Peer Effects on ESG Disclosure: Drivers and Implications for Sustainable Corporate Governance

Abstract

Amid growing global concerns regarding sustainable governance, understanding the drivers of ESG disclosure is vital for promoting transparency and responsible corporate behavior. This study examines the peer effects of ESG disclosure among 32,187 observations from Chinese A-share listed firms between 2010 and 2021. This research employs an instrumental variable approach based on stock-specific idiosyncratic returns estimated via the Carhart four-factor model to address endogeneity concerns. The results confirm significant peer effects, suggesting that firms adjust ESG practices in response to their industry counterparts. These effects are significantly moderated by firm-level characteristics, including information asymmetry, corporate reputation, and market competition, as well as by external conditions such as economic policy uncertainty, business environment volatility, and institutional quality. This research defines peer groups by industry affiliation and conducts robustness tests using ESG risk clustering to address classification bias. This study contributes to the literature by strengthening causal inference and refining the understanding of peer-driven ESG behavior by integrating institutional theory, signaling theory, and information economics. The findings offer practical implications for policymakers, investors, and corporate managers seeking to promote ESG convergence through peer-driven incentives in diverse regulatory contexts.

1. Introduction

The increasing global emphasis on sustainability has elevated Environmental, Social, and Governance (ESG) criteria as a key framework for evaluating corporate performance. Since its inception by the United Nations Environment Program (UNEP) in 2004, ESG disclosure has gained substantial momentum, providing stakeholders with a more comprehensive view of firms’ long-term potential [1,2]. Beyond enhancing transparency and risk management, ESG engagement is now widely recognized as a strategic driver of sustainable growth [3]. This shift aligns with deeper global capital integration and the proliferation of regulatory and voluntary ESG frameworks across jurisdictions.

Prior research on ESG disclosure has primarily focused on three areas: the conceptual development of ESG indicators [4], the determinants of disclosure behavior [5], and the economic implications of ESG engagement [6]. However, one critical dimension remains insufficiently examined: the role of peer effects in shaping ESG disclosure decisions. Although many recent studies have begun exploring ESG peer effects, most of these works emphasize surface-level peer correlations and lack deeper analyses of the transmission mechanisms and contextual moderators that influence peer-driven ESG behavior [7,8]. Emerging evidence suggests that firms may emulate their peers to reduce uncertainty or maintain legitimacy in the eyes of stakeholders [9]. This phenomenon is particularly relevant in emerging markets such as China, where formal disclosure standards remain fluid, and regulatory enforcement is uneven across regions [10].

Understanding peer dynamics in ESG disclosure is essential for at least two reasons. First, peer effects are well established in corporate domains such as investment, executive compensation, and governance [11,12]. Extending this line of inquiry to ESG can illuminate how sustainability practices diffuse across firms, contributing to broader transparency and responsibility. Second, the Chinese ESG landscape presents a hybrid institutional setting marked by soft regulatory signals (e.g., stock exchange guidelines and voluntary rating systems) and mounting stakeholder pressure from media, state ownership, and global investors [13,14]. This creates fertile ground for mimetic behaviors as firms navigate a complex interplay between normative expectations and regulatory ambiguity.

A fundamental challenge in peer effect research concerns the definition of “peer”. While industry-based classification remains the predominant approach, the recent literature has introduced more granular peer metrics, such as product market similarity [15] or ESG risk clustering [16]. These methods seek to capture strategic and sustainability-specific comparability beyond formal industry labels. Addressing this ongoing debate, the present study adopts industry classification as the baseline for peer identification. Still, it rigorously validates this choice through robustness checks using ESG sub-dimension clustering and alternative ESG sources (Bloomberg ESG scores). Moreover, to mitigate confounding from common industry trends, this research constructs peer variables that exclude the focal firm and incorporate industry-by-year fixed effects, ensuring that the estimated effects reflect inter-firm dynamics rather than shared macro-conditions.

To fill this gap, this study aims to conduct the following: (1) assess whether peer effects significantly influence ESG disclosure among Chinese listed firms; (2) identify the internal drivers—such as information asymmetry, reputation concerns, and market competition—that channel these effects; and (3) examine external moderators, including economic policy uncertainty and institutional quality. Empirical analyses employ fixed-effects panel regressions, instrumental variable strategies, and multiple peer group constructions. We find robust evidence of peer-driven ESG behavior, especially among state-owned, younger, and growth-oriented firms. The remainder of the paper is structured as follows: Section 2 reviews the theoretical and empirical literature on peer effects and ESG disclosure. Section 3 describes the dataset, variable construction, and econometric strategy. Section 4 presents the main empirical results. Section 5 offers robustness and endogeneity tests. Section 6 concludes with implications for ESG policy, disclosure standardization, and strategic benchmarking.

2. Literature Review

2.1. Theoretical Foundations

The concept of peer effects, rooted in social and behavioral economics, refers to the influence of reference groups on individual or organizational decisions [17,18,19]. Firms often observe and imitate peer organizations’ behavior in corporate settings, especially under uncertainty or incomplete information. Such imitation can serve strategic functions, mitigating reputational risk, facilitating organizational learning, and signaling alignment with market expectations [11,20]. Empirical studies have documented peer effects across diverse areas of corporate behavior, including investment decisions, capital structure, executive compensation, mergers and acquisitions, and corporate social responsibility (CSR) [21,22].

Institutional theory offers a robust framework to understand these dynamics. According to DiMaggio and Powell (1983), organizations tend to exhibit strategic convergence through institutional isomorphism [23], which operates via three primary mechanisms: coercive (regulatory), normative (professional standards), and mimetic (imitation under uncertainty) [24]. In contexts where formal regulation is weak, ambiguous, or evolving—as is often the case with ESG disclosure—mimetic pressures become particularly salient. Firms may emulate peers perceived as legitimate or successful, reducing the costs and risks of strategic experimentation [25,26]. In the domain of ESG disclosure, such mimetic behavior is increasingly evident. Firms are exposed to shared institutional pressures, including investor activism, reputational evaluations, and third-party ESG ratings [27,28]. These common institutional fields often give rise to convergence in disclosure practices, not necessarily through direct regulatory mandates but via socially embedded expectations.

In China, the ESG disclosure environment is in a state of transition. While formal regulatory mandates remain limited, the CSRC and stock exchanges have issued various guidance documents, pilot schemes, and voluntary disclosure frameworks [29]. However, a unified national ESG reporting requirement has yet to be implemented. This regulatory backdrop has led to an institutional environment shaped more by normative and mimetic forces than coercive legislation [30]. In such a hybrid context, firms increasingly turn to peer firms as reference points in shaping their ESG strategies, reinforcing the relevance of peer effects in emerging markets.

2.2. Peer Effects in ESG Disclosure

Peer effects have been extensively examined in corporate decision-making areas such as investment, capital structure, and executive compensation [31,32]. Yet, their role in ESG disclosure remains a relatively recent area of inquiry. Emerging studies indicate that firms often imitate the ESG practices of their peers to gain legitimacy, signal alignment with market expectations, and adapt to evolving sustainability norms. These effects are particularly prominent in emerging markets, where regulatory ambiguity and institutional voids compel firms to rely on informal cues and peer benchmarks [33,34].

A central methodological challenge lies in defining what constitutes a “peer”. Most empirical studies adopt industry-based classifications—such as SIC, NAICS, or CSRC codes—based on the premise that firms within the same industry face similar stakeholder demands and regulatory environments. This approach is consistent with ESG rating practices in China and captures institutional and competitive proximity [16,17]. However, the recent literature advocates for more refined definitions, including product market similarity [15], ESG risk clustering [35], supply chain relationships, board interlocks, and geographic proximity [36,37,38]. While these advanced methods may improve precision, they demand extensive unstructured data and are often impractical for large-scale studies in emerging markets.

