CSR and Stock Price Crash Risk: Does the Firm Life Cycle Matter? An Emerging Economy Perspective

Abstract

Corporate social responsibility (CSR) plays a growing role in fostering transparency, stakeholder trust, and long-term firm sustainability, particularly in emerging markets. Firms that actively engage in CSR are more likely to disclose credible financial information, which can reduce the incentive to withhold adverse news and thereby limit stock price crash risk (SPCR). This study investigates the impact of CSR on SPCR, while also examining whether this relationship varies across different stages of the firm life cycle (FLC). The analysis is based on an unbalanced panel of listed non-financial firms from the Pakistan Stock Exchange (PSX), covering the period from 2009 to 2023. Financial data were obtained from the State Bank of Pakistan (SBP) and Securities and Exchange Commission of Pakistan (SECP), while market data were collected from the PSX. Employing fixed-effects robust regression models and two crash risk proxies, negative conditional skewness (NCSKEW) and down-to-up volatility (DUVOL), the results reveal a consistent and significant negative association between CSR and SPCR. This suggests that firms with stronger CSR engagement are less prone to extreme negative stock returns. However, the moderating effect of FLC is only evident at the introduction and decline stages, indicating that the effectiveness of CSR in reducing crash risk depends on a firm’s position in its organizational life cycle. These findings contribute to the literature on CSR and financial stability in emerging markets and offer practical implications for investors, managers, and policymakers seeking to promote risk-aware, socially responsible corporate strategies.

1. Introduction

Corporate social responsibility (CSR) reflects a firm’s deliberate commitment to addressing social and environmental issues as part of its broader strategic and operational framework (Ness, 1992). By aligning corporate goals with stakeholder expectations, CSR helps firms build a responsible public image and cultivate trust. Increasingly, large organizations recognize corporate social performance as an equally important metric alongside financial performance when assessing long-term success (Akisik & Gal, 2011). As CSR becomes a mainstream element of corporate governance, it has also gained importance in finance and accounting research (Narullia & Subroto, 2018). While prior research primarily investigated the relationship between firm performance and CSR (Bams et al., 2021), insufficient attention has been given to its implications for firm-specific risk, particularly stock price crash risk (SPCR). Most earlier studies have focused on the interrelation of corporate governance and CSR (El Gammal et al., 2020; Lu et al., 2021), with relatively little investigation into how CSR may influence crash risk through managerial disclosure behavior.

Enterprises with high level of CSR engagement are often more open and less prone toward earnings management or bad news hoarding (Schuler & Cording, 2006). This behavior becomes particularly relevant in today’s context, as firms, investors, and regulators increasingly seek to align business operations with environmental, social, and governance (ESG) standards. CSR has gained prominence in managerial agendas (Albuquerque et al., 2019; Dunbar et al., 2020), prompting firms, especially in developing economies, to channel resources toward socially responsible initiatives (Deng et al., 2013). Given that firms in emerging markets often operate in less mature institutional environments, their approach to CSR may differ significantly from those in developed markets (Kao et al., 2018). Firms in such economies are increasingly aware that reputational capital gained through CSR can help build investor confidence and stakeholder loyalty.

SPCR has become a critical consideration in financial risk management. Triggered in part by the Global Financial Crisis (GFC), recent academic and industry interest has turned to understanding the causes and dynamics of crash risk. Research shows that managers often conceal negative information to safeguard short-term interests or career incentives (Hutton et al., 2009). When these concealed negative signals accumulate beyond a threshold, their eventual disclosure results in a sharp and sudden share price collapse, a crash (Chen et al., 2017). This phenomenon is not only damaging to investors but may also lead to resource misallocation, reduced firm credibility, and broader market instability. Moreover, crash risk is increasingly used as a forward-looking indicator of investor sentiment, particularly in sectors like utilities and energy that are central to economic growth (Huang & Liu, 2021). Timely and transparent disclosure reduces crash risk by ensuring that prices reflect available information, thereby limiting the opportunity for speculative behavior or abnormal returns based on asymmetric information (Y. Kim et al., 2014).

