Does the Market Value Corporate ESG Ratings? A Complex System Driven by Institutional Investors

Abstract

Against the backdrop of China’s dual-carbon goals and the growing emphasis on sustainable development, ESG information has become an important non-financial signal in capital markets; yet whether and how it is priced by investors remains unclear. Using a sample of 2018–2024 Chinese A-share listed firms, this study examines the relationship between corporate ESG ratings and firm market value, with a particular focus on the mediating role of institutional ownership and investor heterogeneity. We find that firms with higher ESG ratings exhibit significantly higher market value, indicating that the market assigns a valuation premium to favorable ESG evaluations. Mediation analyses further show that higher ESG ratings are associated with increased institutional ownership, which in turn enhances firm value. Heterogeneity analyses reveal that this mediating effect is primarily driven by long-term institutional investors, whereas medium-term and short-term institutions neither respond systematically to ESG ratings nor transmit ESG rating information into firm valuation. In additional analyses, we show that ESG rating divergence significantly weakens the positive valuation effect of ESG ratings by increasing informational uncertainty and reducing the credibility of ESG rating signals. Overall, this study provides new evidence on the investor-based mechanisms underlying ESG rating-based pricing and highlights the importance of improving the transparency and comparability of ESG ratings in China’s capital market.

1. Introduction

As concerns over climate change intensify, social responsibility standards deepen [1], and corporate governance practices continue to evolve, Environmental, Social, and Governance (ESG) considerations have moved from the periphery of corporate responsibility debates to the center of capital market analysis [2,3,4]. Since the United Nations Global Compact and the Principles for Responsible Investment formally advocated the integration of ESG factors into investment decision-making and risk assessment, ESG has increasingly reshaped how investors evaluate firms’ long-term sustainability, risk exposure, and value creation potential [5,6,7]. Compared with traditional financial indicators, ESG rating information captures a broader set of non-financial attributes related to environmental stewardship, stakeholder engagement, and governance quality [8], and is therefore widely regarded as an important signal of firms’ long-term prospects.

From the perspectives of information economics and asset pricing theory, ESG ratings, as an important dimension of firms’ non-financial characteristics, can systematically enhance firm market value [9,10,11,12]. First, ESG ratings reflect a firm’s strategy, risk management, and governance, raise market expectations about the level and sustainability of future cash flows, and thus increase firm valuation. Second, high ESG ratings signal investors’ assessments of a firm’s ability to manage regulatory, reputational, and operational risks [13], thereby reducing the likelihood of severe adverse events and improving the firm’s risk profile under conditions of uncertainty. Within an asset pricing framework, this translates into lower required risk premia. Third, high-quality governance structures enhance resource allocation efficiency, mitigate agency problems, and promote consistency and foresight in corporate decision-making, strengthening the contribution of operating activities to long-term value creation [14,15]. Taken together, ESG ratings summarize both contemporaneous and forward-looking assessments of firms’ non-financial attributes but also forward-looking information relevant to their risk–return characteristics and long-term value potential.

In this context, institutional investors have increasingly been recognized as key market participants in the ESG pricing process [3,16,17]. In the Chinese context, institutional investor types have been rapidly diversifying—including state-owned asset managers, insurance and pension funds, mutual funds, securities-asset managers, and foreign institutions—while state ownership, policy-driven mandates, and the historically large presence of retail investors continue to exert important influences on market behavior and investor objectives. Institutional investors, with their scale advantages, professional expertise, and governance engagement, are capable of processing non-financial information and shaping firm valuation. However, prior studies often treat them as homogeneous and rely on aggregate ownership measures, overlooking differences in investment horizons, incentive structures, and governance objectives [18,19]. This omission is particularly noteworthy in China, where state-owned and private ownership backgrounds, policy-driven investment objectives, and market-oriented motives coexist—meaning that the same “institutional investor” label can conceal markedly different behaviors and policy sensitivities. Investor horizon theory suggests that institutional investors with different investment horizons differ fundamentally in their objective functions and behavioral constraints [20,21]. Long-term-oriented institutions place greater emphasis on sustainable cash flows and intertemporal value maximization, making their investment logic more closely aligned with the long-term risk management and value creation attributes reflected in ESG ratings. In contrast, short-term-oriented institutions are more sensitive to near-term returns and trading opportunities; their higher portfolio turnover and short-term incentives may weaken the transmission of ESG rating signals into firm valuation. Although these distinctions are well established in the literature on investor behavior and corporate governance, systematically incorporating investor horizon heterogeneity into ESG pricing research—and empirically testing the mediating roles of different types of institutional investors in the ESG rating–market value relationship—remains an important gap.

Beyond investor heterogeneity, the pricing of ESG information also critically depends on the quality and consistency of ESG measurement. In the Chinese context, the coexistence of international and domestic rating agencies, large variation in local disclosure practices and industry norms, and differing regulatory pacing in disclosure promotion can all amplify or alter how rating divergence affects market valuation associations. The absence of unified ESG standards leads to substantial rating divergence across agencies, and the narrative tone of ESG reports further shapes rating consistency, with positive tone amplifying rating dispersion under low disclosure credibility, while negative tone enhances rating convergence [22]. From a theoretical standpoint, such rating divergence undermines the consistency and credibility of ESG signals, increases investors’ information processing costs, and may impair the incorporation of ESG information into asset prices. While prior studies have documented the existence and statistical properties of ESG rating divergence [23,24,25,26], systematic evidence on whether—and through what channels—rating divergence weakens the market pricing of ESG ratings remains limited [27].

Motivated by these considerations, this study examines whether and how corporate ESG ratings are priced in China’s capital market. Using panel data on Chinese A-share listed firms from 2018 to 2024, we analyze the relationship between ESG ratings and firm market value, with a particular focus on the channels through which ESG information is transmitted into prices. Rather than relying solely on reduced-form correlations, we explicitly incorporate institutional investors into the analytical framework and examine their mediating role in the ESG-valuation relationship. Recognizing substantial heterogeneity in investment horizons, incentives, and governance engagement, we distinguish between long-term, medium-term and short-term institutional investors and examine their respective roles in transmitting ESG information into firm value. This approach characterizes ESG pricing as the outcome of interactions among heterogeneous institutional investors, rather than the behavior of a representative investor. In addition, we investigate the role of ESG rating divergence in shaping the valuation effects of ESG information. From an information-consistency perspective, we examine whether disagreement among rating agencies weakens ESG pricing by increasing uncertainty and reducing signal credibility, thereby highlighting the importance of evaluation consistency for the effectiveness of ESG ratings in capital markets.

This study makes three main contributions within the institutional context of China’s capital market. First, prior research has largely overlooked the role of institutional investors in transmitting ESG information [28]; we examine institutional investors as an important structural factor associated with the pricing of ESG information and reveal the structural complexity of ESG pricing by accounting for investor horizon heterogeneity. Second, existing studies often focus on average effects and pay limited attention to rating divergence [29,30]; we provide empirical evidence that ESG rating divergence is negatively associated with the market valuation effect of ESG ratings, highlighting the importance of consistency and comparability in ESG evaluation systems. Third, the findings offer policy-relevant insights into improving ESG assessment practices, fostering long-term institutional investor participation, and enhancing the informational value of ESG ratings, thereby contributing to the high-quality development of China’s capital market.

The remainder of the paper is organized as follows. Section 2 reviews the related literature and develops the research hypotheses. Section 3 describes the data sources, sample construction, and empirical models. Section 4, Section 5, Section 6 present the main empirical results, robustness checks, and further analyses. Section 7 concludes with a summary of the main findings and a discussion of their practical and policy implications.

2. Hypothesis Development

2.1. The Association Between ESG Ratings and Market Outcomes for Listed Companies

Existing research on ESG has generally recognized that ESG information has become an important dimension of non-financial information in capital markets [31]. Its impact is not confined to short-term stock price fluctuations but operates through multiple channels to influence firms’ long-term market valuation. Compared with stock returns over a single period, firm market value more comprehensively reflects investors’ expectations regarding future cash flows, risk exposure, and sustainable development capacity. Accordingly, examining the economic consequences of ESG ratings from the perspective of firm market value is of greater theoretical and practical relevance.

According to signaling theory [32], under conditions of information asymmetry, corporate managers can convey credible signals of firm quality to capital markets by voluntarily disclosing information that is costly and difficult to imitate. Third-party ESG ratings, which reflect firms’ ESG-related performance, constitute an important form of such signals. On the one hand, high ESG ratings typically reflect firms’ sustained investments in environmental protection, social responsibility, and corporate governance. These investments are long-term, systematic, and largely irreversible, making them difficult for low-quality firms to replicate through short-term window dressing. On the other hand, ESG ratings issued by professional agencies are more standardized, comparable, and credible than ESG reports voluntarily disclosed by firms themselves [33,34,35]. High-quality ESG information disclosure helps reduce investors’ information acquisition and processing costs, alleviates information asymmetry between firms and capital markets, and thereby improves stock price efficiency. Existing studies show that enhanced transparency reduces the likelihood of mispricing and lowers the risk of extreme negative events, such as stock price crashes [10,36,37]. When markets are better able to identify firms’ long-term operating quality through ESG ratings and governance standards, firms’ stocks are more likely to command valuation premiums, leading to higher market value.