China’s ESG disclosure landscape is shaped by a hybrid institutional environment. Although comprehensive national regulations are evolving, the CSRC and stock exchanges have issued guidance and voluntary frameworks promoting mimetic and normative pressures. Firms often respond by aligning with industry peers rather than complying with coercive mandates [21,22]. Given its regulatory relevance and widespread acceptance, this study adopts the CSRC industry classification as the primary basis for identifying peer groups. To mitigate concerns of oversimplification, robustness checks are conducted using alternative peer definitions, such as ESG sub-dimension clustering, and an alternative ESG metric from Bloomberg. Product market similarity is not employed due to data constraints, though the adopted strategies reflect prevailing scholarly approaches. Thus, this research proposes the following:

H1: 

There is a significant industry peer effect in corporate ESG disclosure, influenced by other enterprises in the same industry.

2.3. Internal Factors Influencing Peer Effects in ESG Disclosure

To better understand how peer effects influence corporate ESG disclosure, examining the conditions under which these effects are amplified or attenuated is essential. This section explores three theoretically grounded factors—information asymmetry, corporate reputation, and market competition—commonly identified as contextual forces shaping firm behavior under uncertainty. Importantly, these variables are conceptualized as moderators rather than mediators. They do not explain why peer effects occur, but they examine when and to what extent firms are more likely to emulate peer behavior in ESG-related decision-making. This distinction aligns the empirical approach with the perspectives of core information economics, institutional, and signaling theory.

Information asymmetry is central to understanding how firms address disparities in data access and stakeholder perceptions [39,40]. Corporate reputation highlights the strategic value of intangible assets in competitive environments [41], and market competition reflects the pressures firms face to align with peer practices for survival and growth [9]. Together, these indicators provide a comprehensive framework to examine the forces driving peer effects in ESG disclosure, grounded in information asymmetry, signaling, and competition theories. Information asymmetry arises when firms and external stakeholders have unequal access to relevant information, heightening market uncertainty and stakeholder skepticism [42]. ESG disclosure serves as a signaling mechanism that reduces these asymmetries by enhancing transparency, mitigating adverse selection risks, and strengthening stakeholder confidence [43]. In environments with high information asymmetry, firms experience greater pressures to mimic peers’ ESG practices to align with perceived norms, manage legitimacy concerns, and reduce informational disadvantages [44]. Based on this rationale, we hypothesize the following:

H2: 

Information asymmetry positively moderates the peer effect of corporate ESG disclosure, such that firms facing higher information asymmetry are more likely to imitate their peers’ ESG disclosure behaviors.

Corporate reputation, a key intangible asset, significantly influences firms’ strategic choices, particularly in the context of ESG disclosure [45]. Signaling theory suggests that robust ESG performance sends positive signals to investors and stakeholders, differentiating firms from their competitors and enhancing market valuation [46]. ESG engagement is thus strategically leveraged to build reputational capital, which is critical in markets where intangible assets drive competitive advantage. Effective ESG practices are associated with reduced financing costs, enhanced risk management, and increased market value [47,48]. In prominent industries where reputation is a critical driver of competitiveness, firms often emulate peers’ successful ESG practices to maintain or enhance their reputational standing. This dynamic underscores the importance of reputation-building behaviors in amplifying peer-driven ESG disclosure. Emerging evidence from Pakistan also highlights the role of corporate reputation, alongside CSR and government incentives, in shaping firms’ environmental investment intentions [49]. Consequently, we propose the following:

H3: 

Corporate reputation positively moderates the peer effect of corporate ESG disclosure, such that firms with greater reputational concerns are more inclined to align their ESG disclosure behaviors with peers.

Market competition exerts significant pressure on firms to mimic the ESG practices of their peers. According to competition theory, firms in highly competitive industries face greater uncertainty regarding stakeholder expectations and market dynamics, increasing the incentive to imitate successful peers as a risk-averse strategy [50]. Institutional theory suggests that intensified competition fosters mimetic isomorphism, where firms adopt standard practices like ESG disclosure to gain legitimacy and survive. ESG disclosure has emerged as a key strategy for firms to address competitive challenges, especially in industries with intense rivalry [51]. Moreover, from a resource-based perspective, firms under competitive pressure may replicate peer ESG practices to efficiently allocate scarce resources, strengthen market positioning, and minimize strategic uncertainty. Firms operating under similar resource constraints or volatile external conditions are particularly likely to align their ESG practices with peers to sustain their competitive advantage. Based on this understanding, we hypothesize the following:

H4: 

Market competition positively moderates the peer effect of corporate ESG disclosure, such that firms operating in more competitive environments exhibit stronger peer-driven ESG disclosure behaviors.

2.4. External Factors Influencing Peer Effects in ESG Disclosure

The external environment profoundly influences corporate ESG strategies, shaping how firms interpret and respond to peer behaviors [52]. In particular, when formal ESG mandates are nascent, fragmented, or unevenly enforced, firms are compelled to rely more heavily on informal signals and peer benchmarking to guide strategic decisions [53]. This section investigates three key external moderators: economic policy uncertainty, business environment uncertainty, and institutional embeddedness. These factors are theorized not as mediating mechanisms but as contingencies that affect the strength or salience of peer effects in ESG disclosure.

Economic policy uncertainty represents a macro-level challenge that intensifies information asymmetries and complicates managerial decision-making. Under inconsistent or ambiguous regulatory signals, firms experience heightened difficulties interpreting policy intentions, forecasting compliance requirements, and anticipating enforcement actions [54]. Such uncertainty can exacerbate agency problems and reduce the reliability of formal guidance. Under these conditions, managers increasingly turn to peer firms, especially those perceived as legitimate, reputable, or strategically successful—as reference points to mitigate uncertainty and reputational risk. Information learning theory suggests that firms mitigate uncertainty by referencing peers viewed as legitimate or strategically aligned [55], while competition theory posits that firms mimic successful counterparts to preserve competitive standing [56]. Thus, the following was considered:

H5: 

Economic policy uncertainty positively affects the peer effect of corporate ESG disclosure, such that firms facing higher economic policy uncertainty are more likely to imitate their peers’ ESG disclosure behaviors.

Business environment uncertainty, by contrast, originates from firm-specific operational contexts, such as volatile demand, input instability, or stakeholder pressures. Unlike policy-level shocks, this uncertainty is localized and heterogeneous, affecting firms differently even within the same industry [57]. Firms operating under high environmental volatility often experience bounded rationality, lacking the informational or strategic bandwidth to innovate independently. Instead, they turn to peers as heuristic guides for acceptable ESG behavior. By referencing peers, firms can rapidly validate strategic choices, reduce the perceived risks of deviation, and maintain reputational consistency in unstable markets [58]. Accordingly, the following was hypothesized:

H6: 

Business environment uncertainty positively moderates the peer effect of corporate ESG disclosure, such that firms operating in more volatile business environments are more likely to align their ESG disclosure behaviors with peers.

The institutional environment, defined by shared regulatory norms, enforcement capacity, and cultural expectations, also plays a pivotal role. Institutional theory suggests that firms in similar institutional fields experience stronger isomorphic pressures [59]. In China, variations in provincial government initiatives, regional stock exchange regulations, and local enforcement intensity create a heterogeneous institutional landscape. Firms operating within closer institutional proximity—characterized by similar regulatory norms and enforcement cultures—are more susceptible to peer influences in shaping their ESG strategies. Ecological systems theory further highlights how co-located firms co-evolve under mutual constraints and shared resources [60]. Institutional embeddedness fosters convergence in ESG disclosure practices within peer groups by reinforcing social expectations and normative legitimacy. Therefore, the following was considered:

H7: 

The institutional environment positively moderates the peer effect of corporate ESG disclosure, such that firms embedded in similar institutional contexts exhibit stronger peer-driven ESG disclosure behaviors.