The life cycle of a firm is another critical lens through which business risk and behavior can be understood. The firm life cycle (FLC) reflects how internal dynamics and external forces—such as innovation, competition, and strategic decisions—shape a firm’s evolution over time Dickinson (2011). Different stages of the FLC are associated with varying levels of profitability, information asymmetry, and operational uncertainty (Fama & French, 2000). As firms transition through stages—from introduction to decline—their reporting behaviors and exposure to investor skepticism also change (Hong & Stein, 2003). Firms in early stages, for example, may be more opaque or prone to speculative valuations, while mature firms often operate under tighter governance structures.

Despite growing attention to CSR and SPCR, the literature offers conflicting perspectives. Some scholars argue that CSR enhances disclosure quality and reduces managerial opportunism, thereby lowering crash risk. Others suggest that CSR may be used to mask poor performance or divert attention from negative developments. The relationship becomes even more ambiguous when viewed through the lens of a FLC, which may shape how CSR is implemented and perceived. This research focuses on analyzing the effect of CSR on SPCR to examine whether the FLC stage conditions this relationship. In light of the above backdrop, this study has framed two questions to be addressed: (a) Does CSR reduce SPCR? (b) Does the FLC moderate the relationship between CSR and SPCR?

Supported by stakeholder theory, we argue that enterprises with robust CSR practices are likely to face reduced SPCR, as socially conscious firms are typically subject to greater scrutiny, maintain higher ethical standards, and are more transparent with stakeholders (Cheng et al., 2014). Prior studies have found that CSR limits managers’ ability to conceal bad news (Y. Kim et al., 2014) and may even improve firm performance through stronger governance (Jo & Harjoto, 2011). However, most existing work focuses on developed economies, and scant attention has been given to how CSR functions as a risk mitigation approach in emerging markets like Pakistan. Moreover, the potential interaction between CSR and FLC in influencing crash risk has received limited empirical attention. By filling this gap, this study endeavors to provide insights into more nuanced understanding of the conditions under which CSR influences firm-specific downside risk.

This study finds that CSR is negatively and significantly associated with SPCR, implying that firms with higher CSR involvement face a lower risk of sudden and extreme declines in stock returns. Additionally, the FLC moderates this relationship: CSR is associated with higher crash risk in the introduction stage, likely due to symbolic implementation, but shows stronger risk-reducing effects in the decline stage, where stakeholder trust and reputational capital are more critical. These findings highlight the context-dependent nature of CSR’s impact on crash risk and provide important implications for risk management, investor decision-making, and corporate governance in emerging markets.

The subsequent sections are structured as follows: Section 2 examines the pertinent literature and theoretical background. Section 3 outlines the data, variables, and empirical methodology. Section 4 outlines the study results and discussion. Section 5 concludes with policy implications and directions for future research.

2. Literature Review

Contrary to shareholder theory, stakeholder theory focuses on maintaining stakeholder interests. The stakeholder believes goals may be met by fulfilling the interests of stakeholders. Stakeholders cannot be identified in the absence of CSR, and stakeholder viewpoints can only be fulfilled in the presence of CSR. CSR essentially reduces the inclination of negative news hoarding, and storing unfavorable news increases crash risk, so how CSR might lower business-specific risks should be examined. Prior investigation has yet to thoroughly analyze the nexus between CSR and SPCR. The findings remain uncertain. In particular, the presence of FLC as an external reference has not been investigated in the context of Pakistan. As a result, this topic is currently receiving a lot of attention.

2.1. Stock Price Crash Risk (SPCR)

SPCR denotes the likelihood of severe negative stock returns triggered by the abrupt disclosure of accumulated adverse information. Srivastava et al. (2024) presented the emerging trends in crash risk. Crash risk is typically measured via the negative skewness of firm-specific returns (Callen & Fang, 2015; Jin & Myers, 2006). Underlying SPCR is a classic agency problem: managers motivated by self-interest delay the disclosure of bad news in opaque reporting, leading to information asymmetry. Once the hoarded bad news exceeds a threshold, it precipitates a sharp price collapse (Jin & Myers, 2006). In other words, larger skewness of returns signals higher crash risk (Jin & Myers, 2006; Conrad et al., 2013).