Beyond the information channel, ESG ratings also affect firm market value by influencing corporate risk characteristics and the cost of capital [38]. High ESG ratings are generally associated with greater robustness in environmental compliance, labor relations, product safety, and internal governance, which helps mitigate environmental penalty risks, social conflict risks, and internal agency conflicts [39,40,41]. A reduction in these risks is directly reflected in investors’ assessments of the uncertainty surrounding firms’ future cash flows. From an asset-pricing perspective, firm market value represents the present value of expected future cash flows adjusted for risk. When higher ESG ratings signal lower perceived systematic and idiosyncratic risk, the risk premium required by investors—that is, the cost of capital—declines, thereby increasing market valuation for a given level of expected cash flows. Moreover, prior studies indicate that firms with higher ESG ratings face fewer financing constraints and can obtain equity and debt financing more easily, which further reinforces the role of ESG ratings in enhancing firm market value through the cost-of-capital channel [35,39].

In capital markets, ESG ratings have gradually evolved into reference indicators analogous to credit ratings and are widely used in asset allocation, index construction, and portfolio screening. For investors, ESG ratings not only reflect firms’ current compliance status but also serve as comprehensive signals of long-term sustainable operating capacity [39]. Firms with higher ESG ratings are more likely to be perceived as high-quality firms with sound governance structures and strong long-term value creation potential, thereby elevating investors’ expectations of future cash-flow growth. Although some studies suggest that investor sentiment may overreact to ESG information [42,43,44,45], the fundamental effects of ESG information tend to dominate in long-term market outcomes as captured by firm market value.

In the context of China’s capital market, the association between ESG ratings and firm market value is embedded in a distinctive institutional setting [46]. China’s market has long been characterized by relatively pronounced information asymmetry, a high proportion of retail investors, and strong sensitivity to market sentiment [47]. Under such conditions, standardized and comparable third-party ESG ratings can serve as salient reference indicators, potentially carrying higher marginal informational value than in more mature markets. At the same time, China’s policy-oriented institutional framework closely links corporate ESG performance with regulatory evaluation and resource allocation. With the advancement of the “dual-carbon” goals and high-quality development strategy, ESG-related criteria have been increasingly incorporated into supervisory and financial assessment systems. In this environment, ESG ratings function not only as market signals but also as institutionally embedded indicators associated with firms’ access to financing and strategic resources.

In summary, ESG ratings information exerts a systematic influence on the market valuation of listed firms by reducing information asymmetry, mitigating operational and governance risks, lowering the cost of capital, and improving investors’ expectations of firms’ long-term value. Accordingly, this study proposes the following hypothesis:

H1a. 

Firms with higher ESG ratings exhibit higher market value.

Meanwhile, some studies offer alternative perspectives on the valuation effects of ESG ratings from the viewpoints of cost burden and signal noise. First, ESG-related investments are often accompanied by multiple costs, including environmental governance, employee welfare, compliance monitoring, information disclosure, and governance restructuring. In the short term, these expenditures may compress firms’ profit margins and exert downward pressure on market valuation by reducing distributable cash flows. This effect may be particularly pronounced for firms operating in highly competitive environments, with low profit margins, or facing severe financing constraints [48]. For such firms, improvements in ESG ratings do not necessarily imply simultaneous gains in operational efficiency; instead, they may be interpreted by the market as a “high-cost signal,” whereby firms obtain favorable external evaluations through increased non-core expenditures, without necessarily generating substantive value creation.

Second, ESG ratings may be subject to signal noise. Differences across rating agencies in terms of evaluation criteria, weighting schemes, and information processing methods imply that ESG ratings may not consistently or accurately reflect firms’ true sustainable operating capacity. Under conditions of incomplete information and limited investor cognition, market participants may treat ESG ratings more as superficial reputation indicators rather than credible signals of future cash flows [49]. Moreover, in periods of heightened market attention to ESG, some firms may engage in selective disclosure, symbolic governance practices, or “greenwashing” behaviors to inflate their ratings, leading to a potential divergence between ESG ratings and underlying firm value [50].

Based on the above arguments, the impact of ESG ratings on firm market value may not exhibit a stable positive relationship; instead, it may be insignificant or even negative due to cost pressures, rating biases, or market overreaction. Accordingly, this study proposes the following hypothesis:

H1b. 

Firms with higher ESG ratings exhibit lower market value.

2.2. The Mediating Role of Institutional Investors

In recent years, with the continued development of China’s capital market, the scale of institutional investors has expanded steadily, and their role in firm valuation formation and capital allocation has become increasingly prominent [51]. Compared with individual investors, institutional investors typically operate with longer investment horizons, more sophisticated risk management systems, and stronger capabilities in information collection and processing [52]. As a result, they tend to exert a more stable and rational influence on asset pricing. Prior studies indicate that institutional ownership not only improves market pricing efficiency but also reduces stock price volatility through a “stabilizing” effect, thereby exerting a significant impact on firm market value.

From the perspectives of corporate governance and market pricing, increases in institutional ownership are generally associated with higher firm valuation. On the one hand, institutional investors can exert effective monitoring over corporate managers by virtue of their professional expertise and scale advantages, alleviating agency problems and curbing managerial short-termism [53]. This, in turn, improves firms’ long-term operating performance and enhances market expectations regarding future cash flows [54]. On the other hand, institutional investors are widely regarded as representatives of rational investors in capital markets, and their investment decisions carry demonstrative effects and informational content. When institutional investors increase their holdings in a firm, the market is more likely to interpret such actions as a positive signal of the firm’s long-term value, thereby contributing to an increase in firm market value. Accordingly, institutional investors are commonly viewed as an important bridge linking firm fundamentals to market valuation.

Existing research shows that, relative to individual investors, institutional investors place greater emphasis on non-financial information that reflects firms’ long-term sustainability [55]. With the growing prevalence of sustainable investing, ESG ratings have gradually become an important criterion for institutional investors in assessing firms’ long-term risk exposure and value potential [56]. ESG ratings summarize firms’ fulfillment of environmental and social responsibilities but also systematically reflects their corporate governance quality and risk management capability, and thus contains substantial informational content for institutional investors’ decision-making.

The influence of ESG ratings on institutional investors’ allocation decisions is mainly manifested in two aspects. First, according to signaling theory, under conditions of information asymmetry, firms can convey their underlying quality to investors through observable external indicators. ESG ratings, as a comprehensive reflection of firms’ governance standards, risk control capability, and responsibility orientation, constitutes a credible signal of long-term firm quality. Given their greater emphasis on firm fundamentals and long-term development prospects rather than short-term financial fluctuations, institutional investors are therefore more inclined to allocate capital to firms with higher ESG ratings [57]. Second, from the perspective of long-term value investing and risk management, rational investors jointly consider expected returns and risk characteristics when making asset allocation decisions. Prior studies find that strong ESG ratings signal lower exposure to ESG-related tail risks, enhances operational resilience under conditions of heightened uncertainty, and weakens the adverse impact of negative events on firm value [58,59]. For institutional investors whose primary objectives focus on portfolio stability and downside risk control, firms with high ESG ratings are more consistent with their long-term investment preferences, as they contribute to more stable risk-adjusted returns and greater portfolio robustness [60,61].

In sum, institutional investors exert a significant influence on firm market value by improving corporate governance, enhancing pricing efficiency, and transmitting value-related signals. At the same time, ESG rating information, as an important indicator of firms’ long-term quality and risk characteristics, plays a key role in shaping institutional investors’ ownership decisions. Based on this reasoning, we propose the following hypothesis:

H2. 

Corporate ESG ratings positively affect firm market value through increasing institutional ownership.

Long-term-oriented institutional investors typically pursue stable shareholding strategies and emphasize sustainable value creation over extended investment horizons [62]. From a theoretical standpoint, these investors adopt long-term valuation frameworks that prioritize firms’ future cash-flow sustainability, risk exposure, and governance quality. ESG ratings, which encapsulate firms’ capabilities in managing environmental risks, maintaining stakeholder relationships, and establishing effective governance structures, therefore constitute a particularly relevant source of information for long-term investors. Compared with short-term financial fluctuations, ESG-related signals are more closely aligned with the objectives of long-term institutional investors and are thus more likely to be incorporated into their investment decisions [63].

