Introduction

Prior research indicates that shareholders react negatively in capital markets when firms face earnings pressure (Matsumoto 2002; Pevzner et al. 2015; Gonzalez and Li 2024). The literature typically assesses earnings pressure based on whether a firm meets analysts’ earnings forecasts, as these forecasts serve as key external performance benchmarks for publicly traded firms to attain or exceed (Bartov et al. 2002; Currim et al. 2018; Liu et al. 2021). Meeting these expectations satisfies shareholders, enhances firm value, and improves market performance (Kasznik and McNichols 2002; Brauer and Wiersema 2018; Hirshleifer et al. 2025). Conversely, failing to meet these expectations imposes earnings pressure on firms, leading to declines in firm value (Bhojraj et al. 2009; Kaplan and Zamora 2018; Heater et al. 2025).Footnote 1

Recent debates examine whether CSR moderates the impact of negative events on firm value by mitigating or amplifying their effects. On one hand, some studies suggest that strong CSR profiles buffer firms against crises, legal issues, or product recalls by providing reputational capital that shields firm value (Godfrey 2005; Godfrey et al. 2009; Minor and Morgan 2011; Christensen 2016; Luo et al. 2018; Bae et al. 2020; Kim et al. 2024). This insurance-like effect suggests that CSR can serve as an insurance against the detrimental impacts of such events on firm value. Conversely, others argue that CSR exacerbates negative events, as firms with strong CSR reputations face higher expectations and harsher scrutiny when they fall short, leading to greater market penalties (Janney and Gove 2011; Kim and Lee 2015; Zhang et al. 2022). According to this view, firms with high CSR reputations face heightened scrutiny and higher expectations from stakeholders. When such firms encounter negative events, they may receive harsher criticism compared to firms with lower CSR reputations (Kim and Lee 2015; Liu et al. 2020; Zhang et al. 2025). These contrasting perspectives underscore the need to clarify when CSR mitigates or amplifies the consequences of negative events on firm value.

Grounded in stakeholder theory (Freeman 1984), we argue that the institutional transition from shareholder primacy to stakeholder orientation shapes how stakeholders, including investors, perceive CSR. Consequently, CSR’s moderating role in the relationship between earnings pressure and firm value is contingent on the prevailing institutional logic.Footnote 2 This reasoning aligns with the institutional logics perspective (Lok 2010), which emphasizes that corporate strategies gain legitimacy when aligned with societal norms (Thornton and Ocasio 1999).Footnote 3 Historically, shareholder primacy dominated corporate governance, prioritizing short-term financial outcomes and framing CSR as a value-destroying activity (Friedman 1970; Jensen and Meckling 1976; Smith and Rönnegard 2016). The growing prominence of stakeholder orientation, however, has expanded corporate responsibility to broader societal concerns (Freeman 1984; Stoelhorst and Vishwanathan 2024). Under shareholder primacy, CSR was often regarded as an agency cost, leading investors to penalize firms that underperformed financially while maintaining strong CSR commitments. In contrast, as stakeholder orientation becomes institutionalized, CSR is increasingly perceived as a value-enhancing investment, fostering more favorable investor responses even when firms underperform financially.

Most existing studies on CSR are based on developed markets, where CSR is typically conceptualized as a market-driven, voluntary, and ethically motivated activity. In contrast, CSR practices in China are embedded in a hybrid institutional framework shaped by state intervention, political mandates, and market forces. In this context, CSR is not merely a discretionary initiative rooted in market or moral considerations, but also a strategic response to government policies and institutional pressures. Understanding this hybrid institutional setting is crucial for analyzing how CSR affects firm value in emerging markets.

For example, in China, government regulation operates not only through formal laws but also through informal guidance and administrative incentives, which prompt firms to adopt CSR as a strategy for gaining legitimacy (Marquis and Qian 2014; Zhang et al. 2020). Ownership structure further significantly shapes how CSR is interpreted and valued. State-owned enterprises (SOEs) often treat CSR primarily as a political obligation, which diminishes its signaling value to investors. In contrast, non-SOEs have greater strategic discretion, making their CSR activities more informative about firm value and investor expectations (Jia et al. 2019; Chen et al. 2023). In addition, investors in China—particularly institutional investors—increasingly view CSR not solely through financial or ethical lenses, but also as a proxy for regulatory compliance, political alignment, and consistency with long-term national policy goals (Lu et al. 2023). These distinctive institutional characteristics fundamentally reshape the mechanisms through which CSR operates in China and highlight the importance of situating CSR research within the institutional contexts of emerging markets.

Using a large sample of Chinese listed firms over the past decade, we find that CSR initially amplify the negative impact of earnings pressure on firm value in the early 2010s, but shifted to a mitigating effect by the late 2010s. This transition suggests a decline in shareholder primacy and the growing influence of stakeholder orientation in shaping investor responses to CSR when firms face earnings pressure. This trend aligns with the evolving expectations of Chinese investors, who increasingly reward firms that demonstrate alignment with national priorities and long-term sustainable development. Additionally, the positive amplifying effect of CSR is stronger for non-polluting firms, non-SOEs, and firms located in regions with low environmental regulations, compared to polluting firms, SOEs, and those in highly regulated areas.