Researchers have identified many empirical determinants of crash risk. Key factors include

• Earnings management: Aggressive accrual or real earnings management tends to precede crashes, whereas conservative accounting reduces crash risk (Francis et al., 2016; J.-B. Kim & Zhang, 2016).
• Governance and culture: Weak corporate governance (e.g., poor board oversight) increases SPCR, while strong governance mitigates it (Hutton et al., 2009; An & Zhang, 2013). Cultural factors like religiosity also matter—more religious norms are associated with lower crash risk (W. Li & Cai, 2016).
• Managerial traits: CEO or top-manager characteristics influence SPCR. For example, overconfident CEOs are more prone to crash-inducing behavior (Francis et al., 2016; J.-B. Kim & Zhang, 2016). Likewise, higher managerial ability (signaling better decision-making) has been linked to lower crash risk (Habib & Hasan, 2017b).
• Market factors: Environmental factors like stock liquidity matter; firms with more liquid stocks tend to exhibit higher crash risk (Chang et al., 2017). In general, younger, high-growth firms face more uncertainty and crash vulnerability (Fama & French, 2000; Hong & Stein, 2003).

Together, the theory and evidence imply that SPCR fundamentally stems from information asymmetry and conflicting incentives between managers and shareholders. This suggests that any mechanism that increases transparency or aligns stakeholder interests could reduce crash risk. For example, shareholder monitoring or higher disclosure quality is expected to limit the hiding of bad news. In this context, CSR is a candidate mechanism to improve transparency. Based on the above, we examine whether CSR—by reducing opacity—helps mitigate SPCR (H1).

2.2. Corporate Social Responsibility (CSR)

CSR reflects a firm’s commitment to stakeholders (employees, community, environment) beyond profit-making. Theories of CSR draw on stakeholder theory (Freeman, 1984) and legitimacy theory, which suggest that firms engage in social initiatives to align with the expectations of stakeholders and broader societal norms. By honoring obligations to stakeholders, CSR can reduce agency conflicts and discourage opportunistic behavior (Jo & Harjoto, 2011). Legitimacy theory similarly views CSR as building “moral capital” that sanctions responsible behavior and deters misconduct. In practice, CSR involvement has been linked to higher quality voluntary disclosure and stakeholder trust. For instance, Gelb and Strawser (2001) find that firms engaging in socially conscious activities are “more likely to show high transparency and rarely hide bad news”. Such firms tend to provide more reliable information and exhibit stronger accountability (Gelb & Strawser, 2001; Clarkson, 1995). CSR is a broad concept that incorporates a number of concepts and principles known by various names, including CSR, governance practices, business ethics, stakeholder involvement, responsible capitalism, environmental sustainability, and the environment, among many others (Lu et al., 2018). When a company is going through a crisis, moral capital can act as a risk buffer, preserving firm value (Boubaker et al., 2022; Wu et al., 2024). Further, Yang et al. (2023) found negative and significant impact of CSR on SPCR.

These traits have direct relevance for crash risk. Firms that actively pursue CSR are expected to refrain from withholding negative information; they behave more ethically and transparently. Indeed, empirical studies generally find a negative CSR–SPCR link. For example, Hunjra et al. (2020) document that better CSR performance significantly reduces crash risk in Pakistani and Indian firms. They note that when firms engage in CSR, “they refrain from bad news hoarding, which results in decreasing SPCR.” Similarly, Y. Kim et al. (2014) report lower crash likelihood for firms with proactive CSR disclosure. These findings support the idea that CSR-driven transparency mitigates the buildup of hidden bad news.