These external moderators provide a multi-level framework for understanding when and where ESG peer effects are likely to emerge or intensify. By integrating macro-level uncertainty, meso-level institutional structures, and firm-specific operational volatility, the proposed framework highlights the complex contingencies shaping ESG peer dynamics. Figure 1 visually summarizes the proposed conceptual model, integrating firm-level and contextual factors shaping ESG peer dynamics.

3. Methodology

3.1. Samples and Data Sources

This study investigates peer effects in ESG disclosure using a panel dataset of Chinese A-share listed firms from 2010 to 2021. Financial and firm-level data are drawn from the CSMAR database, while ESG disclosure scores are sourced from the Huazheng ESG Rating System, a widely adopted benchmark in China’s capital markets. To ensure sample reliability, firms in the financial sector are excluded due to distinct regulatory frameworks, along with firms labeled ST, PT, or *ST, which are under special treatment. Observations with missing or extreme values are removed, and continuous variables are winsorized at the 1st and 99th percentiles to reduce outlier bias. The final sample comprises 32,187 firm-year observations. Following the prior literature on peer effects, peer groups are defined by industry affiliation [11].

Specifically, the China Securities Regulatory Commission (CSRC) industry classification is used to identify peer firms—those operating in the same industry and year as the focal firm, excluding the firm itself. This official classification aligns with ESG rating practices in China. It captures shared institutional contexts, stakeholder expectations, and competitive dynamics, making it suitable for assessing peer influences on ESG disclosure behavior. However, it is essential to acknowledge that while widely used, industry affiliation may not fully capture more nuanced strategic, operational, or disclosure-specific similarities among firms. To address this potential limitation, this study conducts robustness checks by constructing alternative peer groups based on ESG risk clustering. This approach provides an additional layer of validation, ensuring that the observed peer effects are not solely driven by industry-level homogeneity.

3.2. Variable Definitions

The dependent variable in this study is the firm-level ESG disclosure score (ESG), derived from the Huazheng ESG Index. The primary independent variable is the industry-level peer ESG score (IndPeerESG), calculated as the average ESG score of all other firms within the same industry and year (excluding the focal firm). This approach helps avoid mechanical correlation with the dependent variable and ensures that the peer construction reflects external reference pressure. The construction model is as follows:

$$\mathit{I}\mathit{n}\mathit{d}{\mathit{P}\mathit{e}\mathit{e}\mathit{r}\mathit{E}\mathit{S}\mathit{G}}_{\mathit{i},\mathit{t}}={\displaystyle \frac{1}{{\mathit{N}}_{\mathit{g},\mathit{t}}-\mathbf{1}}}\sum _{\begin{array}{c}\mathit{j}\in \mathit{g}\\ \mathit{j}\ne \mathit{i}\end{array}}{\mathit{E}\mathit{S}\mathit{G}}_{\mathit{j},\mathit{t}}$$

(1)

To strengthen identification and address concerns over omitted variable bias or industry-wide time trends, we apply firm- and year-fixed effects and conduct additional robustness checks, including the inclusion of industry × year fixed effects and alternative peer group constructions. Importantly, this empirical strategy is guided by the conceptual model developed in Figure 1, which links ESG disclosure to peer behavior and proposes a set of internal and external moderators. Internal moderators include Information Asymmetry (Asy), Corporate Reputation (Rep), and Market Competitiveness (HHI), capturing firm-specific conditions under which peer effects become more salient. External moderators, including Economic Policy Uncertainty (EPU), Business Environment Uncertainty (EU), and Institutional Environment (Market), reflect contextual factors that amplify or constrain firms’ responsiveness to peer disclosure norms.

To test these relationships, we introduce moderator variables into interaction models. In addition, control variables are included to account for firm-level fundamentals such as size, growth, leverage, profitability, liquidity, governance structure, and their peer-level averages.

Table 1. Variable definitions and construction.

Category	Variable Name	Definition and Calculation
Explained Variable	Corporate ESG Disclosure (ESG)	The ESG disclosure score ranging from “AAA” (9) to “C” (1), obtained from the China Huazheng ESG Index.
Explanatory Variable	Industry Peer ESG (IndPeerESG)	The average ESG score of peer firms within the same industry and year, excluding the focal firm, as in Model (1).
Internal Moderators	Information Asymmetry (Asy)	An aggregated measure of trading-based asymmetry derived from three indicators: LR (liquidity ratio), ILL (liquidity), and GAM (yield reversal) [61].
		LR: Captures trading efficiency. ${\mathit{L}\mathit{R}}_{\mathit{i},\mathit{t}}=-{\displaystyle \frac{\mathbf{1}}{{\mathit{D}}_{\mathit{i},\mathit{t}}}}{\sum }_{\mathit{k}=\mathbf{1}}^{{\mathit{D}}_{\mathit{i},\mathit{t}}}\sqrt{{\displaystyle \frac{{\mathit{V}}_{\mathit{i},\mathit{t}}\left(\mathit{k}\right)}{\left|{\mathit{r}}_{\mathit{i},\mathit{t}}\left(\mathit{k}\right)\right|}}}$
		ILL: Reflects price movement relative to trading volume. ${\mathit{I}\mathit{L}\mathit{L}}_{\mathit{i},\mathit{t}}=-{\displaystyle \frac{\mathbf{1}}{{\mathit{D}}_{\mathit{i},\mathit{t}}}}{\sum }_{\mathit{k}=\mathbf{1}}^{{\mathit{D}}_{\mathit{i},\mathit{t}}}\sqrt{{\displaystyle \frac{\left|{\mathit{r}}_{\mathit{i},\mathit{t}}\left(\mathit{k}\right)\right|}{{\mathit{V}}_{\mathit{i},\mathit{t}}\left(\mathit{k}\right)}}}$
		GAM: Measures short-term price corrections. A regression-based approach using excess returns and trading volume.
	Corporate Reputation (Rep)	Corporate Reputation (Rep) is constructed following a multi-stakeholder evaluation framework based on the prior literature [62]. Selecting 12 indicators reflecting perspectives from consumers, creditors, shareholders, and firms can ensure operability, hierarchy, and validity (the selected indicators include (i) consumer and social perspectives: industry rankings of firm assets, revenue, net profit, and firm value; (ii) creditor perspective: the debt-to-asset ratio, current ratio, and long-term debt ratio; (iii) shareholder perspective: earnings per share (EPS), dividend per share (DPS), and audit by a Big Four accounting firm; and (iv) corporate governance perspective: sustainable growth rate and proportion of independent directors). The 12 indicators are then factor-analyzed to extract the principal components representing reputational capital. Firms are then ranked by their factor scores and grouped into ten deciles, with Rep assigned values from 1 (the lowest reputation) to 10 (the highest reputation).
	Market Competitiveness (HHI)	Following established empirical practices, this research adopts the HHI as a proxy for industry-level competitive intensity. Herfindahl–Hirschman Index (HHI) is calculated as $\mathit{H}\mathit{H}\mathit{I}=\mathit{s}\mathit{u}\mathit{m}\left[{\left({\mathit{X}}_{\mathit{i}}/\mathit{X}\right)}^2\right]$, where Xi is firm revenue and X is total revenue. Higher HHI values indicate greater market concentration and lower competition [63,64].
External Moderators	Economic Policy Uncertainty (EPU)	Economic Policy Uncertainty (EPU) is measured using the index developed by Davis, Liu, and Sheng (2019), based on a textual analysis of People’s Daily and Guangming Daily [65]. We compute an annual EPU value as the weighted average of monthly indices. This measure, released regularly, captures the frequency of policy-related uncertainty terms and is widely adopted in studies on China’s macro-financial environment.
	Business Environment Uncertainty (EU)	The industry-adjusted standard deviation of sales revenue over five years reflects operating instability [66,67].
	Institutional Environment (Market)	The marketization index assesses government-market relations, non-state economic development, and the legal environment. Higher values indicate stronger institutions [68].
Control Variables	Firm-Level Controls	This includes firm size, Tobin’s Q, profitability, indebtedness, liquidity, and growth.
	Peer Firm Characteristics	These are peer-level averages (excluding the focal firm) of the above firm-level controls.
	Fixed Effects	These include year, industry, and firm fixed effects.

provides full variable descriptions and definitions.