This study contributes to the literature in several ways. First, grounded in stakeholder theory (Freeman 1984), we demonstrate that the prevailing institutional logic, whether shareholder primacy or stakeholder orientation, shapes the market’s response to CSR. When shareholder primacy dominates, firms with strong CSR but weak financial performance may be penalized, whereas under stakeholder orientation, they may be rewarded. Our findings extend stakeholder theory by showing that stakeholder responses to CSR are contingent on institutional norms, highlighting the dynamic boundary conditions under which CSR is evaluated by stakeholders. Second, it explores the moderating role of CSR in the context of earnings pressure, an underexplored but important setting in CSR research. Unlike external crises or unethical conduct, which are episodic and irregular, earnings pressure is a common challenge that most firms routinely face. Examining CSR in this context sheds light on its role in shaping how financial stress affects firm value. Third, our study contributes to understanding shifts in institutional logic by examining the transition from shareholder primacy to stakeholder orientation in emerging markets. While most existing research focuses on developed countries, our findings shed light on how institutional contexts shape investor perceptions of CSR in developing countries. We show how China’s unique institutional context, including regulatory pressures, policy constraints, and governance expectations, reshapes CSR’s perceived value. This case illustrates broader ongoing institutional transformations in emerging markets and reveals how evolving stakeholder expectations influence CSR outcomes within these contexts.

The paper is structured as follows: Section II reviews the literature. Section III develops the conceptual framework and formulates the hypotheses. Section IV outlines the methodology, including the sample, data, and model specifications. Section V presents the results, followed by additional analyses in Section VI. Finally, Section VII discusses the findings and their implications.

Literature review

Shifting institutional logics: from shareholder primacy to stakeholder orientation

Institutional logics are defined as socially constructed, historically embedded patterns of material practices, assumptions, values, beliefs, and rules that shape how individuals sustain themselves, structure time and space, and interpret their social reality (Thornton and Ocasio 1999). While multiple institutional logics can coexist, transitions occur as one dominant logic weakens and another gains prominence (Thornton and Ocasio 2008). Shareholder primacy has long been the prevailing logic in corporate governance (Stoelhorst and Vishwanathan 2024). Rooted in agency theory, it posits that managers act as agents for shareholders and should prioritize their interests over other stakeholders (Jensen and Meckling 1976). Under this framework, maximizing shareholder value is the primary corporate objective (Friedman 1970). Legal structures in many jurisdictions reinforce this principle, with fiduciary duties primarily directed toward shareholders (Eccles and Youmans 2016).

Despite its prevalence, shareholder primacy has faced growing criticism for its narrow focus and negative societal impacts. Critics argue that prioritizing shareholder value often undermines the interests of other stakeholders, including employees, customers, communities, and the environment (Bowie 1991; Stoelhorst and Vishwanathan 2024). Additionally, the short-termism associated with shareholder primacy can discourage long-term investments, hinder innovation, and impede sustainable growth (Stout 2012; Shin et al. 2022). In response, stakeholder orientation has gained traction. Rooted in stakeholder theory, it asserts that corporations have ethical and moral obligations to consider the interests of all stakeholders, not just shareholders (Freeman 1984). This perspective views organizations as embedded in a network of interdependent relationships, where stakeholders contribute to and are affected by corporate actions. By emphasizing these interconnections, stakeholder theory advocates for balancing competing interests in corporate decision-making.

The shift from shareholder primacy to stakeholder orientation marks a fundamental shift in institutional logics, prioritizing the interests of all stakeholders in corporate decision-making. Prior research has documented this gradual shift and identified key drivers behind it (Stoelhorst and Vishwanathan 2024). Changing societal expectations and growing concerns over corporate responsibility and sustainability have increased pressure on firms to adopt socially responsible practices (Porter and Kramer 2011; Shin et al. 2022). Additionally, institutional investor initiatives, evolving consumer preferences, and regulatory reforms have played significant roles in reshaping the corporate governance towards a stakeholder-oriented approach (Gibson et al. 2022; Bouguerra et al. 2023).

The moderating role of CSR: amplifying or mitigating effect

Extensive research has investigated the negative impact of adverse events on firm value, including uncontrollable external shocks (e.g., financial crises, pandemics, and industrial scandals) and unethical behaviors (e.g., product recalls, environmental damage, and corporate misconduct). For instance, in the U.S. petroleum industry, oil spills lead to negative stock market reactions (Luo et al. 2018). Product recalls erode public trust, prompting negative responses from both consumers and investors (Chen et al. 2009; Lee et al. 2015; Liu et al. 2020). Similarly, workplace misconduct harms employees and triggers adverse market reactions (Chircop et al. 2025). Corporate crimes and illegal activities further diminish firm value (Davidson et al. 1994; Song and Han 2017; Sun et al. 2025).

Building on the recognition that negative events harm firm value, numerous studies have explored whether CSR mitigates or exacerbates this impact, yet findings remain inconclusive. On one hand, many scholars argue that CSR provides an insurance-like protection, buffering firms against the negative effects of adverse events (Godfrey 2005). CSR builds reputational and moral capital, which can reduce negative stakeholder assessments and associated sanctions during crises. Empirical studies support this protective effect across various types of negative events. For instance, Godfrey et al. (2009) find that firms with stronger CSR experience smaller declines in value following legal or regulatory issues. Similarly, Minor and Morgan (2011) show that CSR mitigates the negative impact of product recalls. Christensen (2016) finds that firms publishing CSR reports suffer smaller value losses when facing lawsuits. In the U.S. petroleum industry, CSR weakens the negative market reaction to oil spills (Luo et al. 2018). Additionally, Bae et al. (2020) report that firms with stronger CSR face less market punishment after financial misconduct is exposed. Oh et al. (2025) confirm that CSR mitigates various risks, including market, financial, and operational risks. Ben-Amar et al. (2025) point out that during the COVID−19 pandemic, CSR could not fully serve as an “insurance-like” buffer, and its positive effect on stock returns is pronounced only for financially flexible firms.