The persistent presence of long-term institutional investors is expected to strengthen the market’s pricing of ESG information [39]. Their relatively low portfolio turnover reduces short-term trading noise, allowing ESG-related signals to be reflected more fully in long-term firm valuation. Moreover, long-term institutional investors generally exhibit stronger incentives to engage in corporate governance and to support strategic decisions that enhance firms’ long-run performance, even when such decisions involve short-term costs [64]. In the ESG context, this implies greater tolerance for investments such as environmental innovation or governance reforms that generate delayed but durable benefits [65]. As a result, the involvement of long-term institutional investors enhances the credibility of ESG signals and reinforces the linkage between ESG ratings and firms’ future cash flows, thereby amplifying the positive valuation effects associated with high ESG ratings [63].

In contrast, medium-term and short-term-oriented institutional investors are primarily driven by near-term performance metrics and trading opportunities, with investment decisions that are highly sensitive to short-term information and market sentiment [66,67]. While such investors contribute to market liquidity and facilitate rapid price discovery, their high portfolio turnover may dilute the influence of long-term information in asset prices. From a theoretical perspective, the short investment horizon of these institutions implies a weaker alignment between ESG ratings—whose benefits typically materialize over longer periods—and short-term valuation considerations.

Furthermore, medium-term and short-term institutional investors may implicitly impose short-horizon performance pressures on corporate managers, encouraging a focus on immediate financial outcomes rather than long-term value creation [68,69]. This tendency can reduce firms’ incentives to undertake ESG-related investments with longer payback periods and may weaken the expected association between ESG ratings and future firm value. In addition, frequent trading by short-term institutions can introduce noise and increase price volatility, further obscuring the informational content of ESG ratings. Consequently, medium-term and short-term institutional ownership is unlikely to strengthen—and may even attenuate—the positive market valuation effects of high ESG ratings.

Taken together, differences in investment horizons generate systematic heterogeneity in how institutional investors interpret and transmit ESG information to market prices. Accordingly, we propose the following hypothesis:

H3. 

Long-term-oriented institutional investors exert a stronger positive mediating effect in the relationship between ESG ratings and firms’ market performance, whereas medium-term and short-term-oriented institutional investors exert a negative mediating effect.

3. Research Design

3.1. Sample Selection and Data Source

This study uses A-share listed firms in China over the period 2018–2024 as the research sample to systematically examine the impact of ESG ratings on corporate market performance and further analyze the mediating role of institutional investment. We restrict the sample to 2018–2024 because the Sino-Securities ESG ratings adopted in this paper have been publicly available since 2018; ESG scores for earlier years (2009–2017) reported in some databases are retroactively constructed and therefore are excluded from the baseline analysis to ensure that the information set reflects what was observable to investors in real time. Firm-level ESG rating data are obtained from the Sino-Securities ESG rating system provided by Sino-Securities Index Information Service (Shanghai, China) Co., Ltd. The Sino-Securities ratings are released on a quarterly basis. For the baseline analysis, we construct an annual ESG measure for each firm and year by taking the arithmetic average of all quarterly ratings available within the same calendar year. All empirical analyses are implemented using Stata (v17.0) software.

Firm financial information, ownership structure, and other control variables are collected from the CSMAR database. To ensure sample validity and consistency, the following screening procedures are applied: (1) financial firms are excluded; (2) firms designated as ST or *ST are removed; (3) observations with missing values for key variables are deleted; (4) all continuous variables are winsorized at the 1% level. After these procedures, the final sample consists of 3948 listed firms and 19,119 firm-year observations.

3.2. Variable Definition

3.2.1. Dependent Variable

This study uses Market Value (MarV) to represent firms’ market performance. Market capitalization reflects the aggregate valuation assigned by all market participants and thus encapsulates investors’ overall assessment of firm value. It incorporates not only information about current operating performance but also investors’ expectations regarding future cash flows, growth potential, and risk exposure. Compared with performance measures based on historical accounting data, market value is more forward-looking in nature and can more directly capture the capital market’s evaluation of a firm’s long-term value. In the context of ESG research, ESG-related investments are typically characterized by long-term horizons and uncertainty, and their economic consequences may not be immediately observable in short-term accounting outcomes. Instead, such effects are more likely to be reflected through investors’ expectations and asset-pricing mechanisms in firms’ market valuations. Accordingly, MarV provides an appropriate and effective measure for assessing how ESG information is priced in the capital market.

3.2.2. Independent Variable

As the core explanatory variable, this study employs the ESG composite score released by Sino-Securities Index Information Service (Shanghai) Co., Ltd. (hereafter Sino-Securities ESG score) to measure firms’ ESG ratings. Compared with other ESG evaluation systems, the Sino-Securities ESG framework offers broader coverage and stronger institutional suitability for China’s capital market. It covers nearly the entire population of Chinese A-share listed firms, which helps ensure sample consistency and effectively mitigates potential sample selection bias arising from limited firm coverage. In addition, the long time span and stable updating mechanism of the Sino-Securities ESG data enable a systematic examination of the relationship between ESG ratings and market valuation in a dynamic setting. More importantly, the Sino-Securities ESG evaluation system is constructed under a localized yet internationally comparable framework. While incorporating the core structure of mainstream international ESG standards, it excludes certain indicators that are less compatible with China’s institutional environment or subject to severe data limitations. At the same time, it integrates ESG rating indicators with distinct Chinese characteristics, such as measures related to rural revitalization initiatives and administrative penalties imposed by the China Securities Regulatory Commission (CSRC). This design enhances the economic relevance and interpretability of ESG signals for domestic investors and regulators.

In terms of measurement, the Sino-Securities ESG framework adopts a multi-level indicator system to comprehensively assess firms’ performance across the environmental (E), social (S), and governance (G) dimensions, and ultimately generates a composite ESG score ranging from 0 to 100. Higher scores indicate more favorable ESG evaluations. Compared with discrete rating grades, the continuous ESG composite score more fully captures both cross-sectional variation and intertemporal dynamics in firms’ ESG rating outcomes, thereby improving its suitability for empirical analyses of ESG pricing effects in China’s capital market.

3.2.3. Mediating Variable

Institutional investor ownership (Inst) is defined as the proportion of shares held by institutional investors relative to a firm’s free-float shares during the reporting period.

$${\mathrm{I}\mathrm{n}\mathrm{s}\mathrm{t}}_{i,t}={\displaystyle \frac{\sum _{\mathrm{k}=1}^{{\mathrm{K}}_{\mathrm{i},\mathrm{t}}}{\mathrm{S}}_{\mathrm{k},\mathrm{i},\mathrm{t}}}{{\mathrm{F}}_{\mathrm{i},\mathrm{t}}}}$$

(1)

Specifically, ${\mathrm{S}}_{\mathrm{k},\mathrm{i},\mathrm{t}}$ denotes the number of shares held by institutional investor k in firm i at time t. ${\mathrm{K}}_{\mathrm{i},\mathrm{t}}$ represents the total number of institutional investors holding shares of firm i in period t, and ${\mathrm{F}}_{\mathrm{i},\mathrm{t}}$ denotes the free-float shares of firm i at time t.

Free-float shares capture the number of potentially tradable shares and thus constitute a key factor affecting the frequency of share transactions, changes in ownership structure, and the effective influence of institutional investors. Using total shares as the denominator may underestimate the role of tradable shares in shaping institutional ownership and market interactions, thereby obscuring the relationship between institutional investor behavior, corporate governance, and price discovery. Therefore, this study adopts free-float shares as the normalization benchmark to improve the economic interpretability of the measure.