3.3. Model Construction

This study constructs the following model (2) to test (H1) regarding whether a peer effect exists in corporate ESG disclosure. This model specification builds on Leary and Roberts (2014) and Manski (1993), where peer groups are defined based on industry affiliation, a common practice in the literature [11,69]. By excluding the focal firm from the peer average and incorporating fixed effects, we minimize concerns of reflection bias and time-invariant omitted variables. To strengthen the robustness of our identification strategy, we implement two additional sets of tests. First, we redefine peer groups based on ESG sub-dimension clustering (E, S, and G) to capture similarity in sustainability exposure beyond industry affiliation. Second, we re-estimate the baseline model by including industry-by-year fixed effects, which account for potential confounding due to time-varying industry-level shocks or common disclosure trends.

$${\mathit{E}\mathit{S}\mathit{G}}_{\mathit{i},,\mathit{t}}={\mathsf{\alpha }}_{\mathbf{0}}+{\mathsf{\alpha }}_{\mathbf{1}}{\mathit{I}\mathit{n}\mathit{d}\mathit{p}\mathit{e}\mathit{e}\mathit{r}\mathit{E}\mathit{S}\mathit{G}}_{\mathit{i},\mathit{t}}+{\mathsf{\alpha }}_{\mathbf{2}}{\mathit{C}\mathit{o}\mathit{n}\mathit{t}\mathit{r}\mathit{o}\mathit{l}\mathit{s}}_{\mathit{i},\mathit{t}}+{\mathsf{\alpha }}_{\mathbf{3}}{\mathit{I}\mathit{n}\mathit{d}\mathit{p}\mathit{e}\mathit{e}\mathit{r}\mathit{C}\mathit{o}\mathit{n}\mathit{t}\mathit{r}\mathit{o}\mathit{l}}_{\mathit{i},\mathit{t}}+\sum \mathit{Y}\mathit{e}\mathit{a}\mathit{r}+\sum \mathit{i}\mathit{d}+{\mathsf{\epsilon }}_{\mathit{i},\mathit{t}}$$

(2)

where ${\mathrm{E}\mathrm{S}\mathrm{G}}_{\mathrm{i},\mathrm{t}}$ denotes the ESG disclosure score of firm $\mathit{i}$ in year $\mathit{t}$, and ${\mathit{I}\mathit{n}\mathit{d}\mathit{p}\mathit{e}\mathit{e}\mathit{r}\mathit{E}\mathit{S}\mathit{G}}_{\mathit{i},\mathit{t}}$ is the average ESG disclosure of peer firms (excluding firm $\mathit{i}$) in the same industry and year. ${\mathit{C}\mathit{o}\mathit{n}\mathit{t}\mathit{r}\mathit{o}\mathit{l}\mathit{s}}_{\mathit{i},\mathit{t}}$ represents firm-specific control variables, including firm size, TobinQ, profitability, indebtedness, liquidity, and growth. ${\mathit{I}\mathit{n}\mathit{d}\mathit{p}\mathit{e}\mathit{e}\mathit{r}\mathit{C}\mathit{o}\mathit{n}\mathit{t}\mathit{r}\mathit{o}\mathit{l}}_{\mathit{i},\mathit{t}}$ represents the average of the same control variables among peer firms (excluding firm $\mathit{i}$), and $\sum \mathit{Y}\mathit{e}\mathit{a}\mathit{r}$ and $\sum \mathit{i}\mathit{d}$ present firm and year fixed effects, respectively. ${\mathsf{\epsilon }}_{\mathit{i},\mathit{t}}$ is the random disturbance term.

3.4. Descriptive Statistics

Table 2. Variable descriptive statistics.

Variable	N	Mean	SD	Min	p25	p50	p75	Max
ESG	32,187	4.142	1.095	1	4	4	5	8
IndpeerESG	32,187	4.115	0.366	1	3.941	4.126	4.311	6
Size	32,187	22.17	1.288	19.72	21.23	21.97	22.89	26.46
TobinQ	32,187	2.017	1.353	1.050	1.236	1.608	2.308	13.49
Roe	32,186	0.0712	0.121	−0.827	0.0325	0.0767	0.126	0.406
Lev	32,187	0.415	0.205	0.0298	0.249	0.407	0.569	0.901
Cashflow	32,187	0.0470	0.0686	−0.222	0.00875	0.0465	0.0871	0.258
Growth	32,187	0.181	0.390	−0.648	−0.00786	0.120	0.284	3.541
IndpeerSize	32,187	132.8	238.4	−2.13 × 10−9	16.44	53.75	117.6	4637
IndpeerTobinQ	32,187	12.41	22.95	−2.94 × 10−10	1.566	5.055	11.36	491.6
IndpeerRoe	32,187	0.427	0.780	−0.174	0.0531	0.167	0.382	16.09
IndpeerLev	32,187	2.397	4.306	−0	0.310	0.991	2.073	85.33
IndpeerCashflow	32,187	0.280	0.517	−0.336	0.0319	0.114	0.257	11.87
IndpeerGrowth	32,187	1.117	2.278	−1.135	0.117	0.396	0.980	59.01

presents descriptive statistics for the 32,187 firm-year observations in the final sample. ESG disclosure scores for Chinese A-share listed companies range from 1 to 8, with a mean score of 4.142. This relatively low average reflects the Chinese market’s absence of standardized ESG disclosure norms and evaluation criteria. The interquartile range suggests clustering around the mean, indicating limited variance in ESG scores among firms. The average of the industry-level peer ESG scores (IndPeerESG) is 4.115, closely aligning with the firm-level mean. This convergence highlights the consistency of ESG disclosure practices within industry cohorts and supports the rationale for exploring the peer effect, partially validating H1.

We also examine variable interdependence through Pearson and Spearman correlation matrices (see Appendix A 

Table A1. Correlation coefficient.

