsibility. The conclusion provides strategic references for listed companies to strengthen their internal management while fulfilling their social responsibilities. Finally, there are many controversies about the economic consequences of CSR performance in existing studies. For example, there is debate over whether CSR performance reduces or increases the risk of stock price crashes. The results of this paper support the former conclusion and provide new evidence from the perspective of investor relations management.

The remainder of the paper is arranged in the following ways. Section 2 provides a literature review and empirical studies and hypothesis development. Section 3 is concerned with the empirical model and describes variables and data. Section 4 presents the empirical results of the study. Section 5 presents further mechanism analyses; Section 6 offers an extended analysis of stock price crash risk. Section 7 presents the conclusions of the study.

2. Background and Hypotheses Development

2.1. Literature Review

The academic literature on CSR primarily examines the decision-making process of whether and how to undertake social responsibility, and the resulting economic outcomes [3,27,28]. However, the research covered in this paper introduces another key actor in this context: investors in the capital market. The study primarily explores the impact of enterprises fulfilling their social responsibilities on the performance of the capital market, and addresses the issue of CSR hypocrisy.

2.1.1. CSR and Stock Price Crash Risk

The existing literature primarily focuses on the capital market’s response following the active fulfillment of social responsibilities by enterprises, with particular attention given to stock price crash risk [19,20,29,30,31,32], stock price synchronicity [33], market return rate [34], stock liquidity [35], investor response [36], etc. There are two opposing views on the risk of a share price crash. One is that CSR will reduce the risk of a stock price crash [30,31]. Another view is that CSR will increase the risk of a stock crash [19,20]. CSR is an activity that pursues long-term sustainability. Improving the quality of CSR participation can reduce the synchronicity of stock prices [33]. At the same time, the active implementation of CSR also helps to improve market returns [34] and promote stock liquidity [36].

However, many studies in the aforementioned literature have arrived at similar conclusions, with the relationship between CSR and stock price crash risk being the most contentious. One perspective argues that CSR exhibits a significantly positive correlation with stock price crash risk, implying that improvements in CSR may elevate the likelihood of stock price crashes occurring [19,20]. For instance, Quan et al. [20] assert the presence of a “collapse effect of social responsibility” in China’s capital market, suggesting that the implementation of social responsibility by listed companies reflects more “instrumental characteristics” than “value-creating characteristics”. There are two possible reasons why the increase in CSR will increase the risk of stock price crashes. First, companies may perform their social responsibilities out of self-serving motives to cover up their internal problems and risks. In such cases, CSR can become a “self-serving tool” rather than an expression of genuine concern for social and environmental interests. If companies over-package or falsely advertise their social responsibility activities, once these behaviors are revealed, it may cause distrust and panic among investors, thus increasing the risk of a stock price collapse. Second, fulfilling social responsibility requires enterprises to invest certain resources and energy. If a company excessively pursues social responsibility while neglecting its core business and profitability, it may lead to the uneven distribution of resources and increased operating pressure. In extreme cases, this can have a negative impact on a company’s financial health and share price performance, increasing the risk of a share price crash. In other words, if the company conceals negative information or discloses insufficient information, the information may be suddenly released to the market after accumulation to a certain extent, which will have a huge impact on the stock price, thereby increasing the risk of a stock price collapse.

Nonetheless, a growing number of scholars support the notion that CSR actually serves as a mitigating factor against stock price crash risk [29,30,31,32]. This is because the intuitive feeling is that the fulfillment of corporate social responsibility is often accompanied by higher information disclosure and transparency. When enterprises actively fulfill their social responsibilities, they are more inclined to disclose their business conditions and related information, which helps to reduce information asymmetry and the accumulation of negative information, thus reducing the risk of a stock price crash. Huang et al. [29] pointed out that fulfilling social responsibility significantly increases media coverage, thereby curbing the risk of a stock price collapse. This aligns closely with the research theme of our study, which emphasizes IRM in listed companies. In this context, media reporting plays a crucial role. However, it is important to note that the media reports discussed by Huang et al. [29] primarily pertain to third-party reports and represent a passive strategy. In contrast, the IRM highlighted in our study refers to the active interaction between listed companies and investors in the capital market, utilizing financial communication and marketing regulations as part of the company’s strategic management behavior [37].

2.1.2. Corporate “Hypocrisy” and IRM

Research examining corporate “hypocrisy” in the context of social responsibility suggests that there is a prevalent issue of companies “saying one thing and doing another” when actively fulfilling their social responsibilities [17,18]. In the Chinese capital market, certain companies that have previously received positive ratings for their social responsibility performance have encountered negative incidents. For example, Vanke Group, the recipient of the “China’s Best Corporate Citizen Award”, has been involved in incidents like the “toxic floor incident” and the “paper gate incident” [20]. These events have raised questions among capital market investors regarding the credibility of companies’ outstanding social responsibility performance. Consequently, corporate social responsibility, which should ideally enhance corporate value, can instead be perceived as “hypocrisy” [38,39]. The existing literature has made a lot of efforts to explore ways for companies to promote the fulfillment of social responsibility and establish a positive transmission mechanism, often relying on third-party media reports or self-repair within the capital market. However, proactive strategies employed by companies themselves have received limited attention. Particularly in the context of the “hypocrisy” theory and negative social responsibility events, active IRM strategies become even more crucial for companies.

Investor relations management (IRM) refers to the practice adopted by listed companies of enhancing interactive communication with investors and the capital market through comprehensive, timely, and voluntary information disclosure, facilitating informed decision-making by investors [40]. IRM aims to simultaneously increase shareholders’ wealth and create value for other stakeholders, making it a strategic management behavior that maximizes a company’s overall value [41]. Within this framework, the disclosure of CSR and the active fulfillment of social responsibility fall under the purview of the IRM of listed companies. However, similar to CSR, some criticize IRM as a short-term and superficial endeavor [42]. But in the long run, IRM can increase company value. When scholars measure the level of IRM in detail, they tend to focus more on the interaction and communication between listed companies and investors in the capital market [37]. Active IRM offers significant benefits for companies. For example, it reduces information asymmetry in the capital market and enhances the efficiency of information dissemination [43]. It also lowers capital costs for companies [40] and mitigates the risk of stock price collapse [44]. Furthermore, a higher frequency of investor communication (or greater investor participation) leads to a more considerable increase in a company’s market value. This effect is particularly pronounced in companies with lower information transparency and greater performance volatility [45].

Regrettably, when companies possess a significant level of social responsibility, there is a pressing question of how to effectively and genuinely convey relevant information to the capital market, particularly considering the potential misinterpretation of such social responsibility as “hypocrisy”. Surprisingly, the existing literature offers limited insights into addressing these misunderstandings, thus creating a research gap that this paper aims to fill.

2.2. Hypothesis Development

2.2.1. The Correlation between CRS and IRM

Corporate social responsibility (CSR) means that while creating profits and taking legal responsibility for shareholders and employees, enterprises emphasize their contribution to the environment, consumers and society. The fulfillment of corporate social responsibility not only helps to enhance the social image and brand value of the enterprise, but also conveys the signal to investors that the company pays attention to sustainable development. Theoretically, on the one hand, the fulfillment of corporate social responsibility can significantly enhance the social image and brand value of enterprises, and help enhance investors’ trust and recognition of companies [22,46]. When investors see companies actively taking social responsibility, they are more likely to put money into such enterprises, thus driving up the company’s share price and market value. On the other hand, investor relations management requires companies to disclose relevant information in a timely, accurate and comprehensive manner in order to meet investors’ information needs [47]. The performance of corporate social responsibility is essentially one of the important pieces of information that investors focus on. Therefore, while fulfilling their social responsibilities, enterprises need to strengthen communication with investors and disclose relevant information in a timely manner, so as to enhance investors’ understanding and recognition of the company and avoid misunderstandings about on its true intention of CSR.

Based on the literature reviewed above, despite the ongoing debate surrounding the concepts of “truth and goodness” versus “hypocrisy” in the realm of CSR, it is unquestionable that enterprises aim to cultivate a responsible and virtuous “good citizen” image through their social responsibility endeavors. This is primarily because companies anticipate long-term benefits that outweigh the costs associated with fulfilling social responsibilities [5]. Therefore, when enterprises actively engage in social responsibility initiatives or obtain high third-party CSR ratings, they are more inclined to proactively disclose social responsibility information and provide relevant details to investors in the capital market. Moreover, the fulfillment of CSR significantly amplifies media coverage [29]. In summary, these factors collectively contribute to an increased frequency of interaction between listed companies and investors.

In addition, within the realm of studying the “hypocrisy” phenomenon in CSR, some of the literature suggests that one of the motivations behind enterprises fulfilling their social responsibilities may be to disguise their poor management or unethical conduct [48]. In accordance with signaling theory, whether a company embodies “true goodness” or “hypocrisy” is considered private information, leading even companies with “hypocritical” intentions to actively publicize their social responsibility achievements to garner investor favor. Consequently, investors’ doubts and concerns regarding CSR hold some validity. However, this skepticism can pose additional challenges for genuinely well-intentioned companies, as irrational investors may easily interpret exceptional social responsibility performance as “hypocrisy” or self-interested managerial behavior [19]. Therefore, when enterprises with “true goodness” exhibit high social responsibility performance, they tend to prioritize IRM, fostering stronger interaction and communication with the capital market to prevent misunderstandings. Building upon the aforementioned analysis, this paper proposes the first hypothesis, Hypothesis 1, to be tested, as follows:

Hypothesis 1.