On the other hand, some scholars argue that CSR can amplify the negative impact of adverse events on firm value (Kim and Lee 2015). Firms with strong CSR reputations attract greater stakeholder attention and face higher expectations (Zavyalova et al. 2016). Consequently, when these firms experience negative events, they may be subject to harsher criticism compared to those with weaker CSR reputations. Empirical studies support this negative moderating effect. Research suggests that during crises, CSR may fail to generate value and, in some cases, intensify negative stakeholder reactions (Wang and Qian 2011; Koh et al. 2014; Wang et al. 2021). For instance, Janney and Gove (2011) find that firms with strong CSR reputations are more likely to be perceived as hypocritical and face greater sanctions following corporate governance violations. Similarly, Kim and Lee (2015) highlight that while CSR enhances a firm’s reputation under normal circumstances, it can heighten adverse stakeholder reactions when the firm is implicated in negative events. Liu et al. (2020) reveal that overinvestment in CSR has a boomerang effect on shareholder value when firms announce product recalls. Zhang et al. (2025) find that during an intentional crisis, the stock market reacts more negatively to firms with higher prior CSR performance.

The conflicting findings on CSR’s moderating effect raise an important question: under what circumstances can CSR mitigate or exacerbate the impact of earnings pressure on firm value? Prior research has explored various moderating factors, such as the recurrence of negative events, types of CSR, financial performance, and environmental uncertainty, in shaping CSR’s insurance-like protection effect (Godfrey et al. 2009; Shiu and Yang 2017; Kim et al. 2021). Building on Ioannou and Serafeim (2015), this study takes a dynamic perspective by examining the shifting institutional logics of CSR to better understand this relationship.

Hypotheses Development

From amplifying effect to mitigating effect: the role of transitioning institutional logics

Prior studies have documented that shifts in prevailing institutional logic can alter stakeholders’ interpretations of corporate actions. For instance, Zajac and Westphal (2004) show that market reactions to stock repurchase plans vary with the prevalent logic —positive under a “corporate” logic and negative under an “agency” logic. Similarly, research has examined how market intermediaries, investors, and other external parties evaluate strategies that either align with or challenge the prevailing institutional logic (Philippe and Durand 2011; Ioannou and Serafeim 2015; Shin et al. 2022). In line with this literature, we propose that a transition from shareholder primacy to stakeholder orientation may change shareholders’ interpretations of CSR, ultimately influencing whether CSR mitigates or exacerbates the negative impact of missing analyst forecasts on firm value.

Under shareholder primacy, managers are expected to prioritize maximizing shareholder value over other interests (Jensen and Meckling 1976; Yang et al. 2025). In this context, CSR is often viewed not as a legitimate corporate investment but as managerial misconduct or wasteful expenditure (Friedman 1970). Moreover, managers may use CSR to further their own social, political, or career agendas at the expense of shareholders (Nguyen et al. 2023). Consequently, shareholders may see CSR as value-destroying, favoring enhanced managerial incentives and stricter oversight to curtail CSR spending.

When earnings pressure arises from missing analyst forecasts, shareholders become dissatisfied and react negatively, perceiving that firms have failed to meet expectations (Suchman 1995; Kaplan and Zamora 2018; Heater et al. 2025). Under shareholder primacy, high levels of CSR performance can intensify these negative reactions. Shareholders view CSR as managerial misconduct that erodes firm value and expect firms to limit such expenditures (Friedman 1970; Smith and Rönnegard 2016). Consequently, when dissatisfied shareholders observe substantial CSR investments, their adverse reactions are likely to be amplified.

As institutional logics shift from shareholder primacy to stakeholder orientation, managers prioritize the interests of multiple stakeholders, not just shareholders (Freeman 1984). Under this perspective, CSR is seen as a legitimate corporate investment that enhances competitive advantage by improving reputation (Lloyd-Smith and An 2019), accumulating intangible resources (Khan et al. 2019), reducing risks (Godfrey 2005; Kim et al. 2021), and fostering positive stakeholder responses (Wang and Qian 2011). CSR also attracts talent, enhances employee commitment (Zhao et al. 2022), increases consumer willingness to pay (Ferreira and Ribeiro 2017; Szőcs and Montanari 2025), and strengthens government relationships (Flammer 2018; Chen et al. 2023). Consequently, as stakeholder orientation gains traction, shareholders increasingly perceive CSR as value-enhancing rather than value-destroying (Lu et al. 2023).

Although shareholders may react negatively to earnings pressure resulting from missed analyst forecasts, CSR tends to mitigate these negative reactions when stakeholder orientation prevails. Under stakeholder orientation, shareholders perceive CSR as a value-enhancing investment that aligns with long-term wealth creation, leading them to expect firms to allocate more resources to CSR (Lu et al. 2023). Consequently, dissatisfied shareholders are likely to respond less negatively when they observe firms engaging in CSR, as they recognize its potential benefits.

In summary, we posit that the effect of CSR on the relationship between earnings pressure (i.e., missing analyst forecasts) and firm value depends on the prevailing institutional logic. Specifically, when shareholder primacy dominates, CSR exacerbates shareholders’ negative reactions to earnings pressure. However, as institutional logics shift toward stakeholder orientation, this amplifying effect is expected to weaken and eventually transform into a mitigating effect. Thus, we propose the following:

H1: Over time, the amplifying effect of CSR on the relationship between earnings pressure and firm value is expected to weaken and eventually transition into a mitigating effect.