Variable	ESG	IndpeerESG	Size	Tobinq	Roe	Lev	Cashflow	Growth	IndpeerSize	IndpeerTobinq	IndpeerRoe	IndpeerLev	IndpeerCashflow	IndpeerGrowth
ESG		0.204 ***	0.154 ***	−0.105 ***	0.228 ***	−0.060 ***	0.082 ***	0.065 ***	−0.060 ***	−0.062 ***	−0.056 ***	−0.058 ***	−0.065 ***	−0.055 ***
IndpeerESG	0.194 ***		0.007	−0.115 ***	0.058 ***	0.069 ***	−0.040 ***	0.024 ***	−0.087 ***	−0.113 ***	−0.074 ***	−0.075 ***	−0.118 ***	−0.078 ***
Size	0.185 ***	0.032 ***		−0.488 ***	0.087 ***	0.524 ***	0.074 ***	0.024 ***	−0.043 ***	−0.067 ***	−0.053 ***	−0.016 ***	−0.031 ***	−0.055 ***
Tobinq	−0.100 ***	−0.071 ***	−0.339 ***		0.154 ***	−0.355 ***	0.106 ***	0.094 ***	0.044 ***	0.112 ***	0.052 ***	0.016 ***	0.065 ***	0.077 ***
Roe	0.219 ***	0.039 ***	0.087 ***	0.091 ***		−0.116 ***	0.366 ***	0.369 ***	−0.075 ***	−0.067 ***	−0.035 ***	−0.078 ***	−0.071 ***	−0.035 ***
Lev	−0.069 ***	0.049 ***	0.520 ***	−0.245 ***	−0.183 ***		−0.146 ***	0.004	−0.027 ***	−0.057 ***	−0.032 ***	0.012 **	−0.064 ***	−0.031 ***
Cashflow	0.076 ***	−0.030 ***	0.075 ***	0.117 ***	0.311 ***	−0.155 ***		0.046 ***	−0.037 ***	−0.030 ***	−0.034 ***	−0.047 ***	0.041 ***	−0.055 ***
Growth	0.002	0.007	0.035 ***	0.053 ***	0.262 ***	0.029 ***	0.018 ***		0.002	0.011 **	0.039 ***	−0.002	−0.020 ***	0.094 ***
IndpeerSize	−0.063 ***	−0.022 ***	0.024 ***	−0.000	−0.057 ***	0.034 ***	−0.034 ***	−0.010 *		0.983 ***	0.967 ***	0.994 ***	0.942 ***	0.930 ***
IndpeerTobinq	−0.066 ***	−0.036 ***	0.019 ***	0.032 ***	−0.057 ***	0.024 ***	−0.029 ***	−0.007	0.978 ***		0.959 ***	0.968 ***	0.941 ***	0.931 ***
IndpeerRoe	−0.064 ***	−0.024 ***	0.015 ***	0.007	−0.032 ***	0.024 ***	−0.033 ***	0.015 ***	0.963 ***	0.948 ***		0.959 ***	0.913 ***	0.948 ***
IndpeerLev	−0.061 ***	−0.017 ***	0.030 ***	−0.005	−0.059 ***	0.046 ***	−0.039 ***	−0.010 *	0.997 ***	0.973 ***	0.953 ***		0.924 ***	0.923 ***
IndpeerCashflow	−0.063 ***	−0.025 ***	0.027 ***	0.012 **	−0.054 ***	0.009	0.004	−0.022 ***	0.943 ***	0.933 ***	0.900 ***	0.928 ***		0.854 ***
IndpeerGrowth	−0.060 ***	−0.028 ***	0.019 ***	0.015 ***	−0.027 ***	0.028 ***	−0.040 ***	0.043 ***	0.896 ***	0.896 ***	0.946 ***	0.894 ***	0.789 ***

t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

) and calculate Variance Inflation Factors (VIFs) for all regressors. All VIF values are below the conventional threshold of 10, suggesting multicollinearity is not a concern [70]. These diagnostics confirm that the data are suitable for regression analysis and that the model structure is robust.

4. Empirical Analyses and Results

4.1. Data Analysis of ESG Peer Effect

This study investigates the peer effect on corporate ESG disclosure using regression analyses based on the specified model (2). A two-way fixed effects approach is employed, controlling for firm and year effects to ensure robust results. The regression results are presented in

Table 3. Data analysis of ESG peer effect.

Variable	(1)	(2)	(3)	(4)	(5)	(6)
	ESG (OLS)	ESG (OLS)	ESG (RE)	ESG (RE)	ESG (FE)	ESG (FE)
IndpeerESG	0.1452 ***	0.1392 ***	0.1443 ***	0.1392 ***	0.1442 ***	0.1389 ***
	(27.5954)	(27.2093)	(26.8362)	(27.2093)	(26.7939)	(26.6572)
Control variables	No	Yes	No	Yes	No	Yes
Indpeer_Control variables	No	Yes	No	Yes	No	Yes
Year (FE)	No	No	No	No	Yes	Yes
Firms (FE)	No	No	No	No	Yes	Yes
Constant	0.7375 ***	−0.9139 ***	0.7429 ***	−0.9139 ***	0.7417 ***	−1.0298 ***
	(33.9291)	(−20.5743)	(30.3182)	(−20.5743)	(33.3706)	(−22.7865)
N	32,187	32,183	32,187	32,183	32,187	32,183
adj. R2	0.023	0.095			0.021	0.097

t statistics in parentheses. *** p < 0.01.

, using a stepwise approach for clarity. Columns (1), (3), and (5) present results without firm-specific control variables. These regressions reveal a significantly positive peer effect on ESG disclosure, with significant coefficients at the 1% level. Columns (2), (4), and (6) incorporate firm-specific control variables, year-fixed effects, and firm-level fixed effects. The results consistently indicate a significant positive relationship between peer ESG scores and individual firm ESG disclosures, supporting the presence of an industry peer effect for ESG disclosure. This study compares mixed OLS, random effects, and fixed effects models to validate the model selection. The F-value and Hausman tests confirm that fixed effects models are the most appropriate. The detailed panel data results, particularly for Column (6), are analyzed further in the following sections.

4.2. Robustness Test and Endogeneity Tests

To ensure the robustness and internal validity of the benchmark findings, we conduct a series of additional empirical checks, targeting potential threats from omitted variables, sample-specific shocks, and mechanical correlation. The results are presented in

Table 4. Robustness test.

Variable	(1)	(2)	(3)	(4)	(5)	(6)	(7)
	Esg	ESG	ESG (2010–2019)	ESG	ESG	ESG	Bloomberg_ESG
IndpeerESG	0.139 ***	0.140 ***	0.134 ***	0.078 ***	0.069 ***		0.569 *
	(26.66)	(29.34)	(24.81)	(7.22)	(10.14)		(1.93)
ClusterPeerESG						0.618 ***
						(46.58)
Control variables	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Indpeer_Control variables	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Adding control variables	No	Yes	No	No	No	No	No
Year (FE)	Yes	Yes	Yes	Yes	No	Yes	Yes
Firms (FE)	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Industry (FE)	No	No	No	Yes	No	No	No
Province (FE)	No	No	No	Yes	No	No	No
Industry_Year (FE)	No	No	No	No	Yes	No	No
Constant	−1.030 ***	0.814 ***	−0.186 ***	−0.988 ***	−0.872 ***	−1.399 ***	−5.595
	(−22.79)	(6.02)	(−3.92)	(−6.58)	(−17.55)	(−10.36)	(−1.03)
Observations	32,183	32,183	24,436	32,183	32,175	31,531	11,672
R-squared	0.098	0.133	0.114	0.119	0.127	0.525	0.844
r2_a	0.0972	0.132	0.113	0.117	0.120	0.461	0.824
F	268.5	259.1	242.2	156.6	229.8	192.3	7.092

t-statistics in parentheses. *** p < 0.01, * p < 0.1.

.

In Column (1), the year-end ESG score used in the baseline model is replaced by the average of the firm’s quarterly ESG scores. This adjustment reduces the risk of end-period bias and confirms the temporal consistency of the peer effect, which remains statistically significant. Column (2) incorporates additional control variables, including firm age, board size, and the shareholding ratio of the largest shareholder. These factors may influence ESG disclosure and firms’ responsiveness to peer behavior. The peer effect remains stable after these additions, indicating that it is not confounded by omitted firm-level governance characteristics. To account for potential structural shocks caused by the COVID-19 pandemic, Column (3) excludes the years 2020 and 2021. The estimated coefficient remains statistically significant, suggesting that the peer effect persists even during non-crisis periods. In Column (4), industry and province fixed effects are introduced to capture unobserved heterogeneity across regions and sectors. The results remain robust, underscoring that static differences in institutional or economic environments do not drive the peer effect. Column (5) introduces industry-by-year fixed effects to control for sector-specific temporal shocks, including changes in disclosure mandates, ESG-related campaigns, and rating agency behavior. This specification addresses concerns that the peer effect might reflect industry-level convergence by accounting for common industry trends. The coefficient of IndPeerESG (0.069, t = 10.14) remains highly significant, reinforcing the argument that the observed peer effect reflects genuine behavioral spillovers.