Holding all other factors constant, there is a positive association between the score of CSR and the level of IRM.

2.2.2. The Impact of Business Performance

Because of the covert nature of “hypocritical” behavior, it becomes challenging to directly observe the actual instances of such behavior in enterprises. As a result, it is plausible that the sample of companies with high social responsibility scores includes not only genuinely “good” companies but also a substantial number of “hypocritical” companies. In this case, the mandatory disclosure of enterprise-related information, encompassing both financial and non-financial aspects, becomes the primary observation window for investors in the capital market. Investors often rely on this public information to categorize companies as either “good” or “bad”. Among these, the most sensitive information pertains to the business performance of companies. On the one hand, the business performance typically constitutes the most influential factor impacting a company’s performance in the capital market [49]. Business performance directly reflects the profitability, operation efficiency, cost control ability, market position and other aspects of an enterprise [50,51]. The capital market is a collection of investors’ expectations about the future performance of enterprises. When an enterprise shows strong operating performance, it means that it has high profitability, as well as good asset management and financial robustness, which will directly affect investors’ confidence and expectations regarding the future development prospects of the enterprise. On the contrary, if the enterprise is not well run, investors will lower their expectations for the enterprise, resulting in a decline in stock prices and poor performance of the capital market. On the other hand, the “hypocritical” motives underlying CSR behavior are occasionally perceived as self-serving motives of professional managers or controlling shareholders [20]. This potentially leads to a decline in business performance [52], which will in turn lead to the poor performance of the capital market.

Theoretically, those “hypocritical” companies operating behind a facade may be more inclined to engage in active IRM to conceal their “hypocritical” behavior as much as possible, particularly after attaining a high social responsibility score. The main reason may be information asymmetry. When investors obtain corporate information, they may not be able to fully understand the real situation of the company. If enterprises overemphasize social responsibility and neglect the disclosure of business performance, it will easily lead to investors’ misunderstanding of enterprises. In addition, investors may be affected by their own experience, emotions, biases and other factors when evaluating the value of enterprises, resulting in misunderstandings. For example, when a company’s operating performance is poor, investors may be more likely to view its socially responsible behavior as “hypocrisy” or “showmanship”.

Therefore, when a company exhibits subpar operational performance while maintaining a commendable social responsibility score, it becomes susceptible to misinterpretation and the label of “hypocrisy”. Considering that engaging in CSR endeavors can mitigate the likelihood of financial distress for firms [53], it is expected that these firms will enhance their communication and engagement with investors to evade misunderstandings. While a high likelihood of hypocrisy does not equate to “true” hypocrisy, it is imperative to also focus on companies with a heightened probability of hypocrisy. Empirically speaking, such companies can serve as representative samples of entities with inadequate operational performance yet a strong commitment to social responsibility. Building on the aforementioned analysis, this study proposes the second hypothesis (H2) for examination as follows:

Hypothesis 2.

Holding other conditions constant, the positive influence of CSR on IRM will be more pronounced in the presence of poor operating performance.

3. Research Design

3.1. Data and Sample

First of all, considering the big difference between the financial data of financial companies and that of non-financial companies, we retrieved non-financial companies’ data from China’s “A” stock market. Secondly, the main dependent variable of this study was CSR, which was obtained from Hexun Network www.hexun.com (accessed on 15 June 2023) (Hexun Information Technology LTD. Hexun.com is a leading financial information and online mobile financial services provider in China. Founded in 1996, Hexun was one of the first professional ICPs in China. It is a website with an internet news and information service license, an information network communication audiovisual program license and securities investment consulting qualifications. Hexun actively fulfills its social responsibility by providing high-quality financial information and services to help investors better understand market dynamics and make wise investment decisions), as announced in 2010. We used the Investor Relations Interaction Index (IRII) to measure the level of IRM, obtained from the Nankai University Investor Relations Interaction Index Database, currently updated to 2019. To alleviate the impact of endogenous issues, the explanatory variable was uniformly t + 1 periods. Therefore, the actual sample period was from 2011 to 2020. According to research needs, we processed the initial samples as follows: ① The financial industry and listed firms in the sample symbolized as ST and *ST were removed. ② We Winsorized the main continuous variables at the 1% and 99% percentiles to reduce the impact of extreme values, and other companies’ financial data were obtained from China Stock Market & Accounting Research (CSMAR).

3.2. Measurement of Key Variables

3.2.1. Dependent Variable: IRM Level

We utilized IRII as a measure of the main dependent variable, sourced from the Nankai University Investor Relations Interaction Index Database [37]. This index comprises five primary indicators: communication guarantee, online communication, telephone communication, on-site communication, and communication feedback. Communication guarantee pertains to the organizational assurances for engaging with investors. Online communication involves conducting online roadshows via the internet. Similarly, telephone communication entails hosting teleconferences. On-site communication involves conducting performance briefing meetings, media gatherings, and other activities through face-to-face interactions. Communication feedback primarily occurs through email correspondence. Please refer to Appendix A for a detailed breakdown of the specific secondary indicators.

3.2.2. Independent Variable: CSR Score

The main independent variable in this study is the Corporate Social Responsibility Rating (CSR score), sourced from Hexun Network [54]. The CSR score encompasses five dimensions: shareholder responsibility, employee responsibility, supplier and consumer rights responsibility, environmental responsibility, and social responsibility, accounting for 30%, 15%, 15%, 20%, and 20% of the total score, respectively. Each dimension is associated with its own set of secondary and tertiary indicators. In total, there are 13 secondary indicators and 37 tertiary indicators involved in this measurement. The CSR score provides an accurate assessment of the extent to which CSR is fulfilled. Furthermore, a higher score indicates a greater level of corporate social responsibility fulfillment.

3.2.3. Control Variables

Several driving factors influence levels of IRM [28,29,33,36]. We incorporated various control variables at the firm level, including corporation scale (lnsize), corporation fixed number of year (Age), asset–liability ratio (Lev), operating cash flow (Cashflow), return on assets (ROA), book-to-market (MB), stock turnover rate (Turnover), shares of controlling stockholders (TOP1), the top ten majority shareholding ratio (TOP10), the separation of ownership and control (Sep), executive part-time (Power) and number of professional committees (Committee). Large-scale enterprises typically attract more investors, necessitating a more sophisticated and structured approach to investor relations management [36]. The age of a business serves as a reflection of its historical trajectory and market reputation. Established enterprises tend to cultivate a more enduring brand image and reputation, fostering increased investor confidence [30]. The asset–liability ratio (Lev), operating cash flow (Cashflow), and return on assets (ROA) are utilized to mitigate the impact of an enterprise’s financial health on IRM [31]. Similarly, the book-to-market ratio (MB) and turnover ratio are leveraged to regulate the influence of capital market performance on stock trading outcomes [29]. Factors related to corporate governance, including the shareholding ratio of controlling shareholders (TOP1), the shareholding ratio of the top ten shareholders (TOP10), the separation of ownership and control (Sep), the dual roles of senior executives (Power), and the composition of professional committees (Committee), are employed to control the effects of corporate governance on IRM [36]. In addition, to provide more robust estimation results, we also control for the firm-fixed effect (Firm FE) and time-fixed effect (Year FE). Descriptive statistics of our sample are presented in

Table 1. Summary statistics.

Variables	Observations	Mean	S.D.	Min	p25	p50	p75	Max
Dependent variable
IRII	15,047	0.323	0.118	0.009	0.229	0.328	0.410	0.885
IRII_x1	15,047	0.442	0.261	0	0.250	0.500	0.591	1
IRII_x2	15,047	0.574	0.170	0	0.500	0.609	0.683	1.875
IRII_x3	15,047	0.291	0.279	0	0	0.458	0.474	1.594
IRII_x4	15,047	0.138	0.170	0	0	0.117	0.250	1
IRII_x5	15,047	0.142	0.134	0	0.063	0.116	0.188	2.292
Independent variable
CSR	15,047	26.00	17.18	−18.45	16.99	22.55	28.54	90.87
Control variables
lnsize	15,047	22.21	1.323	15.71	21.29	22.03	22.94	28.51
Age	15,047	2.121	0.823	0	1.609	2.197	2.833	3.367
Lev	15,047	0.431	0.240	0.007	0.257	0.419	0.588	8.256
MB	15,047	0.628	0.241	0.006	0.442	0.634	0.812	1.463
Cashflow	15,047	0.043	0.085	−4.270	0.005	0.042	0.083	0.850
TOP1 (%)	15,047	35.86	15.03	2.197	24.14	34.15	45.88	89.99
Turnover (%)	15,047	581.9	508.6	2.170	246.2	426.7	745.2	4912
TOP10 (%)	15,047	58.87	15.29	3.588	48.06	59.82	70.41	101.2
Sep	15,047	4.760	7.685	−0.012	0	0	7.871	49.40
Power	15,047	1.745	0.436	1	1	2	2	2
Committee	15,047	3.940	0.472	1	4	4	4	8

, and the definitions of the variables can be found in Appendix B.