To capture heterogeneity in institutional investors’ investment styles, this study classifies institutional investors based on their turnover rates [70,71]. For a given institutional investor k in semiannual period t, we first calculate the cumulative market value of stocks purchased and sold across all firms held in the sample. Let ${\mathrm{P}}_{\mathrm{i},\mathrm{t}}$ and ${\mathrm{P}}_{\mathrm{i},\mathrm{t}-1}$ denote the stock price of firm i in periods t and t − 1, respectively; ${\mathrm{S}}_{\mathrm{k},\mathrm{i},\mathrm{t}}$ and ${\mathrm{S}}_{\mathrm{k},\mathrm{i},\mathrm{t}-1}$ denote the number of shares held by institutional investor k in firm i in periods t and t − 1; and $\mathsf{\Delta }{\mathrm{P}}_{\mathrm{i},\mathrm{t}}$ = ${\mathrm{P}}_{\mathrm{i},\mathrm{t}}-{\mathrm{P}}_{\mathrm{i},\mathrm{t}-1}$. The cumulative market value of purchases and sales is calculated as follows:

$${CRbuy}_{\mathrm{k},\mathrm{t}}=\sum _{\mathrm{i}=1}^{{\mathrm{N}}_{\mathrm{k}}}\left|{\mathrm{S}}_{\mathrm{k},\mathrm{i},\mathrm{t}}{\mathrm{P}}_{\mathrm{i},\mathrm{t}}-{\mathrm{S}}_{\mathrm{k},\mathrm{i},\mathrm{t}-1}{\mathrm{P}}_{\mathrm{i},\mathrm{t}-1}-{\mathrm{S}}_{\mathrm{k},\mathrm{i},\mathrm{t}-1}\mathsf{\Delta }{\mathrm{P}}_{\mathrm{i},\mathrm{t}}\right|\left({\mathrm{S}}_{\mathrm{k},\mathrm{i},\mathrm{t}}\ge {\mathrm{S}}_{\mathrm{k},\mathrm{i},\mathrm{t}-1}\right)$$

(2)

$${CRsell}_{\mathrm{k},\mathrm{t}}=\sum _{\mathrm{i}=1}^{{\mathrm{N}}_{\mathrm{k}}}\left|{\mathrm{S}}_{\mathrm{k},\mathrm{i},\mathrm{t}}{\mathrm{P}}_{\mathrm{i},\mathrm{t}}-{\mathrm{S}}_{\mathrm{k},\mathrm{i},\mathrm{t}-1}{\mathrm{P}}_{\mathrm{i},\mathrm{t}-1}-{\mathrm{S}}_{\mathrm{k},\mathrm{i},\mathrm{t}-1}\mathsf{\Delta }{\mathrm{P}}_{\mathrm{i},\mathrm{t}}\right|\left({\mathrm{S}}_{\mathrm{k},\mathrm{i},\mathrm{t}}\ge {\mathrm{S}}_{\mathrm{k},\mathrm{i},\mathrm{t}-1}\right)$$

(3)

${\mathrm{N}}_{\mathrm{k}}$ denotes the number of listed firms held by institutional investor k in period t that are included in the sample. The turnover rate of institutional investor k in period t is defined as:

$${\mathrm{C}\mathrm{R}}_{k,t}={\displaystyle \frac{2\min \left({\mathrm{C}\mathrm{R}\mathrm{b}\mathrm{u}\mathrm{y}}_{\mathrm{k},\mathrm{t}},{\mathrm{C}\mathrm{R}\mathrm{s}\mathrm{e}\mathrm{l}\mathrm{l}}_{\mathrm{k},\mathrm{t}}\right)}{\sum _{\mathrm{i}=1}^{{\mathrm{N}}_{\mathrm{k}}}({\mathrm{S}}_{\mathrm{k},\mathrm{i},\mathrm{t}}{\mathrm{P}}_{\mathrm{i},\mathrm{t}}+{\mathrm{S}}_{\mathrm{k},\mathrm{i},\mathrm{t}}{\mathrm{P}}_{\mathrm{i},\mathrm{t}-1})}}$$

(4)

The numerator adopts $2\min \left({\mathrm{C}\mathrm{R}\mathrm{b}\mathrm{u}\mathrm{y}}_{\mathrm{k},\mathrm{t}},{\mathrm{C}\mathrm{R}\mathrm{s}\mathrm{e}\mathrm{l}\mathrm{l}}_{\mathrm{k},\mathrm{t}}\right)$ to mitigate the distortion caused by one-sided large transactions, while the denominator uses the sum of beginning- and end-of-period portfolio values to normalize trading activity. Given that the study period covers 2018–2024 with semiannual observations, we assume that an individual institutional investor’s investment style remains relatively stable during this period. Therefore, the average turnover rate of investor k is calculated as the arithmetic mean of turnover rates across all semiannual observations during the sample period as follows:

$$\overline{{\mathrm{C}\mathrm{R}}_{\mathrm{k}}}={\displaystyle \frac{1}{\mathrm{N}}}\sum _{\mathrm{j}=1}^{\mathrm{N}}{\mathrm{C}\mathrm{R}}_{\mathrm{k},\mathrm{t}}$$

(5)

N represents the number of semiannual observations for institutional investor k during the sample period. All qualified institutional investors are then ranked according to their average turnover rate $\overline{{\mathrm{C}\mathrm{R}}_{\mathrm{k}}}$ from low to high and divided into three groups. Institutions in the lowest group are defined as long-term institutional investors (LInst), those in the highest group are defined as short-term institutional investors (SInst), and those in the middle group are classified as medium-term institutional investors (MInst).

3.2.4. Moderating Variable

The explanatory variable ESGdif represents the divergence in ESG ratings across rating agencies. With the growing development of responsible investment, multiple ESG rating systems have emerged both domestically and internationally. Because these institutions differ in evaluation standards, indicator systems, and scoring methodologies, they may provide different assessments of the ESG performance of the same firm. To capture such differences, this study uses ESG ratings from Sino-Securities ESG, WIND, SynTao Green Finance, and MioTech as data sources. The ratings from these agencies are first converted into numerical scores and standardized across different rating scales, after which their standard deviation is calculated to measure firm-level ESG rating divergence.

To ensure consistency and comparability with the rating grades of other agencies, this study utilizes the rating grades rather than the raw scores for Sino-Securities ESG. Specifically, the ESG ratings issued by Sino-Securities ESG, WIND, and MioTech consist of nine grades, ranging from C, CC, CCC, B, BB, BBB, A, AA, to AAA. These grades are converted into numerical scores from 1 to 9 in ascending order. The SynTao Green Finance ESG rating system contains ten grades, namely D, C−, C, C+, B−, B, B+, A−, A, and A+, which are assigned values from 0 to 9.

After completing the above transformations, this study calculates the standard deviation of the ESG rating scores for firm i in year t across the five rating systems, and defines this value as ESG rating divergence (ESGdif). A larger value of ESGdif indicates greater disagreement among rating agencies regarding a firm’s ESG performance.

3.2.5. Control Variables

This study includes nine control variables that capture firm-level and financial characteristics [30,72,73]. The precise definitions and measurements for these variables are reported in

Table 1. Definition of variables.

Type	Variables	Symbol	Definition
Dependent variable	Market Value	MarV	Ln(Market value of the firm + 1)
Independent variable	ESG ratings	ESG_SSI	According to the ESG ratings disclosed by Sino-Securities
Mediating variable	Institutional ownership	Inst	The proportion of shares held by institutional investors relative to the firm’s free-float shares
	Long-term institutional investors	LInst	The proportion of shares held by long-term institutional investors relative to the firm’s free-float shares
	Medium-term institutional investors	MInst	The proportion of shares held by medium-term institutional investors relative to the firm’s free-float shares
	Short-term institutional investors	SInst	The proportion of shares held by short-term institutional investors relative to the firm’s free-float shares
Moderating variable	ESG rating divergence	ESGdif	The standard deviation of ESG scores across different ESG rating agencies
Control variables	Asset–liability ratio	Lev	Total liabilities/total assets
	Total assets turnover	ATO	Operating revenue/Average total assets
	Cash flow return on assets ratio	Cashflow	Net cash flow from operating activities/total assets
	Operating revenue growth rate	Growth	(Operating revenue for the year-Operating revenue in the previous year)/Operating revenue in the previous year
	Board size	Board	Natural logarithm of the total number of board members
	Largest shareholding proportion	Top1	Number of shares of the largest shareholder/numbers of total shares
	Listing period	Age	Natural logarithm of (the year-listing year + 1)
	Employee size	Employee	Total number of employees, measured in ten thousand persons
	Nature of property right	SOE	Dummy variable, If the company is a state-owned enterprise, SOEi,t = 1; otherwise, SOEi,t = 0

.

3.3. Empirical Model

In order to test the hypothesis (H1a) of the relationship between ESG ratings and market value of listed companies, model (6) is set up as follows:

$${\mathrm{M}\mathrm{a}\mathrm{r}\mathrm{V}}_{\mathrm{i},\mathrm{t}}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1{\mathrm{E}\mathrm{S}\mathrm{G}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\alpha }}_{\mathrm{j}}{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\mathrm{s}}_{\mathrm{i},\mathrm{t}}+\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}\mathrm{F}\mathrm{E}+\mathrm{F}\mathrm{i}\mathrm{r}\mathrm{m}\mathrm{F}\mathrm{E}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

(6)

In model (6), i represents the firm and t represents the year. The dependent variable, ${\mathrm{M}\mathrm{a}\mathrm{r}\mathrm{V}}_{\mathrm{i},\mathrm{t}}$ captures the market valuation of firm i in year t. The independent variable ESG_i,t is the ESG rating of firm i in year t. The coefficient ${\mathsf{\alpha }}_1$ quantifies the effect of ESG ratings on firms’ market performance. Hypothesis (H1a) is supported if ${\mathsf{\alpha }}_1$ is significantly positive.

To ensure the robustness of the results, standard errors in all regression models are clustered at the firm level to account for potential serial correlation and heteroskedasticity within firms.