Column (6) refines the peer definition by clustering firms based on ESG sub-dimension scores (E, S, and G). This clustering approach captures deeper sustainability-related comparability, reflecting firms’ exposure to similar environmental, social, and governance risks. It aligns with the emerging literature advocating ESG-specific peer grouping over simple industry affiliation [16]. The resulting ClusterPeerESG variable remains significantly associated with firm-level ESG disclosure (coefficient = 0.618, t = 46.58), proving that firms emulate others with similar ESG profiles. These findings validate the robustness of peer effects beyond traditional industry-based definitions, enhancing the reliability of the study’s conclusions. Finally, Column (7) uses Bloomberg ESG scores as an alternative disclosure measure to address concerns about mechanical correlation since the baseline uses ESG scores from the same source for both dependent and explanatory variables. Despite a reduced sample size, the peer effect remains marginally significant at the 10% level, further validating the robustness of the findings. These tests confirm that the observed peer effect is not an artifact of sample design, omitted variables, or measurement source, but rather a consistent behavioral phenomenon.

To address potential endogeneity concerns arising from omitted variables and reverse causality, an instrumental variable (IV) approach is employed. The instrument (IndPeerIdioReturn), defined as the average idiosyncratic return of peer firms in the same industry, captures firm-specific performance shocks exogenous to the focal firm’s ESG disclosure decision but correlated with peer behavior. This satisfies the relevance and exogeneity criteria as it isolates peer-specific noise uncorrelated with common industry trends or macro shocks. Following the approach of Leary and Roberts (2014) [11], firm-level residuals are estimated using the Carhart four-factor model (3).

$${\mathit{r}}_{\mathit{i}\mathit{j}\mathit{t}}-\mathit{r}{\mathit{f}}_{\mathit{t}}={\mathsf{\alpha }}_{\mathit{i}\mathit{j}\mathit{t}}+{\mathsf{\beta }}_{\mathit{i}\mathit{j}\mathit{t}}^{\mathit{M}}{\mathit{M}\mathit{K}\mathit{T}}_{\mathit{t}}+{\mathsf{\beta }}_{\mathit{i}\mathit{j}\mathit{t}}^{\mathit{I}\mathit{N}\mathit{D}}\left({\overline{\mathit{r}}}_{-\mathit{i}\mathit{j}\mathit{t}}-\mathit{r}{\mathit{f}}_{\mathit{t}}\right)+{\mathsf{\beta }}_{\mathit{i}\mathit{j}\mathit{t}}^{\mathit{S}\mathit{M}\mathit{B}}{\mathit{S}\mathit{M}\mathit{B}}_{\mathit{t}}+{\mathsf{\beta }}_{\mathit{i}\mathit{j}\mathit{t}}^{\mathit{H}\mathit{M}\mathit{L}}{\mathit{H}\mathit{M}\mathit{L}}_{\mathit{t}}+{\mathsf{\beta }}_{\mathit{i}\mathit{j}\mathit{t}}^{\mathit{M}\mathit{O}\mathit{M}}{\mathit{M}\mathit{O}\mathit{M}}_{\mathit{t}}+{\mathsf{ϵ}}_{\mathit{i}\mathit{j}\mathit{t}}$$

(3)

where ${\mathit{M}\mathit{K}\mathit{T}}_{\mathit{t}}$, ${\mathit{S}\mathit{M}\mathit{B}}_{\mathit{t}}$,${\mathit{H}\mathit{M}\mathit{L}}_{\mathit{t}}$ and ${\mathit{M}\mathit{O}\mathit{M}}_{\mathit{t}}$ represent the four Carhart factors: market, size, book-to-market ratio, and momentum. Residuals (${\mathsf{ϵ}}_{\mathit{i}\mathit{j}\mathit{t}})$ are used as the IV for peer ESG disclosure. Furthermore, while peer firms’ idiosyncratic stock returns reflect firm-specific shocks influencing peers’ financial behaviors and ESG disclosure incentives, they are unlikely to directly affect the focal firm’s own ESG disclosure decisions, conditional on industry and year fixed effects. This satisfies the exclusion restriction assumption by ensuring that the instrument’s only pathway to influence is through peer firm behavior rather than through omitted correlated shocks.

The Hausman test confirms exogeneity of the IV (p = 0.402, p = 0.848), and both the Cragg–Donald Wald F-statistic (28.19) and Kleibergen–Paap rk Wald F-statistic (34.56) exceed the conventional threshold of 10, indicating no weak instrument concerns. As shown in

Table 5. Instrumental variable approach.

Variable	(1)	(2)
	IndpeerESG	ESG
IndpeerIdioreturn	0.008 ***
	(0.001)
IndpeerESG		−0.285 **
		(2.524)
Constant		2.524 ***
		(0.585)
Observations	18,934	18,934
R-squared		0.930
Cragg–Donald Wald F	28.19
Kleibergen–Paaprk Wald F	34.56

t-statistics in parentheses. *** p < 0.01, ** p < 0.05.

(Column (2)), the second-stage regression reveals a significant and negative coefficient of IndPeerESG at the 5% level, further affirming the robustness of the baseline findings.

4.3. Internal Moderators of Peer Effects in ESG Disclosure

To further unpack the dynamics behind the ESG peer effect, this section investigates the moderating roles of three theoretically motivated firm-level characteristics: information asymmetry (Asy), corporate reputation (Rep), and market competition (HHI). Rather than acting as mediators, these factors are conceptualized as moderators, that is, they influence the magnitude and conditions under which peer effects are more or less pronounced. The estimation follows the extended specification:

$$\begin{gathered} {\mathit{E}\mathit{S}\mathit{G}}_{\mathit{i},\mathit{j},\mathit{t}}={\mathsf{\alpha }}_{\mathbf{0}}+{\mathsf{\alpha }}_{\mathbf{1}}{\mathit{I}\mathit{n}\mathit{d}\mathit{p}\mathit{e}\mathit{e}\mathit{r}\mathit{E}\mathit{S}\mathit{G}}_{\mathit{i},\mathit{t}}+{\mathsf{\alpha }}_{\mathbf{2}}{\mathit{I}\mathit{n}\mathit{d}\mathit{p}\mathit{e}\mathit{e}\mathit{r}\mathit{E}\mathit{S}\mathit{G}}_{\mathit{i},\mathit{t}}\times \mathit{M}+{\mathsf{\alpha }}_{\mathbf{3}}\mathit{M}+ \\ {\mathsf{\alpha }}_{\mathbf{4}}{\mathit{C}\mathit{o}\mathit{n}\mathit{t}\mathit{r}\mathit{o}\mathit{l}\mathit{s}}_{\mathit{i},\mathit{j},\mathit{t}}+{\mathsf{\alpha }}_{\mathbf{5}}{\mathit{I}\mathit{n}\mathit{d}\mathit{p}\mathit{e}\mathit{e}\mathit{r}\mathit{C}\mathit{o}\mathit{n}\mathit{t}\mathit{r}\mathit{o}\mathit{l}}_{\mathit{i},\mathit{j},\mathit{t}}+\sum \mathit{Y}\mathit{e}\mathit{a}\mathit{r}+\sum \mathit{i}\mathit{d}+{\mathsf{\epsilon }}_{\mathit{i},\mathit{t}} \end{gathered}$$

(4)

where M ∈ {Asy, Rep, HHI}. A significant coefficient of the interaction term indicates that the corresponding variable moderates the peer effect in ESG disclosure.

Table 6. Internal moderation analysis of ESG peer effects.