Heterogeneity analysis

Based on the institutional environment context, we explore how China’s unique characteristics influence shareholders’ perceptions of CSR and moderate the effects of shifts in institutional logic. Specifically, we focus on three dimensions—industry type, ownership structure, and the stringency of regional environmental regulation—which reflect differences in political sensitivity, policy pressure, and stakeholder expectations faced by firms, thereby shaping how shareholders interpret and respond to CSR engagement. Understanding these contextual differences is crucial to accurately capturing the dynamics of shareholder evaluation and the effects of CSR in China’s institutional landscape.

Prior research suggests that shareholders’ responses to CSR may vary depending on whether CSR activities align with or exceed baseline expectations (e.g., Godfrey et al. 2009; Guiral et al. 2020). For firms operating in environments where CSR is not strongly expected, CSR initiatives are more likely to be interpreted as credible commitments that enhance reputational or moral capital. Accordingly, institutional logic shifts may exert a stronger influence on how shareholders evaluate CSR. In contrast, for firms embedded in contexts where CSR is mandated or politically expected—such as SOEs or firms in highly regulated industries—the marginal signaling value of CSR is diminished, weakening the influence of shifting logics. Drawing on this reasoning, we conduct cross-sectional analyses to test how these institutional factors condition the relationship between earnings pressure, CSR and firm value under varying institutional logics.

Industry type

Industry type plays a crucial role in shaping shareholders’ perceptions of CSR. Under shareholder primacy, CSR is generally viewed as value-destroying, meaning that when firms fail to meet analyst forecasts, CSR may exacerbate shareholders’ negative reactions, regardless of industry type. However, as stakeholder orientation becomes more prevalent, shareholders’ interpretations of CSR may diverge between polluting and non-polluting firms (Miras-Rodríguez et al. 2015; Tan et al. 2024).

As major contributors to environmental harm, these firms are expected to engage in CSR as a form of compensation for their negative externalities (Zhang et al. 2024). While their CSR initiatives are acknowledged, they tend to receive lower recognition compared to similar efforts by non-polluting firms. In contrast, CSR activities by non-polluting firms generate greater reputational and moral capital, as they are not perceived as merely corrective measures. Consequently, under stakeholder orientation, shareholders are more likely to view CSR in non-polluting firms as value-enhancing, while the mitigating effect of CSR in polluting firms remains weaker. Thus, we propose the following:

H2: As shareholder primacy shifts toward stakeholder orientation, the positive effect of CSR in alleviating the negative impact of earnings pressure on firm value will be more pronounced for firms in non-polluting industries than for those in polluting industries.

Ownership type

Ownership type also influences how shareholders interpret and respond to CSR (Cooper and Weber 2021). In China, listed firms can be broadly categorized into state-owned enterprises (SOEs) and non-state-owned enterprises (non-SOEs). Under shareholder primacy, shareholders generally perceive CSR as a value-destroying activity. When firms fail to meet analyst forecasts, CSR may further exacerbate shareholders’ negative reactions, regardless of ownership type. However, as stakeholder orientation gains prominence, shareholders may differentiate between the CSR initiatives of SOEs and non-SOEs (Wang and Qian 2011; Sun et al. 2025).

In the Chinese context, SOEs serve as an extension of government policy, fulfilling broader social and political functions such as employment generation, industrial development, and poverty alleviation (Jia et al. 2019). In China, SOEs are evaluated not only based on financial performance but also on their contributions to societal well-being. Given this institutional role, CSR investments by SOEs are often perceived as a mandated obligation rather than a strategic initiative aimed at enhancing firm value. In contrast, non-SOEs have greater discretion in CSR engagement, and their CSR efforts are more likely to generate reputational and moral capital. As a result, investors tend to view SOEs’ CSR as policy-driven and less indicative of long-term commitment. Shareholders therefore perceive CSR from SOEs as less value-enhancing compared to non-SOEs. Based on this, we propose the following:

H3: As shareholder primacy shifts toward stakeholder orientation, the positive effect of CSR in alleviating the negative impact of earnings pressure on firm value will be more pronounced for non-SOEs than for SOEs.

Environmental regulation

The stringency of environmental regulations in a firm’s operating region can influence how shareholders interpret and respond to CSR. Under shareholder primacy, CSR is generally perceived as value-destroying. When firms fail to meet analyst forecasts, CSR may further amplify shareholders’ negative reactions, regardless of the regulatory environment.

However, as stakeholder orientation gains prominence, shareholders may differentiate between firms located in regions with high versus low environmental regulations. In areas with stringent environmental policies, local governments prioritize environmental protection and public welfare, imposing stricter compliance requirements on firms (Ren et al. 2023; Fang et al. 2024; Yang et al. 2025). In regions such as ecological pilot zones, CSR initiatives are often perceived as a mandatory compliance response to regulatory pressures rather than a discretionary, value-enhancing activity. Consequently, CSR in these contexts lacks strong signaling value to investors.

In contrast, in areas with more lenient environmental regulations, firms engaging in CSR are seen as exceeding legal requirements, demonstrating proactive commitment to social responsibility. As a result, their CSR initiatives are more likely to enhance reputation and accumulate moral capital (Tang et al. 2024). Thus, shareholders are less likely to perceive CSR as value-enhancing for firms in high-regulation areas compared to those in low-regulation areas. As a result, the mitigating effect of CSR on the impact of earnings pressure on firm value will be weaker for firms operating in highly regulated regions and stronger for those in less regulated regions. Therefore, we propose the following:

H4: As shareholder primacy shifts toward stakeholder orientation, the positive effect of CSR in alleviating the negative impact of earnings pressure on firm value will be more pronounced for firms in areas with low environmental regulations than for those in regions with high environmental regulations.

The theoretical framework of this study is shown in Fig. 1.

Fig. 1
Fig. 1
Full size image

Theoretical framework.