3.3. Methodology

We set a benchmark regression model to establish whether CSR can affect IRM level as follows (establish H1):

$$\mathrm{I}\mathrm{R}\mathrm{I}{\mathrm{I}}_{\mathrm{i},\mathrm{t}+1}={\beta }_0+{\beta }_1\mathrm{C}\mathrm{S}{\mathrm{R}}_{\mathrm{i},\mathrm{t}}+{\displaystyle \sum \delta \mathrm{C}{\mathrm{V}}_{\mathrm{i},\mathrm{t}}}+\mathrm{F}\mathrm{i}\mathrm{r}{\mathrm{m}}_{\mathrm{i}}+\mathrm{Y}\mathrm{e}\mathrm{a}{\mathrm{r}}_{\mathrm{t}}+{\epsilon }_{\mathrm{i},\mathrm{t}}$$

(1)

where the dependent variable ${\mathrm{I}\mathrm{R}\mathrm{I}\mathrm{I}}_{\mathrm{i},\mathrm{t}=1}$ stands for investor relations management level of firm i in year t + 1, and the independent variable ${\mathrm{C}\mathrm{S}\mathrm{R}}_{\mathrm{i},\mathrm{t}}$ expresses the corporate social responsibility rating score of firm i in year t. $\sum {\mathrm{C}\mathrm{V}}_{\mathrm{i},\mathrm{t}}$ represents the control variables. ${\mathrm{F}\mathrm{i}\mathrm{r}\mathrm{m}}_{\mathrm{i}}$ indicates the individual firm-fixed effects, ${Year}_t$ denotes the year-fixed effects, and ${\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$ shows the error term.

To verify H2, we added an interaction term between the operating performance and CSR score into the Model (1). The new model is as follows:

$$\mathrm{I}\mathrm{R}\mathrm{I}{\mathrm{I}}_{\mathrm{i},\mathrm{t}+1}={\beta }_0+{\beta }_1\mathrm{C}\mathrm{S}{\mathrm{R}}_{\mathrm{i},\mathrm{t}}+{\beta }_2\mathrm{C}\mathrm{S}{\mathrm{R}}_{\mathrm{i},\mathrm{t}}\times \mathrm{d}\mathrm{u}\mathrm{m}\_\mathrm{R}\mathrm{O}{\mathrm{A}}_{\mathrm{i},\mathrm{t}}+{\beta }_3\mathrm{d}\mathrm{u}\mathrm{m}\_\mathrm{R}\mathrm{O}{\mathrm{A}}_{\mathrm{i},\mathrm{t}}+{\displaystyle \sum \delta \mathrm{C}{\mathrm{V}}_{\mathrm{i},\mathrm{t}}}+\mathrm{F}\mathrm{i}\mathrm{r}{\mathrm{m}}_{\mathrm{i}}+\mathrm{Y}\mathrm{e}\mathrm{a}{\mathrm{r}}_{\mathrm{t}}+{\epsilon }_{\mathrm{i},\mathrm{t}}$$

(2)

where the variable ROA represents the return on assets of enterprises, and $\mathrm{d}\mathrm{u}\mathrm{m}\_{\mathrm{R}\mathrm{O}\mathrm{A}}_{\mathrm{i},\mathrm{t}}$ is the dummy variable used to measure the business performance of enterprises. When the ROA of a firm is lower than the average ROA in the industry-year dimension, it is set to 1; otherwise, it is set to 0. β₂ indicates the coefficients of the interaction (${\mathrm{C}\mathrm{S}\mathrm{R}}_{\mathrm{i},\mathrm{t}}\times \mathrm{d}\mathrm{u}\mathrm{m}\_{\mathrm{R}\mathrm{O}\mathrm{A}}_{\mathrm{i},\mathrm{t}}$). If ${\beta }_2>0$, we can infer that for companies with poor business performance, the higher the CSR score, the more inclined they are to strengthen IRM. Then, H2 can be verified. Otherwise, H2 could not be established.

4. Empirical Results and Discussion

4.1. Baseline Regression Results

Table 2. Baseline regression.

Variables	(1)	(2)	(3)	(4)
	IRII	IRII	IRII	IRII
CSR	0.0511 ***	0.0332 ***	0.0317 ***	0.0175 **
	(8.162)	(5.255)	(3.765)	(2.083)
dum_ROA			−0.0133 ***	−0.0118 ***
			(−4.061)	(−3.582)
dum_ROA × CSR			0.0302 ***	0.0245 **
			(2.999)	(2.451)
lnsize		0.0370 ***		0.0365 ***
		(15.81)		(15.55)
Age		0.00690		0.00805 *
		(1.527)		(1.772)
Lev		−0.00491		−0.00331
		(−0.872)		(−0.585)
MB		−0.0478 ***		−0.0451 ***
		(−6.750)		(−6.292)
Cashflow		−0.0205 *		−0.0217 **
		(−1.923)		(−2.035)
TOP1		0.000141		0.000123
		(0.879)		(0.765)
Turnover		2.41 × 10−6		2.23 × 10−6
		(0.992)		(0.915)
TOP10		−0.000203		−0.000228 *
		(−1.479)		(−1.658)
Sep		−0.000229		−0.000244
		(−0.905)		(−0.963)
Power		0.00237		0.00233
		(0.831)		(0.816)
Committee		0.00126		0.00109
		(0.398)		(0.345)
Constant	0.311 ***	−0.491 ***	0.319 ***	−0.473 ***
	(175.8)	(−9.933)	(116.9)	(−9.524)
Observations	15,047	15,047	15,047	15,047
Within R2	0.005	0.029	0.007	0.030
Firm FE	NO	YES	NO	YES
Year FE	NO	YES	NO	YES

Note: Inside the parenthesis, “()” is robust t-statistics. Within R2 is the goodness of fit within the group. ***, ** and * represent significance at the 1%, 5% and 10% levels.

shows the results of baseline regression. Columns (1) and (2) are the results of verifying hypothesis 1, and columns (3) and (4) are the results of verifying hypothesis 2. Columns (1)–(2) in Table 2 report the regression results of Model (1), presenting the regression findings regarding the influence of CSR on the extent of investment relations management. The estimated coefficients of CSR, both excluding and including other control variables, are 0.0511 and 0.0332, respectively. Both coefficients exhibit statistical significance at a minimum level of 1%. These results indicate that when holding other factors constant, an increase in the CSR score corresponds to a higher level of investor management within the company. Therefore, Hypothesis 1 of this paper is preliminarily verified. As mentioned in the theoretical part above, when corporate social responsibility records are good, enterprises will strengthen communication with investors and pay more attention to investor relationship management in order to reduce market investors’ perception of hypocrisy and avoid being misunderstood as “hypocritical” enterprises [19].

Columns (3)–(4) in Table 2 report the regression results of Model (2). The estimated coefficients for the interaction term between CSR and business performance (dum_ROA × CSR) are 0.0302 and 0.0245, respectively, without and with other control variables. Both coefficients demonstrate statistical significance at a minimum level of 5%. These findings indicate that when a company experiences poor business performance, a higher CSR score corresponds to a higher level of IRM. Hypothesis 2 is verified. This confirms the hypothesis mentioned above: CSR may not be the initial goal of enterprises, but they have to participate in CSR in order to cover up bad information as a red herring, which often leads to the decline of business performance for enterprises [52]. When the business performance of enterprises is poor, in order to reduce financial distress, the enterprises are likely to strengthen their interaction and communication with investors in order to prevent misunderstandings [53].

4.2. Robustness Test

4.2.1. High-Dimensional Fixed Effect

High dimensional fixed effects can reflect more subtle differences in variables, making the correlation of variables more precise. Although various control variables have been introduced into the baseline regression, there may still be some unobservable factors that affect both corporate social responsibility scores and investor management levels. Therefore, in order to further alleviate the impact of the missing variable problem on the empirical results of this paper, this paper uses the high-dimensional fixed effect method to conduct robustness test.

To mitigate the impact of endogenous problems on the empirical results of this paper, we controlled for individual-fixed effects and year-fixed effects. Furthermore, based on the high-dimensional fixed effect model, the interaction fixed effect between the listed company’s industry and the year (Year × Industry FE), the interaction fixed effect between the listed company’s province and the year (Year × Province FE), and the interaction fixed effect between the company’s ownership and the year (Year × Owner FE) are controlled to mitigate the error caused by omitted variables by as much as possible. The estimated results are shown in

Table 3. High-dimensional fixed effect regression.