To examine hypothesis (H2), this study follows the stepwise regression approach proposed by Baron and Kenny [74]. The following models (7) and (8) are specified as follows:

$${\mathrm{I}\mathrm{n}\mathrm{s}\mathrm{t}}_{\mathrm{i},\mathrm{t}}={\mathsf{\beta }}_0+{\mathsf{\beta }}_1{\mathrm{E}\mathrm{S}\mathrm{G}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\beta }}_{\mathrm{j}}{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\mathrm{s}}_{\mathrm{i},\mathrm{t}}+\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}\mathrm{F}\mathrm{E}+\mathrm{F}\mathrm{i}\mathrm{r}\mathrm{m}\mathrm{F}\mathrm{E}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

(7)

$${\mathrm{M}\mathrm{a}\mathrm{r}\mathrm{V}}_{\mathrm{i},\mathrm{t}}={\mathsf{\theta }}_0+{\mathsf{\theta }}_1{\mathrm{E}\mathrm{S}\mathrm{G}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\theta }}_2{\mathrm{I}\mathrm{n}\mathrm{s}\mathrm{t}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\theta }}_{\mathrm{j}}{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\mathrm{s}}_{\mathrm{i},\mathrm{t}}+\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}\mathrm{F}\mathrm{E}+\mathrm{F}\mathrm{i}\mathrm{r}\mathrm{m}\mathrm{F}\mathrm{E}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

(8)

${\mathrm{I}\mathrm{n}\mathrm{s}\mathrm{t}}_{i,t}$ denotes the institutional ownership of firm i in year t.

To examine how institutional investor heterogeneity influences their mediating role hypothesis (H3), model (9) is specified as follows:

$${\mathrm{M}\mathrm{a}\mathrm{r}\mathrm{V}}_{\mathrm{i},\mathrm{t}}={\mathsf{\gamma }}_0+{\mathsf{\gamma }}_1{\mathrm{E}\mathrm{S}\mathrm{G}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\gamma }}_2{\mathrm{I}\mathrm{n}\mathrm{v}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{o}\mathrm{r}\mathrm{T}\mathrm{y}\mathrm{p}\mathrm{e}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\gamma }}_{\mathrm{j}}{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\mathrm{s}}_{\mathrm{i},\mathrm{t}}+\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}\mathrm{F}\mathrm{E}+\mathrm{F}\mathrm{i}\mathrm{r}\mathrm{m}\mathrm{F}\mathrm{E}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

(9)

In model (9), $\mathrm{I}\mathrm{n}\mathrm{v}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{o}\mathrm{r}\mathrm{T}\mathrm{y}\mathrm{p}\mathrm{e}$ represents the heterogeneity of institutional investors and includes LInst, MInst, and SInst.

To examine the moderating effect of ESG rating divergence (ESGdif), model (10) is specified as follows:

$${\mathrm{M}\mathrm{a}\mathrm{r}\mathrm{V}}_{\mathrm{i},\mathrm{t}}={\mathsf{\mu }}_0+{\mathsf{\mu }}_1{\mathrm{E}\mathrm{S}\mathrm{G}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\mu }}_2{\mathrm{E}\mathrm{S}\mathrm{G}\mathrm{d}\mathrm{i}\mathrm{f}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\mu }}_{\mathrm{j}}({\mathrm{E}\mathrm{S}\mathrm{G}}_{\mathrm{i},\mathrm{t}}\times {\mathrm{E}\mathrm{S}\mathrm{G}\mathrm{d}\mathrm{i}\mathrm{f}}_{\mathrm{i},\mathrm{t}})+\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}\mathrm{F}\mathrm{E}+\mathrm{F}\mathrm{i}\mathrm{r}\mathrm{m}\mathrm{F}\mathrm{E}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

(10)

4. Empirical Results and Discussions

4.1. Descriptive Statistics

Table 2. Overview of variables.

VarName	Obs	Mean	SD	Min	Median	Max
MarV	19,119	4.4888	1.1695	1.9427	4.3056	8.2898
ESG_SSI	19,119	73.9641	5.0506	41.1900	73.9400	100.0000
Inst	19,119	0.4409	0.2544	0.0010	0.4602	0.9230
LInst	19,119	0.0157	0.0301	0.0000	0.0034	0.4263
MInst	19,119	0.4024	0.2529	−0.3903	0.4273	0.9211
SInst	19,119	0.0227	0.0417	0.0000	0.0050	0.5185
ESGdif	15,970	0.7977	0.6620	0.0000	0.7582	4.0965
Lev	19,119	0.3995	0.1907	0.0278	0.3927	0.9104
ATO	19,119	0.6281	0.4029	0.0467	0.5397	2.6445
CashFlow	19,119	0.0579	0.0648	−0.2218	0.0557	0.2822
Growth	19,119	0.1690	0.3519	−0.6735	0.1124	5.0755
Board	19,119	2.1213	0.1968	1.6094	2.1972	2.7081
Top1	19,119	0.3468	0.1489	0.0749	0.3272	0.7579
ListAge	19,119	2.0750	0.7830	0.6931	2.0794	3.5553
SOE	19,119	0.3382	0.4731	0.0000	0.0000	1.0000

presents descriptive statistics for the main variables. Market value (MarV) has a mean of 4.4888 and a standard deviation of 1.1695, indicating substantial variation in firm market valuation among A-share listed companies. The ESG rating score (ESG_SSI) ranges from 41.19 to 100, with an average of 73.9641 and a standard deviation of 5.0506, suggesting generally high ESG performance with notable cross-sectional differences.

Regarding institutional ownership, the aggregate institutional ownership (Inst) has a mean of 0.4409 and a standard deviation of 0.2544, reflecting moderate variation across firms. The ESG rating divergence (ESGdif) averages 0.7977 with a standard deviation of 0.6620, indicating heterogeneity in ESG assessments across rating agencies.

4.2. Baseline Regression Analysis

We compare fixed effects and random effects using the Hausman test and find that the fixed effects model is more appropriate.

Table 3. Hausman test.

Dependent Variable	Test Result	Chi-Square Statistic	p-Value
MarV	Fixed effects	1770.14	0.0000

reports the results of the Hausman test.

To assess potential multicollinearity among explanatory variables, we compute the Variance Inflation Factors (VIFs). As shown in

Table 4. VIF test.

Coefficients
Model	Collinearity Statistics
	Tolerance	VIF
Variable	VIF	1/VIF
SOE	1.48	0.675226
ListAge	1.42	0.705445
Lev	1.24	0.803689
Top1	1.12	0.891529
Board	1.11	0.901617
CashFlow	1.08	0.923200
ATO	1.08	0.923819
Growth	1.05	0.955329
ESG_SSI	1.02	0.977748
Mean VIF	1.18

, all VIF values are well below the conventional threshold of 10, with the maximum being 1.48 (for SOE) and the mean VIF equal to 1.18. This indicates that multicollinearity is not a concern in our model specification.

We further conduct the Wooldridge test for serial correlation in panel data. The results in

Table 5. Wooldridge test.

Wooldridge Test	Coef.
F(1,3124)	1993.618
Prob > F	0.0000

indicate the presence of first-order autocorrelation (p < 0.05). To address this issue, all regressions are estimated with cluster-robust standard errors at the firm level, ensuring the robustness of statistical inference.

Table 6. The effect of ESG ratings on market value.

Variable	(1)	(2)	(3)	(4)	(5)	(6)
	MarV	MarV	MarV	MarV	MarV	MarV
ESG_SSI	0.0570 ***	0.0573 ***	0.0032 ***
	(25.9590)	(26.1292)	(4.8308)
Escore				0.0016 ***
				(3.0569)
Sscore					0.0012 **
					(2.5113)
Gscore						0.0034 ***
						(5.3123)
Lev	2.3052 ***	2.3200 ***	0.8736 ***	0.8658 ***	0.8602 ***	0.8821 ***
	(26.3966)	(26.5773)	(13.7721)	(13.6472)	(13.5278)	(13.8773)
ATO	−0.1752 ***	−0.1708 ***	0.0424	0.0443	0.0408	0.0426
	(−4.5879)	(−4.4968)	(1.3963)	(1.4620)	(1.3398)	(1.3974)
CashFlow	3.1252 ***	3.1224 ***	0.3318 ***	0.3269 ***	0.3337 ***	0.3335 ***
	(17.0741)	(17.0225)	(6.6993)	(6.6165)	(6.7355)	(6.7387)
Growth	0.3935 ***	0.3781 ***	0.1567 ***	0.1556 ***	0.1549 ***	0.1562 ***
	(14.5460)	(13.6959)	(14.4226)	(14.3317)	(14.2960)	(14.4383)
Board	0.6493 ***	0.6772 ***	0.1316 ***	0.1317 ***	0.1304 ***	0.1343 ***
	(8.7812)	(9.1477)	(3.5053)	(3.4978)	(3.4643)	(3.5671)
Top1	0.8336 ***	0.8476 ***	−0.3531 **	−0.3520 **	−0.3529 **	−0.3458 **
	(8.1493)	(8.3086)	(−2.3593)	(−2.3549)	(−2.3567)	(−2.3100)
ListAge	0.5234 ***	0.5133 ***	0.4325 ***	0.4297 ***	0.4311 ***	0.4344 ***
	(27.6197)	(26.8832)	(21.4783)	(21.3437)	(21.3483)	(21.5264)
SOE	0.0991 **	0.1086 ***	−0.0555 *	−0.0542 *	−0.0564 *	−0.0565 *
	(2.5391)	(2.7782)	(−1.7645)	(−1.7144)	(−1.7852)	(−1.7898)
_cons	−3.4971 ***	−3.5689 ***	2.8490 ***	2.9925 ***	3.0035 ***	2.7978 ***
	(−15.3594)	(−15.7266)	(23.3629)	(25.8972)	(26.0910)	(22.4876)
YearFE	No	Yes	Yes	Yes	Yes	Yes
FirmFE	No	No	Yes	Yes	Yes	Yes
F	402.2770	404.0622	159.7270	158.6091	158.2938	161.7470
R2	0.4655	0.4763	0.9635	0.9634	0.9634	0.9635
N	19,119	19,119	19,119	19,119	19,119	19,119