Methodology

Data and sample selection

This study examines listed companies on the Shanghai and Shenzhen stock exchanges to investigate this shift. In China’s capital market, CSR research commonly relies on Hexun rating scores, which assess annual CSR performance across five dimensions: (i) shareholders, (ii) employees, (iii) customers, suppliers, and consumers, (iv) the natural environment, and (v) social contributions. These dimensions are further broken down into 13 level-2 indicators and 37 level-3 indicators, yielding an overall CSR score, with a maximum possible value of 100 and a higher score indicating better CSR performance. The transition from shareholder primacy to stakeholder orientation began in the 1990s in the U.S. (Ioannou and Serafeim 2015) and has similarly occurred in China since the late 2000s. Since Hexun scores have been available from 2010 to 2020, our dataset covers the period from 2010 to 2020.

Following standard practice in the literature, we applied the following exclusion criteria: (1) firms in the financial industry; (2) firms designated as ST or *ST; (3) firms newly listed or suspended in the current year; and (4) firms with missing financial information. We obtain stock return data from RESSET database, financial information and analyst forecast data from CSMAR database. To mitigate the influence of outliers, all continuous variables are winsorized at the top and bottom 1%. The final sample consists of 19,641 firm-year observations.

Variable definition

Dependent variable: cumulative abnormal return (CAR)

To assess abnormal returns associated with the release of annual reports, we designate the annual report disclosure date as the event date. Given the variation in disclosure timing across firms, each observation is treated as an independent event. Cumulative abnormal return (CAR) is used to measure excess returns. Following common practice in event studies (Howe et al. 1992; Lyon et al. 2013; Zhang et al. 2021), we use [−210, −11] as the estimation window and [−1, +1] as the event window.Footnote 4 In model (1), CARi,(t1, t2) represents the cumulative abnormal return of stock i over the event window from t1 to t2. We use the capital asset pricing model (CAPM) to estimate the excess return.

$$CA{R}_{i,(t1,t2)}=\mathop{\sum }\limits_{t=t1}^{t2}A{R}_{i,t}$$
(1)

Independent variable: earnings pressure (P_Epre)

Building on prior research (Bartov et al. 2002; Matsumoto 2002; Clarke et al. 2021), we measure earnings pressure as the difference between actual and forecasted performance. Specifically, when the mean of analysts’ forecasted earnings per share (FEPS) is above (below) actual earnings per share (AEPS), firms experience positive (negative) earnings pressure. Following Zhang and Gimeno (2010), we define the positive earnings pressure variable as Positive Epre (P_Epre), and the negative earnings pressure variable as Negative Epre (N_Epre), as specified in models (2) and (3):

$$PositiveEpr{e}_{i,t}=FEP{S}_{i.t}-AEP{S}_{i.t},\,{\rm{if}}FEP{S}_{i,t}\ge AEP{S}_{i,t}(=0,\,{\rm{else}})$$
(2)
$$NegativeEpr{e}_{i.t}=AEP{S}_{i,t}-FEP{S}_{i.t},\,{\rm{if}}FEP{S}_{i,t}\le AEP{S}_{i,t}(=0,\,{\rm{else}})$$
(3)

Control variables

Consistent with prior research, we include several control variables to capture factors that may influence firm value. First, we control for firm fundamentals, including firm size (Size), financial leverage (Lev), and operating income (OI), which reflect firm scale, capital structure, and profitability (Wei et al. 2017). Second, we include market indicators, including book-to-market ratio (BM) and institutional shareholding (Inshold), to capture how investors perceive and value the firm (Agarwal et al. 2016). Third, we account for organizational attributes, including firm age (Age), state ownership (SOE), and industry environmental characteristics (Polluting), which reflect differences in organizational maturity, ownership identity, and environmental exposure (Wang et al. 2024). Finally, we control for governance structure, including CEO duality (Duality), board size (BS), board independence (PID), and top 1 shareholding (Top1), to reflect internal control, decision-making concentration, and board monitoring effectiveness (An et al. 2020). Table 1 provides definitions for all variables.

Table 1 Variable definitions.
Full size table

Model specification

Building on the above analysis, we use model (4) to estimate the moderating effect of CSR on the relationship between earnings pressure and firm value:

$$CA{R}_{i,t+1}={\beta }_{0}+{\beta }_{1}P\_Epr{e}_{i,t}+{\beta }_{2}P\_Epr{e}_{i,t}\times CS{R}_{i,t}+\beta \sum Control{s}_{i,t}+{\rm{Year}}+{\rm{Firm}}+{\varepsilon }_{i,t}$$
(4)

The subscript i and t denote firms and years. CARi,t+1 represents the cumulative excess return of firm i following the announcement of its annual report in year t + 1. P_Epre denotes the positive earnings pressure, which refers to the extent of a firm’s actual earnings per share falling below analysts’ forecasts. CSRi,t indicates the corporate social responsibility performance of firm i in year t. Control refers to a set of control variables, which are presented in Table 1. We include firm fixed effects (Firm) to control for time-invariant unobservable differences between firms. Additionally, we include year fixed effects (Year) to address unobserved temporal factors. To account for the dependence of multiple observations per firm, we cluster robust standard errors at the firm level.

We are interested in the interaction term, P_Epre×CSR, whose coefficient, β2, captures the moderation effect of CSR on the impact of negative events on firm value. A negative and significant coefficient on P_Epre×CSR (β2 < 0) indicates that investors view financially underperforming firms with strong CSR more negatively.