Variables	(1)	(2)	(3)	(4)	(5)	(6)
	IRII	IRII	IRII	IRII	IRII	IRII
CSR	0.0267 ***	0.0254 ***	0.0226 ***	0.00748	0.00604	0.00440
	(4.100)	(3.841)	(3.416)	(0.865)	(0.691)	(0.504)
1dum_ROA				−0.0147 ***	−0.0148 ***	−0.0141 ***
				(−4.314)	(−4.337)	(−4.143)
dum_ROA × CSR				0.0281 ***	0.0283 ***	0.0264 **
				(2.752)	(2.758)	(2.574)
lnsize	0.0325 ***	0.0318 ***	0.0309 ***	0.0317 ***	0.0311 ***	0.0302 ***
	(12.74)	(12.39)	(11.96)	(12.42)	(12.08)	(11.65)
Age	0.00747	0.00774	0.00468	0.00921 *	0.00949 *	0.00616
	(1.543)	(1.581)	(0.895)	(1.892)	(1.928)	(1.172)
Lev	−0.00126	−0.00209	−0.00227	0.00127	0.000478	0.000191
	(−0.217)	(−0.357)	(−0.388)	(0.218)	(0.0812)	(0.0326)
MB	−0.0339 ***	−0.0315 ***	−0.0285 ***	−0.0292 ***	−0.0267 ***	−0.0238 ***
	(−4.442)	(−4.091)	(−3.692)	(−3.758)	(−3.409)	(−3.034)
Cashflow	−0.0221 **	−0.0232 **	−0.0224 **	−0.0241 **	−0.0254 **	−0.0246 **
	(−2.006)	(−2.095)	(−2.030)	(−2.191)	(−2.286)	(−2.219)
TOP1	0.000251	0.000210	0.000220	0.000228	0.000187	0.000198
	(1.515)	(1.266)	(1.324)	(1.379)	(1.128)	(1.196)
Turnover	1.72 × 10−6	1.70 × 10−6	9.21 × 10−7	1.45 × 10−6	1.39 × 10−6	6.11 × 10−7
	(0.680)	(0.666)	(0.360)	(0.572)	(0.545)	(0.238)
TOP10	−2.18 × 10−5	−2.79 × 10−6	4.02 × 10−5	−5.57 × 10−5	−4.00 × 10−5	5.49 × 10−6
	(−0.152)	(−0.0194)	(0.278)	(−0.389)	(−0.277)	(0.0380)
Sep	−0.000214	−0.000270	−0.000253	−0.000231	−0.000288	−0.000268
	(−0.823)	(−1.034)	(−0.970)	(−0.890)	(−1.104)	(−1.030)
Power	0.00181	0.00202	0.00166	0.00178	0.00198	0.00162
	(0.620)	(0.688)	(0.567)	(0.611)	(0.675)	(0.554)
Committee	−0.000123	−0.000252	0.000549	−0.000366	−0.000504	0.000303
	(−0.0381)	(−0.0777)	(0.170)	(−0.114)	(−0.156)	(0.0937)
Constant	−0.408 ***	−0.394 ***	−0.373 ***	−0.385 ***	−0.371 ***	−0.351 ***
	(−7.616)	(−7.288)	(−6.821)	(−7.157)	(−6.840)	(−6.378)
Observations	15,008	15,006	15,006	15,008	15,006	15,006
Adjusted R2	0.569	0.573	0.575	0.570	0.573	0.576
Firm FE	YES	YES	YES	YES	YES	YES
Year × Industry FE	YES	YES	YES	YES	YES	YES
Year × Province FE	NO	YES	YES	NO	YES	YES
Year × Owner FE	NO	NO	YES	NO	NO	YES

Note: Robust t-statistics are reported in parentheses. ***, ** and * denote significance at the 1%, 5% and 10% levels, respectively.

. Table 3 shows the empirical results of the high-dimensional fixed effect model (one of the robustness tests). Columns (1)–(3) are the results of testing hypothesis 1, and columns (4)–(6) are the results of testing hypothesis 2. Columns (1)–(3) of Table 3 indicate that the estimated coefficients of CSR remain significantly positive, confirming the validity of hypothesis 1 even after controlling for more rigorous fixed effects. Additionally, columns (4)–(6) of Table 3 reveal that the estimation coefficients of the interaction term (dum_ROA × CSR) are also significantly positive, providing further support for hypothesis 2.

4.2.2. Endogeneity Issue: The Instrumental Variable Estimation

The main function of instrumental variables is to solve the endogenous problem in the model. When the explanatory variables in the model are related to the error term, traditional regression analysis may not be able to accurately estimate the causal relationship between the variables. Instrumental variables help identify and estimate true causality by introducing exogenous variables that are related to explanatory variables but not to the error term. Considering that corporate social responsibility and investor relations management may have the endogenous problem of mutual causation, this paper introduces the instrumental variable method for further analysis.

To further address the influence of endogenous issues on the regression results, we employ the instrumental variable method. Specifically, the relevant literature on corporate finance provides numerous discussions on the relationship between religious factors, a typical informal institutional arrangement, and CSR, offering ideas for this study.

Business ethics exerts normative and restrictive effects on enterprise behavior, encompassing perspectives such as the monetary view, contract view, integrity view and social responsibility view. It provides a normative framework and code of conduct. Religious culture, which shapes value concepts, is closely associated with regulating the ethical behavior of enterprises. The impact of religion on CSR has been extensively discussed in the literature [55]. Religion influences business practices through ethical practices, including the fulfillment of CSR [56] Brammer et al. [57] directly investigated the impact of religious belief on CSR, and found that individuals with religious beliefs exhibit a significantly more positive attitude toward CSR compared to those without religious beliefs. However, Griffin and Sun [58] discovered a positive correlation between the number of denominations and the disclosure of social responsibility information; they also found a significant positive correlation between the intensity of religious belief, the number of sectarian affiliations, and the disclosure of environmental information. It is evident that existing research generally supports the notion that religious factors significantly promote the fulfillment of CSR.

Based on the above ideas, this paper selects the number of Taoist and other temples within a 200 km radius of listed companies’ locations as an instrumental variable of CSR. In terms of relevance conditions, the existing literature on religion and CSR provides solid theoretical support for this choice. In terms of exogeneity, the number of Taoist and other temples is often affected by the local long-term history and culture, with less direct influence of listed companies. Thus, it can be regarded as satisfying the exogenous conditions. Since the number of Taoist and other temples at the location of the listed company is a time-invariant variable, the instrumental variable used in the actual estimation is the interaction term between the number of Taoist and other temples and the average CSR value of other enterprises in the same industry and year. In addition, to ensure the robustness of the instrumental variable, we also provide the number of Taoist and othr temples within 150 km and 250 km as robustness indicators.

The estimated results are presented in

Table 4. Instrumental variable regression.

Variables	(1)	(2)	(3)	(4)	(5)	(6)
	The First-Stage Regression			The Second-Stage Regression
CSR				1.173 ***	1.053 ***	1.105 ***
				(2.985)	(2.702)	(3.081)
IV1	0.000294 ***
	(3.410)
IV2		0.000449 ***
		(3.181)
IV3			0.000211 ***
			(3.577)
lnsize	0.0550 ***	0.0548 ***	0.0548 ***	−0.0262	−0.0193	−0.0223
	(16.66)	(16.62)	(16.63)	(−1.176)	(−0.879)	(−1.100)
Age	0.0132 **	0.0133 **	0.0133 **	−0.00866	−0.00701	−0.00766
	(2.051)	(2.070)	(2.059)	(−0.856)	(−0.736)	(−0.800)
Lev	−0.0702 ***	−0.0701 ***	−0.0701 ***	0.0751 **	0.0667 **	0.0704 ***
	(−8.779)	(−8.782)	(−8.767)	(2.540)	(2.298)	(2.594)
MB	−0.103 ***	−0.103 ***	−0.103 ***	0.0704	0.0575	0.0639
	(−10.29)	(−10.28)	(−10.31)	(1.634)	(1.360)	(1.618)
Cashflow	0.0737 ***	0.0765 ***	0.0761 ***	−0.106 ***	−0.0985 ***	−0.102 ***
	(4.818)	(5.024)	(4.998)	(−2.988)	(−2.789)	(−3.032)
TOP1	0.000378 *	0.000351	0.000348	−0.000307	−0.000227	−0.000252
	(1.654)	(1.539)	(1.526)	(−0.891)	(−0.713)	(−0.787)
Turnover	−8.17 × 10−6 **	−7.98 × 10−6 **	−8.12 × 10−6 **	1.15 × 10−5 **	1.04 × 10−5 **	1.09 × 10−5 **
	(−2.357)	(−2.304)	(−2.343)	(2.048)	(1.974)	(2.068)
TOP10	0.000365 *	0.000386 **	0.000384 **	−0.000616 **	−0.000608 **	−0.000616 **
	(1.859)	(1.972)	(1.963)	(−2.051)	(−2.124)	(−2.144)
Sep	0.000126	0.000139	0.000133	−0.000418	−0.000404	−0.000410
	(0.348)	(0.385)	(0.368)	(−0.856)	(−0.893)	(−0.877)
Power	0.00582	0.00586	0.00581	−0.00441	−0.00379	−0.00410
	(1.432)	(1.442)	(1.429)	(−0.744)	(−0.684)	(−0.728)
Committee	0.00406	0.00403	0.00408	−0.00339	−0.00290	−0.00314
	(0.901)	(0.896)	(0.906)	(−0.545)	(−0.503)	(−0.528)
Observations	14,971	14,986	15,003	14,971	14,986	15,003
Adjusted R2	0.521	0.520	0.521	−3.276	−2.650	−2.916
Firm FE	YES	YES	YES	YES	YES	YES
Year FE	YES	YES	YES	YES	YES	YES
F statistic	11.63	10.12	12.80

Note: Robust t-statistics are reported in parentheses. ***, ** and * denote significance at the 1%, 5% and 10% levels, respectively.