However, evidence is not entirely uniform. Some recent work, especially in emerging markets, finds mixed or even opposite effects. One study (Z. Li & Wu, 2025) used Chinese data to show that, under certain conditions, CSR can increase crash risk. They argued that investors may view high CSR spending skeptically (due to its costs or “rhetoric”), which can heighten crash risk, unless the local business environment improves. This highlights that the CSR–SPCR relationship can be contingent on context. Overall, most studies point to CSR as a tool for limiting information opacity (through stakeholder accountability), but the evidence is still evolving, especially in developing economies. Notably, Pakistani firms in particular have received little attention beyond the Hunjra et al. (2020) study. Thus, there is a gap in understanding how CSR affects crash risk in Pakistan’s market, and whether theoretical frameworks like legitimacy theory apply there. In summary, theory suggests that CSR should reduce crash risk by curbing the hoarding of negative news (stakeholder and legitimacy perspectives), and most empirical findings are consistent with CSR attenuating SPCR. Building on the previous discussion, this study proposes the following hypothesis:

H1. 

CSR negatively influences SPCR.

2.3. Firm Life Cycle (FLC)

FLC refers to the stages of a firm’s development, typically introduction, growth, maturity, shake-out, and decline (Gort & Klepper, 1982). Each stage is distinguished by different resource needs, investment opportunities, and risk profiles. For example, Dickinson (2011) shows that profitability patterns and cash flow characteristics vary significantly across FLC stages. Likewise, Fama and French (2000) note that firms that are heavily reliant on future growth have greater uncertainty. Hong and Stein (2003) argue that such high-growth (immature) firms are prone to greater fluctuations and crash risk due to broader belief heterogeneity. In contrast, mature firms have more predictable earnings and lower information risk.

These life-cycle effects can shape the impact of CSR on crash risk. Younger, fast-growing firms may have limited capacity for CSR or may prioritize growth over stakeholder transparency. In contrast, mature firms often have more established social obligations and stakeholder relationships, so CSR initiatives may carry greater legitimacy and be more credible. Prior research finds that FLC stage moderates other CSR outcomes (e.g., CSR–performance links) (Jan et al., 2021). Similarly, Habib and Hasan (2017a) find that corporate risk-taking follows life-cycle patterns. Therefore, it is reasonable to expect that the effect of CSR in reducing crash risk is contingent on the FLC stage. For instance, CSR might be most effective at reducing crash risk in mature firms that have the organizational stability to implement CSR meaningfully.

In the context of Pakistan, no prior study has examined the interplay of CSR, crash risk, and FLC. This gap motivates our investigation of FLC as a moderator. In line with the above reasoning, we propose that the strength of CSR’s influence on SPCR will vary by FLC stage. Hence,

H2. 

FLC moderates the relationship between CSR and SPCR.

3. Methodology

3.1. Sampling and Data Collection

This study is based on an unbalanced panel dataset comprising all listed non-financial firms on the Pakistan Stock Exchange (PSX) and part of KSE-100 index over the period from 2009 to 2023. Firms from the financial sector were excluded due to their distinct regulatory structure and accounting practices. Weekly stock price data were gathered from the official websites of PSX and Business Recorder, while annual financial and accounting information was obtained from the SBP and SECP. All continuous variables were winsorized at the 1% level on both ends to limit the effect of extreme outliers. Firms were excluded if they had fewer than 26 weekly return observations in any year or less than seven years of available data across the sample period, in order to mitigate thin-trading effects and ensure robustness in crash risk measurement. Applying these criteria yielded a final sample of 917 firm-year observations drawn from qualifying non-financial firms. Diagnostic tests were conducted to examine data quality and ensure suitability for panel regression analysis. Prior to estimation, summary statistics were computed to offer an overview of the key variables’ central tendencies and dispersion.

3.2. Measurement of Variables

For hypothesis testing, the study employed SPCR as dependent variable, CSR as independent variable and FLC as moderator. A number of control variables were employed, such as return on asset (ROA), firm size (FS), leverage (LEV), standard deviation of weekly returns (SIGMA), and average weekly returns (RET).