Analysis and results

Descriptive analysis

Table 2 presents the descriptive statistics. The mean value of P_Epre is 0.309, while the mean value of N_Epre is 0.021, indicating that analysts are more likely to overestimate rather than underestimate a company’s earnings per share. The mean CSR score is 26.178, with a standard deviation of 15.980, reflecting significant variation in CSR performance across firms. The variance inflation factor (VIF) for the variables is 1.98, suggesting that multicollinearity is not a concern.

Table 2 Descriptive statistics.
Full size table

Baseline results

Table 3 presents the main findings. The first column estimates the model using data from 2010 to 2012, while each subsequent column extends the sample by adding one additional year (e.g., the second column includes data from 2010 to 2013, the third from 2010 to 2014, and so forth). The final column reports the estimated results for the full sample period from 2010 to 2020. This stepwise approach allows us to analyze how the relationship evolves over time. Consistent with Hypothesis 1, the coefficient of P_Epre×CSR is initially significant and negative but gradually becomes less negative and eventually turns significantly positive. These results support our hypothesis, suggesting that shareholders’ reactions to CSR have become progressively less unfavorable over time and ultimately favorable.

Table 3 Baseline results.
Full size table

The results reveal a notable shift in shareholder perceptions of CSR over time. In 2012, the negative relationship between the interaction term (P_Epre×CSR) and CAR is both statistically and economically significant. The coefficient of –0.030 implies that a one-standard-deviation increase in CSR is associated with 47.94% decrease (0.030×15.98/0.969) in firm value’s response to earnings pressure.Footnote 5 However, over the sample period from 2010 to 2020, this relationship gradually turned positive. By 2020, the interaction coefficient increases to 0.022, indicating that a one-standard-deviation increase in CSR is associated with a 72.4% increase (0.022×15.98/0.486) in firm value’s response to earnings pressure.

These findings suggest that in the early 2010s, shareholders react negatively to firms with strong CSR performance but weak financial conditions. By 2013, this negative effect has diminished and is no longer significant. Between 2014 and 2017, shareholder sentiment toward these firms turns positive, although the effect remained statistically insignificant. After 2018, however, shareholders exhibit a significantly positive response to firms with high CSR performance despite earnings pressure. This timeline highlights the evolving nature of shareholder perceptions, reflecting a gradual shift from a shareholder primacy perspective to a stakeholder-oriented view in the context of earnings pressure.

Heterogeneity analysis result

We further investigate how the transition from shareholder primacy to stakeholder orientation varies across firms at the industry, firm, and regional levels.

First, at the industry level, we examine differences in the transition of CSR moderation effects between firms operating in polluting and non-polluting industries.Footnote 6 Panel A in Table 4 shows that for polluting firms, CSR’s moderating effect is negative but not significant from 2010 to 2013, and positive but still insignificant from 2014 to 2017. After 2018, the moderating effect of CSR gradually turns slightly positive. This suggests a shift toward stakeholder orientation among polluting firms, though the trend remains weak. It is consistent with shareholders perceiving CSR activities in polluting industries as an inherent obligation rather than a discretionary effort, thereby limiting the positive moderating effect of CSR when firms underperform.

Table 4 Heterogeneity analysis of industry: polluting vs non-polluting firms.
Full size table

Panel B in Table 4 shows that for non-polluting firms, CSR’s moderating effect is initially negative but not significant from 2010 to 2013. Between 2014 and 2017, this effect becomes positive, albeit still insignificant. However, after 2018, the moderating effect of CSR becomes significantly positive. This result support our conjecture that non-polluting firms face lower CSR expectations compared to polluting firms. As a result, when non-polluting firms engage in CSR activities but fail to meet analysts’ earnings forecasts, shareholders may view their CSR efforts as exceeding expectations, leading to a more favorable evaluation.

Overall, we find that as the focus shifts from shareholder primacy to stakeholder orientation, the mitigating effect of CSR on the impact of earnings pressure on CAR is more pronounced for firms in non-polluting industries than for those in polluting industries.

Second, at the firm level, we explore the differences in the transition of CSR effects between SOEs and non-SOEs. Panel A in Table 5 shows that for SOEs, CSR’s moderating effect is negative but not significant from 2010 to 2013, and while it turns positive from 2014 to 2017, it remains insignificant. After 2018, the moderating effect of CSR becomes slightly positive, indicating a gradual shift in investor perception from shareholder primacy to stakeholder orientation. However, this shift is not statistically significant, suggesting that shareholders view CSR as an inherent responsibility of SOEs rather than a discretionary effort that could enhance firm value under earnings pressure.

Table 5 Heterogeneity analysis of ownership: SOEs and Non-SOEs.
Full size table

Panel B in Table 5 shows that for non-SOEs, CSR’s moderating effect remains negative but insignificant from 2010 to 2017. However, after 2018, the effect becomes significantly positive. This result suggests that non-SOEs face lower CSR expectations than SOEs. When non-SOEs demonstrate strong CSR performance but fail to meet analysts’ earnings forecasts, shareholders may perceive their CSR efforts as exceeding expectations, leading to a more favorable evaluation.

Overall, as the shift from shareholder primacy to stakeholder orientation progresses, the mitigating effect of CSR on the impact of earnings pressure on CAR is more pronounced for non-SOEs than for SOEs.

Third, at the regional level, we explore the differences in the transition of CSR effects between firms located in areas with high and low environmental regulations. Following previous literature (Xie et al. 2017), regional environmental regulation is measured by the ratio of energy-saving and environmental protection expenditure to general budget expenditure.

Panel A in Table 6 shows that for firms located in regions with high environmental regulations, the moderating effect of CSR is negative but insignificant from 2010 to 2013, and positive but insignificant from 2014 to 2017. After 2018, the moderating effect of CSR becomes slightly positive. This trend indicates that investors’ perceptions of CSR for firms in highly regulated areas are shifting from shareholder primacy to stakeholder orientation, but this shift is not statistically significant. It is consistent with that shareholders tend to view CSR in these firms as a response to mandatory regulations, rather than a discretionary effort that could enhance firm value under earnings pressure.