, where columns (1)–(3) represent the first-stage regression of instrumental variables, while columns (4)–(6) represent the second-stage regression of instrumental variables (instrumental variables regression). The F-values of the first-stage regression, namely, 11.63, 10.12, and 12.80, are all above the critical value of 10. The estimated coefficients of the instrumental variables are all significantly positive, indicating that a greater presence of religious institutions near a listed company corresponds to higher CSR scores among other enterprises in the same industry, and higher CSR for the listed company itself. Additionally, the regression coefficient in the second stage is significantly positive, suggesting that CSR ratings have a positive impact on IRM. This finding further supports the core regression results of this study.

4.2.3. Retesting Based on the Difference-in-Differences Method (DID)

To further verify the causality of the empirical results, this study employs the DID method and categorizes the sample into two groups: those with a CSR score higher than the 75th percentile (dCSR_p75) are labeled as 1, while the rest are labeled as 0. Similarly, the sample with an enterprise asset return lower than the 25th percentile (dROA_p25) is labeled as 1, and the rest are labeled as 0. The focus of this paper is the interaction coefficient between dCSR_p75 and dROA_p25. The inclusion of dummy variables for the 75th percentile of CSR and the 25th percentile of ROA aims to create the treatment and control groups required for the DID analysis. Moreover, given that CSR scores and business performance are continuous variables, it is generally assumed that samples above the 75th percentile of CSR represent enterprises that excel in fulfilling their social responsibilities, while those below the 75th percentile can be considered relatively weak in this aspect. Particularly, the samples above the 75th percentile of CSR and below the 25th percentile of CSR provide a clear contrast between “good” and “poor” social responsibility performance. Similarly, the samples below the 25th percentile of ROA and above the 75th percentile of ROA serve as a sharp contrast between “poor” and “good” business performance. By conducting DID tests on samples falling below the 25th percentile of ROA and above the 75th percentile of CSR, the robustness of the empirical results can be further underscored. Additionally, this approach helps mitigate the measurement error associated with CSR and business performance to a certain extent.

Based on the estimated results in

Table 5. Difference-in-Difference regression.

Variables	(1)	(2)
	Total Sample	Below the 25th Percentile and above the 75th Percentile
dCSR_p75	0.00225	−0.0161
	(0.848)	(−0.967)
dROA_p25	−0.0168 ***	−0.0402 ***
	(−6.944)	(−2.597)
dCSR_p75 × dROA_p25	0.00963 *	0.0333 *
	(1.833)	(1.936)
Control variables	YES	YES
Constant	−0.477 ***	−0.474 ***
	(−9.618)	(−4.305)
Observations	15,047	4079
Within R2	0.031	0.032
Firm FE	YES	YES
Year FE	YES	YES

Note: Robust t-statistics are reported in parentheses. ***, and * denote significance at the 1% and 10% levels, respectively.

, the coefficients of dCSR_p75 × dROA_p25 are both significantly positive, indicating that enterprises with strong social responsibility performance but poor business performance attach more importance to IRM. Notably, the coefficient in column (2) is significantly larger than that in column (1), indicating that the net effect on IRM becomes more pronounced as the gap between CSR and business performance widens. Considering that enterprises with good social responsibility performance but poor business performance are susceptible to misinterpretation, the empirical findings of this study demonstrate that such enterprises tend to prioritize IRM, further reinforcing the main conclusion of this research.

4.2.4. Modification of the Measurement of the Independent Variables

Considering that the disclosure of information related to CSR by listed companies is predominantly voluntary, such information is often presented in two forms: standalone social responsibility reports and the inclusion of social responsibility information in annual reports. Accordingly, this study conducts grouped regressions for different types of disclosure, as presented in columns (1)–(2) of

Table 6. Robustness test based on replacement core explanatory variables.

Variables	(1)	(2)	(3)	(4)
	Annual Report	Independent CSR Report	WIND	Runling (RKS)
CSR	0.0675 ***	0.0286 ***
	(4.237)	(2.735)
CSR (WIND)			0.00350 *
			(1.793)
CSR (Runling)				0.00116 ***
				(4.211)
Control variables	YES	YES	YES	YES
Constant	−0.461 ***	−0.565 ***	−0.251 ***	0.0349
	(−7.975)	(−4.014)	(−3.053)	(0.546)
Observations	10,610	4254	2256	1519
Within R2	0.029	0.017	0.062	0.022
Firm FE	YES	YES	NO	NO
Year FE	YES	YES	NO	NO

Note: Robust t-statistics are reported in parentheses. ***, and * denote significance at the 1% and 10% levels, respectively.

. Furthermore, to ensure the robustness of CSR measurement, the regression results are also provided after replacing the scores with Runling’s social responsibility rating and WIND’s ESG rating [31]. It should be noted that due to a significant number of missing values for these two CSR scores when matched with the selected samples in this study, only OLS regression results are presented, as shown in columns (3)–(4) of Table 6. The estimated coefficient of CSR remains significantly positive, thus supporting the fundamental conclusion of this research.

5. Further Discussion: Mechanism Analysis

Part 5 is an extension of the basic results, seeking mainly to further analyze the specific path by which corporate social responsibility affects the investor relations management level.

5.1. Mechanism Analysis Based on IRM Level Index

IRII is an index representing the level of investor relations interaction and comprises five primary indicators: communication guarantee (IRII_X1), online communication (IRII_X2), telephone communication (IRII_X3), on-site communication (IRII_X4), and communication feedback (IRII_X5). To investigate the specific pathways through which CSR affects the investor relations interaction index, this study examines each of the five IRII indicators as dependent variables.

Table 7. Regression results based on the sub-index of IRII.

Variables	(1)	(2)	(3)	(4)	(5)
	IRII_x1	IRII_x2	IRII_x3	IRII_x4	IRII_x5
CSR	−0.0184	0.0295 ***	0.0148	0.0539 ***	−0.00356
	(−1.372)	(3.352)	(1.128)	(5.319)	(−0.220)
lnsize	0.0232 ***	0.0345 ***	0.0285 ***	0.0574 ***	0.0268 ***
	(4.655)	(10.57)	(5.853)	(15.31)	(4.486)
Age	−0.00330	0.0483 ***	0.00409	0.0219 ***	−0.0630 ***
	(−0.343)	(7.662)	(0.435)	(3.019)	(−5.450)
Lev	−0.00497	0.00117	0.00122	0.000326	−0.0183
	(−0.415)	(0.149)	(0.104)	(0.0361)	(−1.273)
MB	−0.0554 ***	−0.0221 **	−0.0818 ***	−0.0565 ***	−0.0245
	(−3.678)	(−2.239)	(−5.567)	(−4.985)	(−1.353)
Cashflow	0.0145	−0.0535 ***	−0.0377 *	−0.0347 **	−0.00520
	(0.640)	(−3.601)	(−1.702)	(−2.029)	(−0.191)
TOP1	0.000104	−0.000313	0.000489	−3.24 × 10−5	0.000378
	(0.306)	(−1.404)	(1.470)	(−0.126)	(0.925)
Turnover	−8.40 × 10−6	8.52 × 10−7	8.32 × 10−6 *	1.05 × 10−5 ***	8.49 × 10−7
	(−1.623)	(0.251)	(1.647)	(2.685)	(0.137)
TOP10	0.000846 ***	−0.000377 **	−5.18 × 10−5	−0.00105 ***	0.000402
	(2.890)	(−1.965)	(−0.181)	(−4.747)	(1.145)
Sep	−0.000712	−0.000563	0.000276	−0.000126	0.000322
	(−1.320)	(−1.593)	(0.523)	(−0.310)	(0.497)
Power	0.00198	0.00497	−0.00225	0.00226	−0.00430
	(0.325)	(1.250)	(−0.380)	(0.493)	(−0.590)
Committee	0.0140 **	−0.00524	0.00364	−0.00522	−0.00298
	(2.080)	(−1.191)	(0.554)	(−1.031)	(−0.369)
Constant	−0.159	−0.218 ***	−0.347 ***	−1.073 ***	−0.286 **
	(−1.515)	(−3.169)	(−3.379)	(−13.56)	(−2.263)
Observations	15,047	15,047	15,047	15,047	15,047
Adjusted R2	0.572	0.527	0.610	0.492	0.368
Firm FE	YES	YES	YES	YES	YES
Year FE	YES	YES	YES	YES	YES

Note: Robust t-statistics are reported in parentheses. ***, ** and * denote significance at the 1%, 5% and 10% levels, respectively.

reports the results of mechanism analysis based on IRM. Columns (1), (3), and (5) present the testing results for (IRII_X1), (IRII_X3), and (IRII_X5), respectively. The estimated coefficient of CSR is not significant, indicating that CSR does not significantly promote the communication guarantee, telephone communication and communication feedback of enterprise investment relationship management. Columns (2) and (4), which pertain to (IRII_X2) and (IRII_X4) as dependent variables, respectively, reveal significantly positive estimated coefficients for CSR at a significance level of 1%. This suggests that CSR can facilitate online communication in IRM and on-site communication during interaction processes.