Note: *, **, and *** indicate significance at the level of 10%, 5%, and 1%, respectively; T-statistics for the regression coefficients are in parentheses.

reports the baseline regression results on the relationship between ESG ratings and firm market value (MarV). Columns (1)–(3) present the regression results using ESG_SSI as the key explanatory variable. Without controlling for fixed effects (column (1)), ESG_SSI is positively and significantly associated with firm market value (coefficient = 0.0570, t = 25.9590). After including year fixed effects in column (2), the coefficient remains significant (coefficient = 0.0573). When both firm and year fixed effects are further controlled for in column (3), the coefficient remains positive and statistically significant at the 1% level (coefficient = 0.0032, t = 4.8308).

Columns (4)–(6) further examine the effects of the three ESG sub-dimensions—environmental (Escore), social (Sscore), and governance (Gscore)—on firm market value. After controlling for both firm and year fixed effects, the scores of all three dimensions are positively and significantly related to firm market value. Specifically, the coefficient of Escore is 0.0016 (t = 3.0569), indicating that improvements in environmental performance contribute to higher market valuation. The coefficient of Sscore is 0.0012 (t = 2.5113), suggesting that firms with stronger social responsibility performance tend to receive more favorable valuation from investors. In contrast, the coefficient of Gscore is 0.0034 (t = 5.3123), which is not only more statistically significant but also larger in magnitude, implying that the governance dimension plays a particularly important role in the market valuation process.

The positive association between ESG_SSI and market value implies that investors are able to recognize and price ESG-related information beyond traditional financial fundamentals [75]. In particular, ESG rating appears to be valued not merely as a symbolic or compliance-oriented attribute, but as an indicator associated with firms’ long-term sustainability, risk management capability, and value creation potential. Overall, the baseline regression results consistently support hypothesis (H1a).

4.3. Mediation Effect Analysis

Table 7. The mediating role of institutional investor shareholding.

Variable	(1)	(2)	(3)
	MarV	Inst	MarV
ESG_SSI	0.0032 ***	0.0004 **	0.0032 ***
	(4.8308)	(2.0715)	(5.0895)
Inst			1.4878 ***
			(23.9521)
Lev	0.8736 ***	−0.0072	0.8834 ***
	(13.7721)	(−0.5515)	(15.3267)
ATO	0.0424	0.0060	0.0299
	(1.3963)	(0.7372)	(1.1983)
CashFlow	0.3318 ***	−0.0017	0.3287 ***
	(6.6993)	(−0.1440)	(6.9927)
Growth	0.1567 ***	0.0175 ***	0.1311 ***
	(14.4226)	(6.4931)	(13.0470)
Board	0.1316 ***	0.0405 ***	0.0699 **
	(3.5053)	(4.1355)	(2.0681)
Top1	−0.3531 **	0.3122 ***	−0.8342 ***
	(−2.3593)	(7.4253)	(−6.5631)
ListAge	0.4325 ***	−0.0290 ***	0.4778 ***
	(21.4783)	(−6.5600)	(25.0250)
SOE	−0.0555 *	0.0725 ***	−0.1581 ***
	(−1.7645)	(6.7437)	(−5.0527)
_cons	2.8490 ***	0.2408 ***	2.4603 ***
	(23.3629)	(7.4768)	(21.3930)
YearFE	Yes	Yes	Yes
FirmFE	Yes	Yes	Yes
F	159.7270	27.9494	207.9376
R2	0.9635	0.9591	0.9676
N	19,119	19,119	19,119

Note: *, **, and *** indicate significance at the level of 10%, 5%, and 1%, respectively; T-statistics for the regression coefficients are in parentheses.

reports the baseline mediation analysis. Column (2) shows that ESG_SSI has a positive and statistically significant effect on institutional ownership (Inst), with a coefficient of 0.0004 (t = 2.0715), indicating that firms with higher ESG ratings tend to attract a larger proportion of institutional investors. Column (3) further introduces institutional ownership into the market value regression. The coefficient of Inst is positive and highly significant (coefficient = 1.4878, t = 23.9521). Taken together, these results indicate a partial mediation effect: higher ESG ratings not only directly enhance firm market value but also indirectly increase firm value by attracting greater institutional investor participation. Overall, the mediation analysis lends strong support to hypothesis (H2).

To further explore heterogeneity in the mediation mechanism,

Table 8. Heterogeneous mediation effects by institutional investment horizon.

Variable	(1)	(2)	(3)	(4)	(5)	(6)
	LInst	MarV	MInst	MarV	SInst	MarV
ESG_SSI	0.0005 ***	0.0026 ***	−0.0003 *	0.0034 ***	−0.0002 ***	0.0035 ***
	(7.7240)	(4.0059)	(−1.8891)	(5.2377)	(−3.7950)	(5.3263)
LInst		1.3103 ***
		(8.0560)
MInst				1.2414 ***
				(16.8348)
SInst						1.4080 ***
						(11.5716)
Lev	−0.0004	0.8741 ***	−0.0123	0.8943 ***	0.0089 **	0.8611 ***
	(−0.1038)	(13.9263)	(−1.0245)	(14.9110)	(2.1339)	(13.6457)
ATO	0.0035 *	0.0379	0.0023	0.0405	0.0044 **	0.0363
	(1.8481)	(1.2594)	(0.3137)	(1.5333)	(2.3341)	(1.2024)
CashFlow	−0.0020	0.3344 ***	−0.0060	0.3439 ***	0.0113 **	0.3159 ***
	(−0.5304)	(6.7758)	(−0.5592)	(7.0927)	(2.3526)	(6.4184)
Growth	−0.0009	0.1578 ***	0.0077 ***	0.1496 ***	0.0107 ***	0.1416 ***
	(−1.1623)	(14.5242)	(3.1107)	(14.2922)	(10.6586)	(13.1474)
Board	0.0006	0.1309 ***	0.0425 ***	0.0817 **	−0.0009	0.1329 ***
	(0.2035)	(3.5350)	(4.8218)	(2.2963)	(−0.2799)	(3.5934)
Top1	−0.0142 **	−0.3345 **	0.3575 ***	−0.7866 ***	−0.0135 **	−0.3341 **
	(−2.1637)	(−2.2555)	(8.8009)	(−5.9332)	(−2.1517)	(−2.2409)
ListAge	0.0005	0.4319 ***	−0.0234 ***	0.4554 ***	−0.0150 ***	0.4536 ***
	(0.3223)	(21.7285)	(−5.6777)	(23.0257)	(−7.8839)	(22.7364)
SOE	−0.0099 ***	−0.0425	0.0743 ***	−0.1476 ***	0.0024	−0.0589 *
	(−3.3489)	(−1.3725)	(7.1284)	(−4.4844)	(1.2709)	(−1.9283)
_cons	−0.0133	2.8664 ***	0.2312 ***	2.5806 ***	0.0629 ***	2.7604 ***
	(−1.5006)	(23.8638)	(7.8994)	(21.5093)	(7.0234)	(22.8771)
YearFE	Yes	Yes	Yes	Yes	Yes	Yes
FirmFE	Yes	Yes	Yes	Yes	Yes	Yes
F	8.8188	148.9450	24.2914	171.1810	21.7969	168.3098
R2	0.6525	0.9639	0.9645	0.9660	0.5286	0.9642
N	19,119	19,119	19,119	19,119	19,119	19,119

Note: *, **, and *** indicate significance at the level of 10%, 5%, and 1%, respectively; T-statistics for the regression coefficients are in parentheses.

distinguishes among long-term, medium-term, and short-term institutional investors. Column (1) shows that ESG_SSI has a significantly positive impact on long-term institutional ownership (LInst) (coefficient = 0.0005, t = 7.7240), suggesting that firms with higher ESG ratings are more likely to attract long-term institutional capital. When LInst is incorporated into the market value regression in column (2), the coefficient of LInst is positive and statistically significant (coefficient = 1.3103, t = 8.0560), indicating that long-term institutional investors play an important role in transmitting ESG information into firm valuation [76].