Table 6 Heterogeneity analysis of environmental regulation.
Full size table

Panel B in Table 6 shows that for firms located in regions with low environmental regulations, the moderating effect of CSR is negative but insignificant from 2010 to 2013 and positive but insignificant from 2014 to 2017. However, after 2018, the moderating effect becomes significantly positive. This suggests that firms in regions with low environmental regulations are expected to bear less CSR. When these firms demonstrate strong CSR performance despite failing to meet earnings forecasts, shareholders may perceive this as exceeding expectations, leading to a positive evaluation. Therefore, we find that as shareholder primacy transitions to stakeholder orientation, the mitigating effect of CSR on the impact of earnings pressure on CAR is more pronounced for firms in areas with low environmental regulations than for those in highly regulated regions.

Robustness test

To ensure the robustness of our results, we employ several checks, as shown in Table 7. The first check in Panel A involves altering the measurement approach for the independent variables. Initially, we use the annual report release date as the event date, with an estimation window of [−210, −11] and an event window of [−1, 1] to identify abnormal returns. To test the robustness of our findings, we adjust the estimation window to [−10, 10], and recalculate the cumulative abnormal returns (CAR) using this alternative window. This adjustment allows us to assess whether the results remain consistent across different time frames, thereby enhancing the validity of our conclusions and ensuring that our findings are not dependent on the initial window settings.

Table 7 Robustness test.
Full size table

The second robustness check in Panel B examines alternative measurements of the independent variable. Following Brown et al. (2022), we replace the mean with the median of analysts’ forecasts and recalculate positive earnings pressure (P_Epre2) and negative earnings pressure(N_Epre2). As shown in Table 7, this adjustment did not significantly alter the results, reinforcing the robustness of our measurement approach.

The third robustness check involves controlling for alternative fixed effects. While our primary models included firm and year fixed effects, Panel C extends this by incorporating industry×year fixed effects to account for industry-specific variations over time. As shown in Table 7, this refinement does not alter our conclusions, confirming that our findings remain robust even after accounting for additional fixed effects.

Fourth, we test the robustness of our results by adjusting the sample. To ensure consistency across time periods, we select 492 companies with complete financial and CSR data from 2010 to 2020. This fixed sample eliminates potential biases from missing data and allows for more stable regression analyses. The results in Panel C remain consistent, confirming the robustness of our findings and enhancing the reliability of our conclusions.

Conclusion and discussion

This study examines how shifts in mainstream institutional logic, from shareholder primacy to stakeholder orientation, affect the moderating role of CSR in the relationship between earnings pressure and firm value. Grounded in stakeholder theory (Freeman 1984), we argue that CSR influences firm value by shaping stakeholder perceptions and expectations, which are contingent on the prevailing institutional logic. Drawing on data from Chinese listed firms from 2010 to 2020, we find that the amplifying effect of CSR on the negative impact of earnings pressure gradually diminishes over time, eventually reversing into a mitigating effect. This dynamic pattern highlights that the perceived value and legitimacy of CSR are contingent upon evolving societal expectations rather than fixed.

Prior research presents mixed evidence regarding CSR’s moderating effect under negative events. Some studies suggest that CSR can amplify the adverse impact because investors may perceive CSR spending as a misallocation of resources (Christensen 2016; Luo et al. 2018; Bae et al. 2020; Oh et al. 2025). In contrast, other studies report a mitigating effect, emphasizing that CSR helps preserve legitimacy and stakeholder trust (Janney and Gove 2011; Kim and Lee 2015; Liu et al. 2020; Zhang et al. 2025). Our findings extend the stakeholder theory by showing that CSR’s impact on firm value depends on the institutional logic, highlighting the importance of contextual interpretation by stakeholders.

In the early period, when shareholder primacy dominated, financial performance served as the central criterion of legitimacy. In this context, CSR was interpreted as a resource diversion from profitability, amplifying the negative consequences of earnings pressure. As stakeholder orientation gradually became institutionalized, however, the evaluative standards of legitimacy broadened beyond short-term financial outcomes. Under this logic, CSR increasingly functioned as a forward-looking investment signaling resilience and long-term value creation. The observed transition from amplification to mitigation thus reflects a temporal shift in institutional logics and a reconfiguration of the market’s perception of CSR activities.

Heterogeneity analysis further reveals that this institutional transition is particularly salient for certain firms, such as non-polluting firms, non-SOEs, and firms located in regions with less stringent environmental regulation. These firms often engage in CSR activities that go beyond their core obligations, and thus receive additional benefits when aligning with evolving societal expectations, consistent with Tang et al. (2024). In contrast, firms such as polluting firms, SOEs, or those in heavily regulated regions face stable and policy-driven expectations, which constrain the benefits they can gain from CSR activities. Consistent with Jia et al. (2019), Ren et al. (2023), and Fang et al. (2024), the additional market rewards for these firms are limited, thereby reducing the marginal effect of shifts in institutional logic on their CSR engagement.