In summary, CSR mainly improves the level of IRM by strengthening online communication and on-site communication. Communication channels play a crucial role in information transmission [37]. While IRM can bring benefits to the controlling shareholders, it also incurs certain costs. Therefore, controlling shareholders need to carefully evaluate costs and benefits when choosing communication strategies. Enhancing the communication mechanism involves establishing a professional investor relations team, which serves as the organizational foundation for efficient communication.

5.2. Analysis of Moderating Effects Based on Financial Distress, Financing Constraints, and Information Disclosure Quality

Based on the principal-agent theory, the disclosure of social responsibility information may serve as a tool for window dressing, allowing managers to camouflage their lack of profitability or other negative news [48]. Theoretically, the disclosure of social responsibility information is likely to serve as a “disguise” for financial difficulties, severe financing constraints, or poor information disclosure. By testing hypothesis 2, we find that when a company has a high social responsibility rating but poor business performance, it may enhance IRM to avoid misinterpretation. We can infer that when an enterprise with a high social responsibility score experiences financial distress, tight financing constraints, or poor information disclosure quality, it may also prioritize interactions with investors so as to convey positive information and prevent being perceived as engaging in “hypocritical behavior” to conceal its poor business performance.

Therefore, this study introduces interaction terms between financial distress, financing constraints, information disclosure quality, and CSR, in order to examine this mechanism. Financial distress is measured using the AltmanZ index, with a value of 1 (dum_Z) when the AltmanZ index exceeds the industry’s annual median, and 0 otherwise. Financing constraints are measured by the KZ index, with a value of 1 (dum_KZ) when the KZ index is higher than the industry’s annual median, and 0 otherwise. Accounting information disclosure quality is categorized as “qualified” or “unqualified”, represented by 1 (dum_AQI), while “excellent” or “good” is represented by 0. Columns (1)–(3) in

Table 8. Regression results of moderating effects based on financial distress, financing constraints, and information disclosure quality.

Variables	(1)	(2)	(3)
	IRII	IRII	IRII
CSR	0.0131	0.0198 **	0.00166
	(1.267)	(2.275)	(0.185)
dum_Z	−0.00611
	(−1.629)
dum_Z × CSR	0.0271 **
	(2.469)
dum_KZ		−0.0102 ***
		(−3.067)
dum_KZ × CSR		0.0206 **
		(2.066)
dum_AQI			−0.0218 ***
			(−4.362)
dum_AQI × CSR			0.0418 **
			(2.043)
lnsize	0.0373 ***	0.0371 ***	0.0451 ***
	(15.78)	(15.82)	(15.73)
Age	0.00647	0.00629	−0.000837
	(1.429)	(1.390)	(−0.155)
Lev	−0.00540	−0.00283	−0.0145 **
	(−0.942)	(−0.499)	(−2.037)
MB	−0.0490 ***	−0.0490 ***	−0.0612 ***
	(−6.672)	(−6.916)	(−7.105)
Cashflow	−0.0204 *	−0.0275 **	−0.0143
	(−1.915)	(−2.481)	(−0.934)
TOP1	0.000138	0.000152	0.000505 **
	(0.861)	(0.948)	(2.553)
Turnover	2.41 × 10−6	2.65 × 10−6	−2.38 × 10−6
	(0.992)	(1.088)	(−0.851)
TOP10	−0.000202	−0.000231 *	2.16 × 10−5
	(−1.470)	(−1.676)	(0.127)
Sep	−0.000240	−0.000244	−0.000320
	(−0.945)	(−0.962)	(−1.008)
Power	0.00243	0.00240	3.77 × 10−5
	(0.850)	(0.841)	(0.0115)
Committee	0.00127	0.00119	0.00139
	(0.402)	(0.377)	(0.352)
Constant	−0.490 ***	−0.482 ***	−0.586 ***
	(−9.838)	(−9.737)	(−9.730)
Observations	15,047	15,047	9241
Adjusted R2	0.563	0.563	0.443
Firm FE	YES	YES	YES
Year FE	YES	YES	YES

Note: Robust t-statistics are reported in parentheses. ***, ** and * denote significance at the 1%, 5% and 10% levels, respectively.

present the interaction results for financial distress, financing constraints, and accounting information disclosure quality. The estimation results show that the estimated coefficient for the interaction term dum_Z × CSR is 0.0271, for dum_KZ × CSR it is 0.0206, and for dum_AQI × CSR it is 0.0418, all of which are statistically significant at the 5% level.

In summary, when an enterprise has a high social responsibility score and experiences severe financial distress, tight financing constraints, or poor information disclosure quality, it tends to engage in positive interaction with investors in the capital market, aligning with our earlier inferences. As CSR can potentially serve as a “hypocritical” cover-up for poor business conditions, companies actively interact with investors to prevent misinterpretations and convey positive information. This finding is consistent with the logic of hypothesis 2 presented in this study, indirectly confirming the analytical reasoning behind hypothesis 2.

6. The Extended Analysis Based on Stock Price Crash Risk

In relation to fulfilling social responsibilities, the performance of the capital market, particularly the risk of a stock crash, has become a pivotal topic for scholars. There are two contrasting views in the existing literature on how CSR influences stock price crash risk. One perspective posits that disclosing CSR information can also generate a “communication effect” by reducing information asymmetry and improving the capital market performance of enterprises, thereby mitigating the risk of a stock crash [30,31]. Conversely, another perspective maintains that active CSR disclosure or performance serves as a “self-serving tool” for management to divert the public’s attention and conceal negative information. When the negative information accumulates to a critical point, it can lead to an instantaneous stock crash, further exacerbating the risk [19,20]. So how exactly does CSR affect the stock price crash risk? Our research points out that investor management plays an important role.

The risk of stock price collapse stems primarily from information asymmetry. Theoretically, as a strategic communication approach of enterprises, IRM can effectively mitigate the level of internal and external information asymmetry by engaging in two-way communication with capital market investors. By increasing external investors’ trust and favor towards managers, it can enhance investors’ confidence and reduce their overreaction to corporate risks, thus significantly reducing the stock price crash risk for listed companies.

Drawing from this perspective and integrating the aforementioned empirical findings in this paper, we can deduce the analytical framework: “corporate social responsibility → investor relations management → stock price crash risk”. In essence, a higher CSR score corresponds to a higher level of IRM, ultimately leading to a reduced stock price crash risk. Conversely, a lower CSR score corresponds to a lower level of IRM, increasing the stock price crash risk. Therefore, IRM serves as a critical mechanism for CSR to influence the stock price crash risk of listed companies. This section will delve into the causal relationship between CSR and stock price crash risk.

We here refer to the specific process of calculating stock price crash risk as set out by existing scholars [20,31]. Firstly, the weekly rate of return $R_{i,t}$ of individual stocks is regressed according to the average weekly return rate $R_{m,t}$ of the stock market to eliminate the impact of the economic cycle and other market factors on the return of individual stocks. The model to be tested is as follows:

$$R_{i,t}={\alpha }_0+{\alpha }_1{R}_{m,t-2}+{\alpha }_2{R}_{m,t-1}+{\alpha }_3{R}_{m,t}+{\alpha }_4{R}_{m,t+1}+{\alpha }_5{R}_{m,t+2}+{\epsilon }_{it}$$

(3)

where $R_{i,t}$ represents the rate of return of individual stock i in week t considering cash dividend reinvestment, $R_{m,t}$ stands for the average rate of return weighted by the circulating market value of all stocks in week t of the stock market, and ${\epsilon }_{it}$ represents the part of the rate of return of individual stock that cannot be explained by the fluctuation of market yield. By calculating $W_{i,t}=Ln\left(1+{\epsilon }_{it}\right)$, we denote $W_{i,t}$ as the specific weekly yield of individual stock i. Since there is a certain degree of non-synchronization between stock market yield and trading behavior, the lead term and lagged term of market yield are added to the model (3) to reduce the potential deviation caused by non-synchronous trading.

Secondly, the measure of stock price crash risk is calculated according to $W_{i,t}$:

$$NCSKE{W}_{i,t}=-{\displaystyle \frac{n{\left(n-1\right)}^{3/2}{\displaystyle \sum W_{i,t}^3}}{\left(n-1\right)\left(n-2\right){\left({\displaystyle \sum W_{i,t}^2}\right)}^{3/2}}}$$

$$DUVO{L}_{i,t}=\ln {\displaystyle \frac{\left(n_u-1\right){\displaystyle \sum _{Down}{W}_{i,t}^3}}{\left(n_d-1\right){\displaystyle \sum _{Up}{W}_{i,t}^2}}}$$

where $NCSKE{W}_{i,t}$ represents the coefficient of negative return skewness, $DUVO{L}_{i,t}$ represents the ratio of earnings’ up and down fluctuation, n represents the number of weeks in which individual stock i is listed for trading each year, and $n_u$ and $n_d$ respectively represent the number of weeks in which the weekly yield of stock i is higher than and lower than the average annual yield of the stock. The higher the value of $NCSKE{W}_{i,t}$ and $DUVO{L}_{i,t}$, the higher the degree of negative deviation of stock prices, and the higher the fluctuation ratio of earnings, the greater the possibility of a crash.