Variable	(1)	(2)	(3)	(4)	(5)	(6)
	ESG	ESG	ESG	ESG	ESG	ESG
IndpeerESG	0.147 ***	0.139 ***	0.142 ***	0.148 ***	0.166 ***	0.156 ***
	(27.86)	(26.57)	(25.54)	(26.66)	(27.07)	(26.09)
IndpeerESG_Asy	0.061 ***	0.058 ***
	(5.17)	(5.03)
Asy	−0.152 ***	−0.035 ***
	(−36.11)	(−6.18)
IndpeerESG_Rep			0.015 ***	0.017 ***
			(2.93)	(3.40)
Rep			0.081 ***	0.056 ***
			(42.20)	(15.05)
IndpeerESG_HHI					−0.171 ***	−0.171 ***
					(−10.67)	(−11.06)
HHI					−0.092 ***	−0.142 ***
					(−6.67)	(−10.56)
Control variables	No	Yes	No	Yes	No	Yes
Indpeer_Control variables	No	Yes	No	Yes	No	Yes
Year (FE)	Yes	Yes	Yes	Yes	Yes	Yes
Firms (FE)	Yes	Yes	Yes	Yes	Yes	Yes
Constant	0.695 ***	−0.819 ***	0.760 ***	−0.249 ***	0.664 ***	−1.103 ***
	(31.80)	(−14.57)	(33.07)	(−3.36)	(25.88)	(−23.61)
Observations	32,181	32,177	28,658	28,655	32,187	32,183
R-squared	0.060	0.100	0.081	0.103	0.026	0.103
Number of years	12	12	12	12	12	12
r2_a	0.0600	0.0989	0.0801	0.102	0.0253	0.102
F	689.1	237.1	836.2	218.7	283.7	246.5

t-statistics in parentheses. *** p < 0.01.

reports the results. For information asymmetry (H2), Columns (1) and (2) show that the interaction term IndPeerESG × Asy is positive and significant at the 1% level (0.061 and 0.058, respectively), indicating that firms facing greater asymmetry are more likely to align their ESG disclosure with peers to mitigate uncertainty. For corporate reputation (H3), the interaction terms in Columns (3) and (4) are also positive and significant (0.015 and 0.017), consistent with signaling theory—firms mimic peer behavior to strengthen or safeguard intangible reputational capital. For market competition (H4), Columns (5) and (6) reveal that IndPeerESG × HHI is significantly negative (−0.171 at 1%), supporting the idea that peer effects are more potent in more competitive environments, where mimicking ESG behavior may serve strategic differentiation or legitimacy goals. These findings collectively provide evidence for the theoretical mechanisms proposed in Section 2.3 and reinforce that firm-specific characteristics and strategic considerations shape the intensity of ESG peer influence.

4.4. External Moderators of Peer Effects in ESG Disclosure

To complement the analysis of firm-level mechanisms, this section examines how external environmental factors, including economic policy uncertainty (EPU), business environment uncertainty (EU), and the strength of institutional environments (Market), moderate the peer effect in ESG disclosure. The following regression model is used:

$$\begin{gathered} {\mathit{E}\mathit{S}\mathit{G}}_{\mathit{i},\mathit{j},\mathit{t}}={\mathsf{\alpha }}_{\mathbf{0}}+{\mathsf{\alpha }}_{\mathbf{1}}{\mathit{I}\mathit{n}\mathit{d}\mathit{p}\mathit{e}\mathit{e}\mathit{r}\mathit{E}\mathit{S}\mathit{G}}_{\mathit{i},\mathit{t}}+{\mathsf{\alpha }}_{\mathbf{2}}{\mathit{I}\mathit{n}\mathit{d}\mathit{p}\mathit{e}\mathit{e}\mathit{r}\mathit{E}\mathit{S}\mathit{G}}_{\mathit{i},\mathit{t}}\times \mathit{N}+{\mathsf{\alpha }}_{\mathbf{3}}\mathit{N}+ \\ {\mathsf{\alpha }}_{\mathbf{4}}{\mathit{C}\mathit{o}\mathit{n}\mathit{t}\mathit{r}\mathit{o}\mathit{l}\mathit{s}}_{\mathit{i},\mathit{j},\mathit{t}}+{\mathsf{\alpha }}_{\mathbf{5}}{\mathit{I}\mathit{n}\mathit{d}\mathit{p}\mathit{e}\mathit{e}\mathit{r}\mathit{C}\mathit{o}\mathit{n}\mathit{t}\mathit{r}\mathit{o}\mathit{l}}_{\mathit{i},\mathit{j},\mathit{t}}+\sum \mathit{Y}\mathit{e}\mathit{a}\mathit{r}+\sum \mathit{i}\mathit{d}+{\mathsf{\epsilon }}_{\mathit{i},\mathit{t}} \end{gathered}$$

(5)

where $\mathit{N}$ denotes the external factors (EPU, EU, and Market). A statistically significant coefficient of the interaction term (${\mathit{I}\mathit{n}\mathit{d}\mathit{p}\mathit{e}\mathit{e}\mathit{r}\mathit{E}\mathit{S}\mathit{G}}_{\mathit{i},\mathit{t}}\times \mathit{N})$ indicates that the respective external factor influences the strength of the peer effect.

Table 7. External moderation analysis of ESG peer effects.

Variable	(1)	(2)	(3)	(4)	(5)	(6)
	ESG	ESG	ESG	ESG	ESG	ESG
IndpeerESG	0.148 ***	0.146 ***	0.143 ***	0.136 ***	0.143 ***	0.138 ***
	(27.64)	(28.20)	(27.12)	(26.82)	(26.64)	(26.53)
IndpeerESG_EPU	0.000 **	0.000 *
	(2.09)	(0.24)
EPU	−0.000 ***	−0.000 ***
	(−4.94)	(−9.95)
IndpeerESG_EU			0.097 ***	0.118 ***
			(2.72)	(3.42)
EU			−0.287 ***	−0.322 ***
			(−20.89)	(−21.33)
IndpeerESG_Market					0.006 **	0.006 **
					(2.15)	(2.05)
Market					0.005 ***	0.004 ***
					(3.93)	(3.15)
Control variables	No	Yes	No	Yes	No	Yes
Indpeer_Control variables	No	Yes	No	Yes	No	Yes
Year (FE)	Yes	Yes	Yes	Yes	Yes	Yes
Firms (FE)	Yes	Yes	Yes	Yes	Yes	Yes
Constant	0.745 ***	−0.948 ***	0.798 ***	−0.913 ***	0.699 ***	−1.080 ***
	(34.13)	(−21.24)	(36.64)	(−20.92)	(28.35)	(−22.52)
Observations	32,187	32,183	31,437	31,435	32,187	32,183
R-squared	0.024	0.098	0.036	0.113	0.022	0.098
Number of years	12	12	12	12	12	12
r2_a	0.0239	0.0974	0.0357	0.113	0.0220	0.0976
F	263.5	232.4	392.8	268.0	245.8	233.6

t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

presents the findings on the moderating effects of external factors on the ESG peer effect. Columns (1) and (2) reveal that economic policy uncertainty (EPU) significantly amplifies the ESG peer effect. The interaction term is positive and significant at the 1% level, suggesting that firms rely more on peer ESG disclosure practices when navigating uncertain macro-policy environments. This finding is consistent with information learning theory, where firms emulate peers to reduce strategic ambiguity under uncertainty. Columns (3) and (4) examine business environment uncertainty (EU). After controlling for firm-level variables, the interaction term remains positive and highly significant, indicating that firms are more inclined to align with peer behavior to signal stability and responsiveness under unstable operational conditions. Columns (5) and (6) explore the role of the institutional environment (Market). The results indicate that firms in more market-oriented regions are more sensitive to peer ESG behavior, as reflected by a positive and significant coefficient of the interaction term. This suggests that institutional support strengthens the normative and mimetic pressures that drive peer conformity in ESG practices. These findings collectively highlight that external conditions significantly shape the extent to which firms internalize peer behavior in ESG disclosure. They reinforce the theoretical claim that the peer effect is not uniform across environments but varies with the degree of institutional maturity and market uncertainty.

5. Heterogeneity in ESG Peer Effects Across Organizational and Contexts

To further explore the dynamics of peer effects in ESG disclosure, this section conducts a heterogeneity analysis based on key firm-level and regional characteristics. Specifically, four dimensions are examined: ownership structure, geographical region, firm size, and corporate life cycle stage. These factors are theorized to shape how firms respond to peer influence by conditioning their institutional exposure, resource constraints, and strategic priorities.