Theoretical contributions

Our study contributes to the literature in several ways. First, it contributes to the debate on whether CSR mitigates or amplifies the impact of negative events on firm value by applying an institutional logic perspective. Grounded in stakeholder theory (Freeman 1984), we argue that CSR affects firm value by shaping stakeholders’ perceptions, which in turn are influenced by prevailing institutional logics, including shareholder primacy and stakeholder orientation. Prior research has yielded mixed findings on CSR’s moderating effect, due to differences in performance measures, event alignment, CSR targets, societal culture, and consumer-brand connections (Godfrey et al. 2009; Janney and Gove 2011; Zeidan 2013; Shiu and Yang 2017; Zhang et al. 2023, 2025; Hoang and Phang 2023; Kim et al. 2024; Gutknecht 2024; Ryoo 2025). Integrating stakeholder theory with an institutional logic lens, we highlight the contingent mechanisms through which CSR can either mitigate or amplify the effect of earnings pressure on firm value.

Second, it extends research on CSR’s moderating role by examining earnings pressure as a new context. Prior studies have focused on CSR’s influence in external crises or unethical corporate conduct, such as pandemics, financial crises, product recalls, and financial irregularities (Minor and Morgan 2011; Shiu and Yang 2017; Poursoleyman et al. 2024). By investigating earnings pressure, our study offers a more comprehensive understanding of CSR’s moderating role in business practices. Unlike external crises or unethical corporate conduct, earnings pressure stems from market expectations, such as analyst forecasts. Examining this context not only broadens the scope of CSR research, but also reveals how CSR interacts with routine financial pressures that directly affect firm value and market reactions.

Third, our study advances understanding of institutional logic shifts by highlighting the transition from shareholder primacy to stakeholder orientation in emerging markets. Prior research has primarily focused on developed economies. For instance, Ioannou and Serafeim (2015) examine how the rise of stakeholder orientation over time influences analysts’ perceptions of CSR in U.S. firms, while Shin et al. (2022) investigate CEO dismissals in financially underperforming firms with strong CSR performance using U.S. data. Our findings demonstrate that this institutional shift is not confined to developed markets but is also highly relevant in emerging economies. In the Chinese context, CSR has developed alongside institutional shifts from state-planned mandates to market-driven legitimacy mechanisms. Focusing on an emerging market, this study offers new insights into the evolving role of CSR within the global institutional landscape. By explicitly integrating China’s unique institutional characteristics, our research deepens the understanding of CSR’s changing function in emerging markets beyond the Western context.

Practical implications

Our study also has important practical implications for investors, managers, and policymakers. First, investors should evaluate CSR performance within its specific institutional context rather than applying a uniform standard. By considering the dominant institutional logics, they can more accurately assess how CSR contributes to sustainable competitive advantages and improve portfolio performance. In shareholder-oriented environments, CSR is often viewed as a nonessential cost, especially when firms underperform. Conversely, in stakeholder-oriented environments, CSR is increasingly seen as a positive signal of long-term value creation and legitimacy. Understanding these differences helps investors make informed decisions aligned with changing market dynamics.

Second, managers should actively align CSR strategies with prevailing institutional logics. When stakeholder orientation dominates, CSR becomes a strategic tool to stabilize firm value and sustain legitimacy during periods of financial strains. Managers can enhance CSR effectiveness by prioritizing substantive stakeholder engagement and ensuring strategic alignment with long-term value creation, especially in contexts where CSR expectations are rising. Additionally, integrating CSR into strategic and operational processes can help firms build deeper stakeholder trust and improve resilience against external shocks.

Third, policymakers can foster meaningful CSR engagement by introducing incentive-compatible policies, such as integrating CSR into performance evaluations, tailoring standards to ownership structures, and easing compliance burdens for firms with high-quality disclosures. Our findings highlight the critical role of institutional support in promoting substantive CSR in emerging markets. By embedding stakeholder-oriented expectations within institutional frameworks, policymakers can help firms internalize externalities and align corporate behavior with broader societal goals.

Limitations and recommendations for future research

This research has several limitations that could create avenues for further studies. First, our analysis covers the period 2010–2020, as Hexun CSR scores are consistently available only for this period, with limited coverage thereafter. Although this period captures important institutional changes in China, a decade of data may not be sufficient to capture the long-term evolution of CSR or the broader shift from shareholder primacy to stakeholder orientation. Future research could extend the time frame and incorporate more recent ESG (Environmental, Social, and Governance) metrics to examine whether the observed patterns persist or vary under different institutional and market conditions. Such an extension would enhance the temporal validity of our findings and provide deeper insights into the dynamic role of CSR across different stages of China’s economic transformation.

Second, although this study focuses on earnings pressure as a specific type of negative event, firms are exposed to a wide range of adverse events. These include industry scandals, public crises, product recalls, financial misconduct, and environmental violations, which vary in severity, visibility, and potential impact on firm value. The moderating effect of CSR may therefore differ depending on the type and context of the event. While our findings show that CSR shifts from amplifying to mitigating the adverse impact of earnings pressure, future research should examine whether the observed institutional logic transition applies to other types of negative events. Analyzing multiple event types would help delineate the boundary conditions of CSR’s moderating role and provide deeper insights into how firms can strategically leverage CSR to navigate diverse adverse situations. Moreover, exploring how event characteristics interact with CSR may reveal more nuanced mechanisms through which CSR influences firm value across different contexts.

Third, another limitation of this study relates to the measurement of CSR. While we rely on the Hexun overall CSR score to provide a broad overview of corporate social responsibility, CSR encompasses diverse practices, including environmental sustainability, employee welfare, and community engagement. Aggregating these dimensions into a single score may obscure how specific CSR activities differentially moderate the impact of earnings pressure on firm value. Future research could adopt a more granular approach, examining individual CSR dimensions to identify which aspects are most influential under varying institutional contexts. Additionally, incorporating alternative CSR measures, such as third-party ESG ratings or textual analyses of CSR reports, could provide deeper insights into the mechanisms through which CSR shapes firm value across different types of negative events.