The results of the fixed effects model regression applied in examining the impact of CSR on stock price crash risk are reported in columns (1) and (2) of

Table 9. Regression results based on CSR and stock price crash risk.

Variables	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	Fixed Effects Model		IV1		IV2		IV3
	NCSKEW	DUVOL	NCSKEW	DUVOL	NCSKEW	DUVOL	NCSKEW	DUVOL
CSR	−0.210 ***	−0.122 ***	−6.448 **	−3.284 **	−6.217 **	−3.083 **	−6.384 ***	−3.098 **
	(−3.684)	(−3.395)	(−2.470)	(−2.195)	(−2.261)	(−1.969)	(−2.578)	(−2.221)
lnsize	0.196 ***	0.121 ***	0.543 ***	0.298 ***	0.527 ***	0.285 ***	0.538 ***	0.286 ***
	(9.264)	(9.098)	(3.672)	(3.515)	(3.410)	(3.234)	(3.837)	(3.631)
Age	−0.140 ***	−0.111 ***	−0.0548	−0.0686 *	−0.0578	−0.0704 *	−0.0564	−0.0713 *
	(−3.429)	(−4.296)	(−0.816)	(−1.782)	(−0.860)	(−1.839)	(−0.853)	(−1.915)
Lev	0.0337	0.0158	−0.407 **	−0.207 *	−0.390 *	−0.193 *	−0.402 **	−0.194 *
	(0.663)	(0.492)	(−2.074)	(−1.844)	(−1.903)	(−1.656)	(−2.145)	(−1.839)
MB	−1.100 ***	−0.726 ***	−1.747 ***	−1.057 ***	−1.716 ***	−1.030 ***	−1.740 ***	−1.037 ***
	(−17.21)	(−18.02)	(−6.112)	(−6.451)	(−5.755)	(−6.066)	(−6.378)	(−6.748)
Cashflow	−0.0103	−0.00284	0.444 *	0.234 *	0.447 *	0.229	0.458 **	0.231 *
	(−0.107)	(−0.0467)	(1.883)	(1.732)	(1.793)	(1.610)	(1.971)	(1.761)
TOP1	−0.00224	−0.00153 *	7.58 × 10−5	−0.000384	9.65 × 10−6	−0.000431	7.22 × 10−5	−0.000402
	(−1.551)	(−1.673)	(0.0331)	(−0.293)	(0.00431)	(−0.338)	(0.0326)	(−0.322)
Turnover	−7.82 × 10−6	−1.03 × 10−5	−5.50 × 10−5	−3.49 × 10−5	−5.38 × 10−5	−3.35 × 10−5	−5.57 × 10−5	−3.39 × 10−5 *
	(−0.356)	(−0.744)	(−1.482)	(−1.637)	(−1.451)	(−1.587)	(−1.530)	(−1.652)
TOP10	0.00603 ***	0.00324 ***	0.00855 ***	0.00452 ***	0.00844 ***	0.00443 ***	0.00844 ***	0.00437 ***
	(4.856)	(4.135)	(4.284)	(3.957)	(4.175)	(3.849)	(4.251)	(3.906)
Sep	0.00155	0.00251 *	0.00245	0.00297	0.00247	0.00297	0.00249	0.00298
	(0.677)	(1.735)	(0.754)	(1.597)	(0.772)	(1.634)	(0.772)	(1.636)
Power	−0.00351	0.00442	0.0307	0.0224	0.0321	0.0224	0.0331	0.0224
	(−0.136)	(0.272)	(0.779)	(0.993)	(0.821)	(1.004)	(0.851)	(1.025)
Committee	−0.0229	−0.00502	0.00305	0.00835	0.00164	0.00721	0.00244	0.00739
	(−0.801)	(−0.279)	(0.0738)	(0.352)	(0.0402)	(0.311)	(0.0594)	(0.320)
Constant	−3.930 ***	−2.405 ***
	(−8.804)	(−8.543)
Observations	15,047	15,047	14,971	14,971	14,986	14,986	15,003	15,003
Adjusted R2	0.110	0.112	−1.325	−0.917	−1.237	−0.824	−1.298	−0.831
Firm FE	YES	YES	YES	YES	YES	YES	YES	YES
Year FE	YES	YES	YES	YES	YES	YES	YES	YES

Note: Robust t-statistics are reported in parentheses. ***, ** and * denote significance at the 1%, 5% and 10% levels, respectively.

. The estimated coefficients of CSR are −0.210 and −0.122, both of which are significant at the 1% level, indicating that under identical conditions, superior CSR performance can effectively mitigate the risk of stock price collapse. Columns (3)–(8) of

Table 10. Regression results based on CSR and risk of stock price collapse (high-dimensional fixed effects).

Variables	(1)	(2)	(3)	(4)	(5)	(6)
	F.NCSKEW1	F.NCSKEW1	F.NCSKEW1	F.DUVOL1	F.DUVOL1	F.DUVOL1
CSR	−0.142 **	−0.157 ***	−0.156 ***	−0.0751 **	−0.0850 **	−0.0860 **
	(−2.420)	(−2.618)	(−2.602)	(−2.031)	(−2.253)	(−2.270)
lnsize	0.246 ***	0.254 ***	0.259 ***	0.151 ***	0.154 ***	0.155 ***
	(10.73)	(10.93)	(11.06)	(10.44)	(10.50)	(10.47)
Age	−0.123 ***	−0.114 ***	−0.105 **	−0.104 ***	−0.100 ***	−0.102 ***
	(−2.823)	(−2.584)	(−2.218)	(−3.779)	(−3.593)	(−3.405)
Lev	0.0293	0.0172	0.0179	0.0133	0.00879	0.00770
	(0.558)	(0.324)	(0.337)	(0.402)	(0.263)	(0.230)
MB	−1.109 ***	−1.129 ***	−1.140 ***	−0.739 ***	−0.746 ***	−0.747 ***
	(−16.15)	(−16.22)	(−16.32)	(−17.04)	(−16.99)	(−16.95)
Cashflow	−0.0105	0.0246	0.0217	−0.000140	0.0223	0.0205
	(−0.106)	(0.246)	(0.217)	(−0.00224)	(0.353)	(0.324)
TOP1	−0.00186	−0.00159	−0.00163	−0.00111	−0.000894	−0.000909
	(−1.253)	(−1.062)	(−1.086)	(−1.180)	(−0.943)	(−0.958)
Turnover	−2.54 × 10−5	−2.73 × 10−5	−2.57 × 10−5	−1.79 × 10−5	−1.88 × 10−5	−1.84 × 10−5
	(−1.116)	(−1.184)	(−1.105)	(−1.246)	(−1.295)	(−1.255)
TOP10	0.00470 ***	0.00492 ***	0.00475 ***	0.00217 ***	0.00218 ***	0.00216 ***
	(3.653)	(3.777)	(3.625)	(2.670)	(2.656)	(2.612)
Sep	0.000980	0.000899	0.000857	0.00229	0.00215	0.00218
	(0.420)	(0.381)	(0.363)	(1.554)	(1.445)	(1.462)
Power	−0.00420	−0.00370	−0.00377	0.00320	0.00409	0.00393
	(−0.160)	(−0.140)	(−0.142)	(0.193)	(0.244)	(0.235)
Committee	−0.0358	−0.0297	−0.0309	−0.00882	−0.00462	−0.00505
	(−1.237)	(−1.016)	(−1.054)	(−0.482)	(−0.250)	(−0.274)
Constant	−4.962 ***	−5.176 ***	−5.289 ***	−3.015 ***	−3.099 ***	−3.110 ***
	(−10.29)	(−10.60)	(−10.67)	(−9.908)	(−10.06)	(−9.942)
Observations	15,008	15,006	15,006	15,008	15,006	15,006
Adjusted R2	0.127	0.127	0.127	0.128	0.129	0.130
Firm FE	YES	YES	YES	YES	YES	YES
Year × Industry FE	YES	YES	YES	YES	YES	YES
Year × province FE	NO	YES	YES	NO	YES	YES
Year × Owner FE	NO	NO	YES	NO	NO	YES

Note: Robust t-statistics are reported in parentheses. ***, ** and denote significance at the 1%, 5% and 10% levels, respectively.

present the regression results obtained under three different instrumental variable scenarios, revealing that the estimated coefficients of CSR are consistently negative and significant at least at the 5% level. Additionally, this study provides the regression results under the high-dimensional fixed effect model (see Table 10) and various stock price crash risk measurement indicators (

Table 11. Robust regression results based on stock price crash risk.