In contrast, column (3) indicates that ESG_SSI has a weakly negative effect on medium-term institutional ownership (MInst) (coefficient = −0.0003, t = −1.8891). Similarly, column (5) shows that ESG_SSI is negatively related to short-term institutional ownership (SInst) (coefficient = −0.0002, t = −3.7950).

Taken together, the results reveal a clear asymmetry in the mediating role of institutional investors. ESG ratings primarily attract long-term institutional investors, who in turn contribute to the transmission of ESG information into firm valuation. By contrast, medium-term and short-term institutional investors do not increase their holdings in response to higher ESG ratings, even though their holdings remain positively associated with firm value. These findings suggest that the valuation effects of ESG ratings are mainly transmitted through investors with longer investment horizons [77], highlighting the importance of investor horizon in ESG pricing and providing empirical support for hypothesis (H3).

5. Robustness and Endogeneity Tests

5.1. Robustness Tests

To further assess the robustness of the main findings,

Table 9. Robustness and endogeneity tests results.

Variable	(1)	(2)	(3)	(4)	(5)	(6)
	BM	MarV	MarV	MarV	ESG_SSI	MarV
L.ESG_SSI				0.0017 *
				(1.9144)
LInst
ESG_SSI	0.0010 ***	0.0038 ***	0.0031 ***			0.2792 ***
	(3.6066)	(4.6024)	(4.7484)			(8.4516)
IV					0.0012 ***
					(2.6833)
Lev	0.0957 ***	0.7541 ***	0.8679 ***	1.0032 ***	−3.5887 ***	2.5045 ***
	(4.5670)	(10.8555)	(13.9909)	(13.0809)	(−4.9424)	(35.3939)
ATO	−0.1113 ***	−0.0200	0.0443	0.0086	0.1707	−0.1506 ***
	(−9.3937)	(−0.6088)	(1.4543)	(0.2509)	(0.5071)	(−5.1230)
CashFlow	−0.1927 ***	0.3939 ***	0.3270 ***	0.4057 ***	−0.1294	1.4869 ***
	(−9.2710)	(6.5283)	(6.6231)	(7.2616)	(−0.1826)	(4.9278)
Growth	−0.0464 ***	0.2159 ***	0.1582 ***	0.1618 ***	−0.5771 ***	0.4492 ***
	(−10.7735)	(15.4975)	(14.4488)	(13.6668)	(−4.1637)	(11.8613)
Board	0.0191	0.1243 ***	0.1221 ***	0.1173 ***	−0.1254	0.5411 ***
	(1.3540)	(2.6783)	(3.2815)	(2.7622)	(−0.2370)	(9.0640)
Top1	0.0339	−0.4354 ***	−0.3562 **	−0.0914	−0.1387	0.5915 ***
	(0.9166)	(−2.5923)	(−2.3836)	(−0.5925)	(−0.1027)	(7.0052)
ListAge	−0.1045 ***	0.4711 ***	0.4317 ***	0.5358 ***	−0.8693 ***	0.5847 ***
	(−14.0185)	(19.8121)	(21.4108)	(18.8430)	(−3.0309)	(32.2767)
SOE	−0.0122	−0.0447	−0.0567 *	−0.0993 **	0.1359	−0.1975 ***
	(−1.1122)	(−1.2309)	(−1.8495)	(−2.4885)	(0.3038)	(−3.7571)
_cons	0.8036 ***	2.7485 ***	2.8777 ***	2.7247 ***	77.4547 ***	−19.7271 ***
	(18.5111)	(18.8716)	(24.1823)	(17.9278)	(55.0007)	(−8.1643)
YearFE	Yes	Yes	Yes	Yes	Yes	Yes
FirmFE	Yes	Yes	Yes	Yes	Yes	Yes
F	75.4368	126.3130	158.3956	111.4321	8.4301	590.9683
R2	0.8563	0.9578	0.9636	0.9600	0.5349	−0.4747
N	19,119	13,395	19,119	15,584	19,119	19,182
Kleibergen–Paap rk Wald F						38.45
Stock–Yogo Critical Values (10% maximal IV size)						16.38
Durbin–Wu–Hausman Statistics						78.58 p = 0.0000

Note: *, **, and *** indicate significance at the level of 10%, 5%, and 1%, respectively; T-statistics for the regression coefficients are in parentheses.

reports regression results based on alternative model specifications. In column (1), the dependent variable is replaced with the book-to-market ratio (BM), which captures the valuation structure of firms from the perspective of the capital market. A higher BM generally indicates lower market valuation relative to book value, reflecting investors’ expectations regarding growth prospects, risk, or intangible assets [78]. The coefficient of ESG_SSI is positive and statistically significant (coefficient = 0.0010, t = 3.6066), indicating that ESG_SSI is systematically associated with firms’ valuation characteristics as measured by BM. This evidence supports the stability of the baseline relationship under an alternative valuation metric.

Second, when the sample is restricted to manufacturing firms in column (2), the coefficient of ESG_SSI remains positive and statistically significant (coefficient = 0.0038, p < 0.01). Manufacturing firms are characterized by high resource dependence, heavy energy consumption, and labor-intensive operations, making them more prone to negative incidents such as environmental pollution, lapses in worker protection, and corporate governance failures. As a result, the market and institutional investors pay closer attention to their ESG ratings. The findings indicate that the positive association between ESG ratings and market performance persists even within a relatively homogeneous industry setting.

Third, column (3) further strengthens the baseline specification by incorporating province fixed effects, thereby controlling for time-invariant regional heterogeneity such as differences in local regulatory intensity, institutional environments, and regional development levels. Under this more stringent specification, the positive effect of ESG_SSI on firm market value remains highly significant and economically meaningful, with an estimated coefficient of 0.0031 (p < 0.01). The magnitude of the coefficient is very close to that obtained in the baseline model, indicating that the valuation premium associated with higher ESG ratings are not driven by unobserved provincial characteristics or region-specific institutional factors.

Finally, in column (4), we replace the core explanatory variable ESG_SSI with its one-period lag (L.ESG_SSI) to conduct a robustness check of the main results. The use of lagged ESG is justified for several reasons. First, investors require time to observe and interpret the actual implications of ESG ratings for firm prospects. Even when firms release high-quality ESG reports, investors typically do not immediately incorporate the information into their investment decisions. Instead, they integrate ESG ratings with multiple dimensions of information, including financial performance, strategic plans, market position, and industry context, to assess the actual impact of ESG on long-term firm value. Second, for large or long-term holdings, adjustments are usually executed in stages to mitigate trading costs and avoid substantial market impact. In particular, unwinding sizable long-term positions gradually is necessary to prevent significant price declines, which extends the time it takes for ESG information to be transmitted into market valuation. Based on this rationale, we re-estimate the market value regression using L.ESG_SSI as the explanatory variable. The results show that the coefficient of L.ESG_SSI is 0.0017 (t = 1.9144), which is marginally significant at the 10% level. The positive relationship indicates that an increase in ESG ratings in the previous period is associated with higher firm market value in the subsequent period, supporting the notion that ESG information requires time to be digested and gradually capitalized by the market. Overall, this robustness check suggests that the main conclusion remains valid even when only the lagged ESG rating is considered.

5.2. Endogeneity Tests

We examined the impact of ESG ratings on corporate market performance. First, reverse causality is a potential concern: firms with superior market performance may possess greater financial and organizational resources to invest in ESG activities, thereby obtaining higher ESG ratings. At the same time, firms with higher market valuations are more likely to attract institutional investors, which in turn may shape the observed ownership structure. Second, omitted variable bias may occur if unobservable firm characteristics—such as managerial long-term orientation, corporate culture, or governance philosophy—simultaneously affect ESG engagement, institutional investors’ investment, and market performance. Third, ESG ratings may be subject to measurement error, which could further bias the estimated coefficients.