Table 8. Heterogeneity analysis methodology.

Dimension	Grouping	Grouping Criteria
Ownership Structure	SOEs vs. non-SOEs	Based on whether firms are classified as state-owned or privately held
Geographic Region	Eastern, Central, Western	Based on standard provincial classifications reflecting economic development levels
Firm Size	Large vs. Small Firms	Firms are categorized by whether total assets exceed the annual industry median
Life Cycle Stage	Growth, Maturity, Decline Phases	Determined by a composite index of sales growth, retained earnings, and capital investment, following Dickinson (2011) [71]

outlines the classification criteria used for subgroup analysis. The regression model (2) constructed in Section 3.3 is estimated separately for each subsample. The results are reported in Appendix A 

Table A2. Heterogeneity test.

Variable	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
	ESG (Soe)	ESG (Private)	ESG (Growth)	ESG (Mature)	ESG (Recession)	ESG (East)	ESG (Mid)	ESG (West)	ESG (Big)	ESG (Small)
IndpeerESG	0.155 ***	0.125 ***	0.141 ***	0.139 ***	0.145 ***	0.146 ***	0.153 ***	0.090 ***	0.143 ***	0.138 ***
	(23.47)	(18.84)	(19.68)	(17.81)	(13.11)	(26.00)	(11.02)	(7.45)	(22.74)	(19.08)
Control variables	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Indpeer_Control variables	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year (FE)	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Firms (FE)	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Constant	0.878 ***	1.066 ***	0.661 ***	0.816 ***	0.784 **	0.485 **	1.321 ***	0.286	0.257	1.655 ***
	(4.21)	(5.87)	(3.42)	(3.69)	(2.23)	(2.05)	(4.04)	(0.89)	(1.43)	(7.83)
Observations	11,129	21,054	14,770	11,288	6019	22,899	4233	5051	17,141	15,042
R-squared	0.161	0.125	0.112	0.128	0.168	0.136	0.137	0.112	0.122	0.141
r2_a	0.158	0.123	0.110	0.126	0.164	0.135	0.131	0.107	0.121	0.139
F	111.8	157.6	97.55	87.27	63.82	190.0	35.16	33.32	125.6	129.5

t-statistics in parentheses. *** p < 0.01, ** p < 0.05.

, and a comparative summary is illustrated in Figure 2.

Across all subgroups, IndPeerESG remains positively associated with firm-level ESG disclosure at the 1% significance level, confirming the robustness of the peer effect. However, notable differences emerge across organizational and regional characteristics.

State-owned enterprises (SOEs) exhibit more substantial peer effects than non-SOEs, likely due to stricter regulatory oversight and institutional pressures that reinforce alignment with policy-driven ESG norms. In contrast, non-SOEs, operating under looser disclosure mandates, demonstrate weaker conformity. By region, firms in the central region show the most pronounced peer effects, followed by those in the western and eastern regions. The central region’s evolving institutional environment may increase reliance on peer signals, whereas the eastern region’s advanced market systems reduce such dependence. In terms of firm size, smaller firms demonstrate more substantial peer effects, possibly due to constrained internal capabilities and the heightened need for external legitimacy. Larger firms tend to act as industry benchmarks and rely less on peer imitation. Finally, firms in the growth stage exhibit the most substantial peer effects, consistent with greater information asymmetry and agency concerns. These firms mimic their peers to enhance market credibility. Conversely, mature and declining firms, with more established systems, show reduced peer reliance.

Figure 2 provides a comparative visual representation of the ESG peer effect across the heterogeneity dimensions. The findings confirm significant variations in the peer effect based on property rights, regional contexts, firm size, and life cycle stage. These results underscore the importance of contextual factors in shaping firms’ ESG disclosure practices.

6. Conclusions and Discussions

6.1. Key Findings and Contributions

This study investigates the peer effects in corporate ESG disclosure using panel data of Chinese A-share listed firms from 2010 to 2021. By employing rigorous econometric techniques—most notably, instrumental variable (IV) estimation based on stock-specific idiosyncratic returns—our analysis provides robust evidence that firms adjust their ESG disclosure behaviors in response to those of industry peers. These peer effects remain consistent across various model specifications and alternative peer classifications, including ESG sub-dimension clustering and third-party ESG ratings from Bloomberg.

Beyond identifying peer effects, we uncover critical firm-level and contextual factors that moderate their intensity. Specifically, firms with higher information asymmetry, stronger reputational concerns, and more intense market competition are more susceptible to peer influence. Externally, economic policy uncertainty, institutional conditions, and volatility in the business environment further reinforce peer-driven ESG convergence. The effect is particularly pronounced among state-owned enterprises, firms in central China, smaller firms, and those in earlier stages of the corporate lifecycle, underscoring the context-specific nature of ESG diffusion in transitional economies.

This research contributes to the ESG literature in several ways. First, it enriches understanding of ESG disclosure behavior in the Chinese context—an increasingly essential yet underexplored setting [72,73,74]. Second, it addresses recent scholarly calls for more nuanced definitions of peer firms by validating industry-based classifications through ESG risk-based clustering. Third, it mitigates endogeneity concerns by employing an IV strategy grounded in the Carhart four-factor model, enhancing the credibility of the peer effect estimations. Methodologically, this study improves upon prior research by separating peer behavior from firm-specific ESG decisions, controlling for time-invariant unobservables through firm, year, and industry-by-year fixed effects, and rigorously constructing peer variables. Moreover, integrating moderating mechanisms rooted in information economics, signaling theory, and institutional theory provides a multi-level analytical framework linking firm strategy with broader institutional structures [75].

Beyond theoretical insights, this study offers practical implications. Regulators are encouraged to institutionalize ESG benchmarking tools and promote industry-wide ESG disclosure dashboards to strengthen peer-driven behavioral alignment. Stock exchanges may pilot sector-specific ESG ranking portals to enhance visibility and activate mimetic pressures. International standard-setting organizations such as ISSB, UN PRI, and GRI can facilitate cross-market convergence by aligning disclosure taxonomies and promoting shared knowledge platforms across emerging economies. At the firm level, managers and ESG officers should view peer alignment as reputationally strategic and potentially prone to herd behavior. Therefore, firms are advised to complement peer-driven learning with internal materiality assessments to ensure the authenticity of ESG disclosure.

6.2. Limitations and Future Research

As ESG regulations and market expectations evolve, the strength and mechanisms of peer effects may shift. In the early stages of ESG institutionalization, peer influence may be more prominent due to regulatory ambiguity; however, as mandatory disclosure requirements mature, formal compliance mechanisms could partially displace informal peer benchmarking. Future research could examine these temporal dynamics by comparing stages of regulatory development or using policy shocks as quasi-natural experiments. Furthermore, this study assumes a unidirectional influence from peers to focal firms. High-performing ESG firms may also shape peer behaviors through feedback loops or co-evolutionary dynamics. Investigating these reciprocal relationships may require network-based methods or structural equation modeling.

While peer-driven alignment can promote convergence and learning, it also poses risks. Firms may mimic ESG behaviors symbolically—without genuine sustainability commitments—thus increasing the likelihood of greenwashing and stakeholder misperception. These dynamics could ultimately undermine the credibility and effectiveness of ESG disclosure, particularly in less-regulated environments. Future research should explore the conditions under which peer effects foster substantive ESG practices rather than superficial conformity.

Although this study focuses on Chinese A-share firms, the underlying mechanisms—peer imitation under uncertainty, signaling via reputational channels, and stakeholder-driven alignment—apply theoretically to other emerging or transitional markets facing institutional ambiguity. Future studies could replicate this framework in different country contexts (e.g., India, Brazil, South Africa), examine cross-listed firms operating under dual disclosure regimes, or utilize high-frequency ESG disclosure data to capture evolving peer network dynamics. Researchers may further elucidate how peer-based incentives can be leveraged to promote credible, effective ESG practices globally.