Variables	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
	NCSKEW2	DUVOL2	NCSKEW3	DUVOL3	NCSKEW4	DUVOL4	NCSKEW5	DUVOL5	NCSKEW6	DUVOL6
CSR1	−0.181 ***	−0.120 ***	−0.175 ***	−0.117 ***	−0.252 ***	−0.163 ***	−0.230 ***	−0.159 ***	−0.244 ***	−0.166 ***
	(−3.263)	(−3.354)	(−3.161)	(−3.259)	(−4.431)	(−4.537)	(−4.257)	(−4.536)	(−4.465)	(−4.721)
lnsize	0.192 ***	0.114 ***	0.198 ***	0.117 ***	0.183 ***	0.105 ***	0.156 ***	0.0838 ***	0.160 ***	0.0890 ***
	(9.344)	(8.569)	(9.628)	(8.836)	(8.720)	(7.906)	(7.817)	(6.442)	(7.903)	(6.818)
Age	−0.164 ***	−0.117 ***	−0.142 ***	−0.0961 ***	−0.198 ***	−0.164 ***	−0.220 ***	−0.168 ***	−0.212 ***	−0.160 ***
	(−4.137)	(−4.561)	(−3.566)	(−3.757)	(−4.866)	(−6.393)	(−5.688)	(−6.669)	(−5.407)	(−6.357)
Lev	0.0461	0.0155	0.0506	0.0243	0.0140	−0.00529	0.0220	−0.00397	0.0236	−0.00221
	(0.934)	(0.487)	(1.024)	(0.764)	(0.277)	(−0.165)	(0.456)	(−0.127)	(0.483)	(−0.0702)
MB	−0.998 ***	−0.669 ***	−1.013 ***	−0.670 ***	−1.080 ***	−0.710 ***	−0.957 ***	−0.619 ***	−0.956 ***	−0.614 ***
	(−16.09)	(−16.68)	(−16.31)	(−16.74)	(−16.98)	(−17.66)	(−15.82)	(−15.73)	(−15.62)	(−15.57)
Cashflow	−0.0351	−0.0426	−0.0417	−0.0416	−0.0289	−0.0140	−0.0799	−0.0878	−0.0744	−0.0776
	(−0.375)	(−0.705)	(−0.446)	(−0.690)	(−0.302)	(−0.231)	(−0.878)	(−1.481)	(−0.807)	(−1.306)
TOP1	−0.00138	−0.000536	−0.00144	−0.000542	−0.00255 *	−0.00154 *	−0.00256 *	−0.00143	−0.00257 *	−0.00155 *
	(−0.982)	(−0.590)	(−1.024)	(−0.598)	(−1.775)	(−1.697)	(−1.869)	(−1.601)	(−1.855)	(−1.737)
Turnover	−8.32 × 10−6	−3.15 × 10−6	−7.41 × 10−6	5.40 × 10−7	−7.11 × 10−6	−7.45 × 10−6	1.71 × 10−5	1.42 × 10−5	8.52 × 10−6	9.38 × 10−6
	(−0.390)	(−0.229)	(−0.347)	(0.0392)	(−0.325)	(−0.539)	(0.825)	(1.052)	(0.405)	(0.692)
TOP10	0.00409 ***	0.00185 **	0.00411 ***	0.00178 **	0.00684 ***	0.00394 ***	0.00581 ***	0.00350 ***	0.00606 ***	0.00351 ***
	(3.389)	(2.375)	(3.406)	(2.284)	(5.538)	(5.036)	(4.944)	(4.581)	(5.092)	(4.578)
Sep	0.00124	0.00177	0.00127	0.00197	0.00112	0.00244 *	0.00166	0.00131	0.00124	0.00143
	(0.559)	(1.234)	(0.569)	(1.376)	(0.491)	(1.696)	(0.768)	(0.928)	(0.567)	(1.014)
Power	−0.0146	−0.00729	−0.0138	−0.00450	−0.00394	−0.00720	−0.0104	−0.00477	−0.0102	−0.000669
	(−0.584)	(−0.451)	(−0.550)	(−0.279)	(−0.154)	(−0.444)	(−0.428)	(−0.301)	(−0.413)	(−0.0421)
Committee	−0.0127	−0.00108	−0.0134	0.00445	−0.0163	−0.0145	−0.0101	0.00415	−0.0108	0.000872
	(−0.459)	(−0.0601)	(−0.483)	(0.249)	(−0.574)	(−0.809)	(−0.373)	(0.236)	(−0.394)	(0.0495)
Constant	−3.706 ***	−2.133 ***	−3.899 ***	−2.293 ***	−3.577 ***	−1.903 ***	−2.861 ***	−1.441 ***	−2.996 ***	−1.589 ***
	(−8.557)	(−7.613)	(−8.988)	(−8.202)	(−8.058)	(−6.774)	(−6.775)	(−5.245)	(−7.007)	(−5.769)
Observations	15,047	15,047	15,047	15,047	15,047	15,047	15,047	15,047	15,047	15,047
Adjusted R2	0.098	0.101	0.098	0.098	0.113	0.116	0.103	0.103	0.103	0.102
Firm FE	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES
Year FE	YES	YES	YES	YES	YES	YES	YES	YES	YES	YES

Note: Robust t-statistics are reported in parentheses. ***, ** and * denote significance at the 1%, 5% and 10% levels, respectively.

). The estimated coefficients of CSR are all significantly negative, aligning with our analytical logic.

Based on the above analysis and empirical regression results, the findings of this study indicate that enterprises that actively fulfill their social responsibilities can effectively reduce the risk of a stock price crash. Different from previous studies performed by scholars, we provide new evidence from the perspective of IRM. It is worth noting that, for a considerable period of time, skepticism has existed regarding the common belief in “running a good business by doing good deeds”, with some even viewing the fulfillment of CSR as a “hypocritical” act. Even in this context, when enterprises exhibit excellent social responsibility performance, well-managed investor relations can still reduce the risk of a stock price crash. This is because companies not only deliver positive news to investors, but they also strive to actively and effectively communicate with investors in the capital market to avoid potential misunderstandings. Such communication can effectively mitigate the risk of a stock price collapse when listed companies face extreme events.

7. Conclusions

Through the proactive fulfillment of social responsibilities and cultivation of robust investor relations, enterprises can not only bolster their sustainability, but also garner investor backing, and fortify the stability and longevity of their developmental trajectory. CSR performance is not only a way to win the favor of investors, but also an important means to improve the long-term sustainability of enterprises. However, within the field of CSR, scholars have continuously debated whether enterprises actively fulfilling social responsibilities are “genuinely altruistic” or merely engaging in “corporate hypocrisy”. CSR implementation can sometimes not only fail to bring positive effects to enterprises, but may bring negative effects because it is misunderstood as an attempt to cover up bad facts [26]. How to better realize the win–win situation of social responsibility and shareholder value has long been a challenge in both theoretical and practical circles. Investor relations management plays a crucial role in facilitating information communication for listed companies. When a company actively fulfills its social responsibility, it may sometimes be misinterpreted as enacting “hypocritical” behavior. How, then, can a company truly and effectively convey relevant information to the capital market? There is limited literature addressing this “prevention of misinterpretation”. Based on the perspective of the interaction between listed companies and investors, this article studies the relationship between CSR and IRM.

It is found that when enterprises actively fulfill their social responsibilities, they place greater emphasis on IRM in the capital market. On the one hand, when enterprises engage in proactive social responsibility efforts, they aim to establish a positive and responsible image among their investors. On the other hand, concerns arise regarding the potential misinterpretation of outstanding social responsibility performance as “hypocritical”. Therefore, companies seek to actively engage with investors to mitigate any negative perceptions that may impact the company’s value. The above conclusion remains valid after accounting for endogeneity issues and conducting a series of robustness tests. In the mechanism analysis, it is found that CSR primarily tends to improve the level of IRM by strengthening online and on-site communication. Furthermore, when financial distress is more serious, the financing constraint are tighter or the quality of accounting information disclosure is poor, enterprises are more inclined to actively interact with investors in the capital market to prevent “hypocritical” behavior that may be misinterpreted as social responsibility. In the extended analysis, the research results of this paper support the interpretation that the proactive implementation of CSR significantly mitigates the risk of a stock price crash. It is postulated that IRM serves as a crucial influencing mechanism in this context. The fulfillment of CSR is conducive to promoting the sustainable development of enterprises. The research conclusion provides a new way to promote the sustainable development of enterprises from the perspective of CSR and IRM.

Our findings carry significant implications for the strategy of disclosure of social responsibility information. Firstly, the fulfillment of CSR represents a long-term strategic imperative for enterprises, thus the scope of the mandatory disclosure of social responsibility information can be further expanded. We must emphasize the punitive measures for enterprises that fail to disclose, as enforced by regulations, and increase the strength of incentives for enterprises to voluntarily disclose. Secondly, enterprises ought to prioritize positive communication with investors during social responsibility construction, as failure to do so may be misconstrued as “hypocrisy” and backfire. Positive investment relations management can not only enable enterprises to fulfill their social responsibilities effectively, but also procures broader returns in the capital market, such as enhancing business performance and reducing the risk of a stock crash. This is achieved through a balanced approach that encompasses both doing “good” and running the business well.

Research limitations and future prospects: Through CSR practice, enterprises can have a positive social impact, achieve sustainable development, and enact more robust capital market performance and business performance. However, the specific impact of investor relations management from the perspective of corporate social responsibility on the capital market and business performance of enterprises still needs to be further explored and studied. In the future, corporate social responsibility and investor relationship management can be further studied from different aspects, such as investor trust and loyalty, risk management, etc., in order to reveal the deep mechanisms and paths of influence of these relationships, an thus provide enterprises with more effective investor relationship management strategies and promote sustainable development and value creation in enterprises.