To further address potential endogeneity arising from reverse causality, omitted variables, and measurement error, this study employs an instrumental variable approach for ESG ratings. The instrument is regional ESG awareness, proxied by the ESG-related Baidu Search Index for the region in which a firm is headquartered. This measure captures the intensity of public attention and social concern regarding ESG issues at the local level. Regional ESG awareness is unlikely to affect the market value of an individual firm through direct firm-specific economic channels other than ESG-related behavior. From the relevance perspective, regional search intensity primarily reflects the degree of public, media, and regulatory attention to ESG-related issues. Such attention is more likely to operate through its influence on firms’ information disclosure and governance practices [79]. For example, in regions with higher public attention to ESG issues, media coverage and social monitoring tend to be more frequent, and regulatory authorities may place greater emphasis on firms’ sustainable development practices. These pressures may encourage firms to strengthen ESG disclosure and improve their environmental, social, and governance practices [80]. In this process, rating agencies are able to obtain more comprehensive and transparent information and may adjust their ESG evaluations accordingly [81]. At the same time, long-term institutional investors are more likely to identify firms with sustained improvements in ESG performance and may increase their shareholdings or engage in active governance. From the exclusion restriction perspective, the instrumental variable mainly captures the allocation of public attention across different regions and does not directly affect firm market value. Therefore, in theory, the selected instrumental variable satisfies the exclusion restriction and can be considered a valid instrument.

The results of Durbin–Wu–Hausman test in Table 9 strongly reject the null hypothesis of exogeneity, indicating that ESG_SSI is endogenous. This finding justifies the use of the instrumental variable approach. The first-stage regression results, reported in Column (5) of Table 9, indicate that regional ESG awareness is strongly and positively associated with ESG_SSI, with a coefficient of 0.0012 (p < 0.05). This confirms the relevance condition, suggesting that firms located in regions with higher ESG-related search intensity tend to exhibit significantly higher ESG ratings.

The second-stage results, shown in Column (6), reveal that the instrumented ESG_SSI continues to exert a positive and statistically significant impact on corporate market performance, with a coefficient of 0.2792 (p < 0.01). The magnitude and significance of this effect indicate that correcting for endogeneity strengthens, rather than weakens, the estimated impact of ESG ratings on firm value.

The Kleibergen–Paap rk Wald F statistic exceeds the Stock–Yogo critical value at the 10 percent maximal size, indicating that weak instrument concerns are unlikely.

6. Further Analysis: ESG Rating Divergence

Given the widespread differences across ESG rating agencies in China’s capital market, this section further examines whether ESG rating divergence (ESGdif) moderates the effect of ESG ratings on firms’ market value [24,82].

Table 10. The Moderating Effect of ESG Rating Divergence.

Variable	(1)	(2)
	MarV	MarV
ESG_SSI	0.0032 ***	0.0069 ***
	(4.8308)	(5.6367)
ESGdif		−0.0277 ***
		(−3.2750)
ESG_SSI × ESGdif		−0.0020 *
		(−1.8273)
Lev	0.8736 ***	0.8741 ***
	(13.7721)	(10.9703)
ATO	0.0424	−0.0679 *
	(1.3963)	(−1.8804)
CashFlow	0.3318 ***	0.4365 ***
	(6.6993)	(5.9716)
Growth	0.1567 ***	0.0934 ***
	(14.4226)	(7.0210)
Board	0.1316 ***	0.2628 ***
	(3.5053)	(5.5061)
Top1	−0.3531 **	0.0182
	(−2.3593)	(0.1037)
ListAge	0.4325 ***	0.4691 ***
	(21.4783)	(15.5152)
SOE	−0.0555 *	−0.1585 ***
	(−1.7645)	(−3.2220)
_cons	2.8490 ***	2.5321 ***
	(23.3629)	(13.7801)
YearFE	Yes	Yes
FirmFE	Yes	Yes
F	159.7270	55.8811
R2	0.9635	0.9316
N	19,119	15,553

Note: *, **, and *** indicate significance at the level of 10%, 5%, and 1%, respectively; T-statistics for the regression coefficients are in parentheses.

reports the estimation results.

Column (2) introduces ESG rating divergence and its interaction with ESG_SSI. The interaction term ESG_SSI × ESGdif is negative and statistically significant (−0.0020, p < 0.1). This result indicates that ESG rating divergence attenuates the positive effect of ESG ratings on market value. In other words, while firms with higher ESG scores tend to enjoy higher market valuations, this valuation premium becomes smaller when ESG ratings are less consistent across agencies. When rating divergence is high, investors are likely to discount ESG signals due to ambiguity regarding their reliability and comparability.

From a theoretical perspective, these results are consistent with information asymmetry and signal credibility arguments. ESG ratings serve as external signals of firms’ sustainability performance; however, substantial divergence across rating providers dilutes the clarity of this signal [83]. As a result, institutional investors may respond more cautiously, reducing their willingness to price ESG rating aggressively into firm value.

7. Conclusions and Implications

7.1. Research Conclusions

Based on a panel of Chinese A-share listed firms from 2018 to 2024, this study systematically examines the pricing mechanism of corporate ESG ratings in the capital market, with particular attention to the mediating role of institutional investors and the moderating effect of ESG rating divergence. The baseline regression results indicate that ESG_SSI is positively and significantly associated with firm market value, suggesting that the market recognizes and partially capitalizes the value premium of ESG information. From a valuation theory perspective, enhanced ESG performance can influence firm value by improving the sustainability of future cash flows, reducing both systematic and firm-specific risks, and lowering capital costs. Furthermore, the mediation analysis demonstrates that institutional investors serve as an important channel through which ESG information is transmitted to firm market value, with heterogeneity across investment horizons: holdings by long-term institutional investors significantly amplify the positive effect of ESG on firm value, while medium- and short-term investors exhibit weaker responses. In addition, ESG rating divergence significantly dampens the valuation effect of ESG ratings, indicating that inconsistency among rating agencies reduces signal credibility and increases informational uncertainty. These findings theoretically validate the transmission chain of “signal consistency–signal credibility–market response” and underscore that the effectiveness of ESG ratings as non-financial corporate information signals depends on their credibility and consistency. Robustness checks and endogeneity analyses further support the main findings, enhancing the reliability and interpretability of the conclusions.

7.2. Practical Implications

For corporate managers: To enhance market recognition of ESG information, firms should establish systematic and verifiable ESG and disclosure mechanisms, including clearly defined key performance indicators, optimized internal data governance processes, and appropriate third-party verification. Compared with pursuing high scores from individual rating agencies, consistent and verifiable practices are more likely to foster long-term market trust. Firms should incorporate ESG into long-term strategic considerations, covering capital allocation, supply chain management, and major investment decisions, to effectively improve future cash flows and mitigate corporate risk. Additionally, firms should maintain communication with major rating agencies, clarify disclosure standards and data sources, and provide supplementary explanations or independent audit reports when necessary to enhance rating consistency.

From the perspective of institutional investors: Investors should avoid relying solely on a single ESG score, especially when rating divergence is substantial or information transparency is limited, and should instead conduct comprehensive assessments based on firm fundamentals, governance structure, industry context, and liquidity. The study shows that long-term institutional investors are more likely to capitalize ESG information into valuation premiums; thus, pension funds, insurance companies, and other long-horizon investors should incorporate high-quality and consistent ESG indicators into long-term portfolio allocation and active engagement strategies. Short- and medium-term traders should recognize that ESG information may have limited impact on firm value in the short term. In portfolio management, fund managers should treat rating divergence as a risk factor, employ multi-source consensus or hedging strategies to mitigate potential valuation errors, and pay attention to shareholding size and tradability, as these factors affect transaction costs and the feasibility of divestment.

7.3. Limitations and Future Research Directions

Despite our contributions, this study has several limitations. First, the Sino-Securities (SSI) ESG rating incorporates a number of indicators that are specific to the Chinese institutional and regulatory context. While this localization enhances the relevance of the ESG measure for China’s capital market, it may limit the direct comparability of the findings with studies based on international ESG rating systems.

In the instrumental variable robustness analysis, although this study employs regional ESG awareness as an instrumental variable to alleviate potential endogeneity concerns, the identification strategy relies on the exclusion restriction assumption—that is, the instrument affects firm market value only through its impact on firms’ ESG behavior. However, in practice, regional ESG awareness may still indirectly influence firm value through other regional institutional or informational channels. Therefore, the exclusion restriction condition may be subject to a certain degree of potential bias.

In addition, this study has limitations regarding the identification method for institutional investors’ investment styles. Specifically, we classify institutional investors into long-term, medium-term, and short-term categories based on their average turnover ratio over the sample period. This classification approach primarily relies on the overall investment behavior characteristics of investors, rather than dynamically categorizing them based on their specific holding behaviors in individual firms. Consequently, while an institutional investor may exhibit different investment horizon structures across different companies, this firm-level heterogeneity is not further distinguished in our analysis. Furthermore, since institutional ownership stakes and trading behaviors may evolve over time, the classification method based on the average turnover ratio over the sample period may not fully capture such dynamic adjustment processes.