This investigation adopted two proxies to measure firm-specific SPCR such as negative conditional skewness (NCSKEW) and down-to-up volatility (DUVOL) (Chen et al., 2001). Both NCSKEW and DUVOL rely on the extended market model (Y. Kim et al., 2014). Following Chen et al. (2001) and Hutton et al. (2009), the model was used to draw residuals for the measurement of firm-specific weekly returns (FSWRs). The extended market model is shown below:

$${\rho }_{i,\tau } = {\alpha }_i + {\theta }_{1,i} {\rho }_{m,\tau -2} + {\theta }_{2, i} {\rho }_{m,\tau -1} + {\theta }_{3,i} {\rho }_{m,\tau } + {\theta }_{4,i} {\rho }_{m,\tau +1} + {\theta }_{5,i} {\rho }_{m,\tau +2} + {\epsilon }_{i,\tau }$$

(1)

where ${\rho }_{i,\tau }$ represents the stock return of firm i for week τ, ${\rho }_{m,\tau }$ represents the value-weighted market return (VWMR) for week τ, and ${\epsilon }_{i,\tau }$ is an error term of firm i for week τ. Two lags and lead values of VWMR were considered in computation to account for non-synchronous trading (Dimson, 1979). FSWR (${\omega }_{i,\tau }$) of firm i for week τ was then determined by using residual return from Equation (1):

$${\omega }_{i,\tau } = {\ln (1+\epsilon }_{i,\tau })$$

(2)

NCSKEW is the proxy to measure SPCR based on skewness of the returns distribution (Zaman et al., 2021). This approach limits the asymmetries of returns distribution and has been described numerous times in prior studies. NCSKEW was calculated by taking the inverse of the 3rd moment of FSWR every year, stabilizing it by raising the standard deviation (SD) of FSWR to the 3rd power, and was multiplied by −1. NCSKEW with higher value reflects a greater crash risk. NCSKEW is computed as follows:

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

(3)

The other measure of SPCR is DUVOL. FSWR results are split into two main sections: ‘weeks with down returns’ and ‘weeks with up returns.’ The ‘down-weeks’ have lower returns than the annual mean, while the ‘up-weeks’ have higher returns. The SD of FSWR is then computed for each category. Finally, ${DUVOL}_{i,t}$ is measured as:

$${DUVOL}_{i,t}= log{\displaystyle \frac{\left[{(n}_u-1){(\sum }_{Down}{\omega }_{i,\tau }^2)\right]}{\left[{(n}_d-1){(\sum }_{Up}{\omega }_{i,\tau }^2)\right]}}$$

(4)

where $n_u$ stands for the total of up-weeks, and $n_d$ measures the count of down-weeks for firm i in year t. A high crash risk indicates a greater ${DUVOL}_{i,t}$ value (Chen et al., 2001).

Following Feng et al. (2018) and Javeed and Lefen (2019), in this study, CSR was measured by social contribution value per share (SCV). SCV was developed by using an equation that addresses the environmental, sociological, and economic elements of emerging economies’ CSR policies. It also helped us in understanding the societal performance of the business organization’s and how they lead to community ideals. This index includes all criteria required for social benefits. The included factors represent the generation of value for various stakeholders, such as EPS for its shareholders; taxation for society; staff expenses for employees; interest charges for lenders; and public welfare expenses for the community. Furthermore, environmental contamination was considered as a social cost and hence eliminated. CSV was calculated by using the following formula:

$$SCV= EPS +{\displaystyle \frac{\left(Total Tax + Employees’ Salaries + Public Welfare Expenses + Interest on Loan - Social Cost\right)}{Total Equity}}$$

FLC was considered as moderating variable. The literature is divided on the optimal method to measure FLC (Jan et al., 2021). Different researchers propose varying sets of FLC stages for a thorough examination (Argyres & Bigelow, 2007; Matos & Hall, 2007). Following Dickinson (2011), this study employed cash flow pattern (CFP) as a proxy to measure FLC stages. The combination of CFP represents the interaction of organizations’ resource utilization and ability to execute with their strategic decisions (

Table 1. Firm life cycle stages.

Stages	Introduction	Growth	Maturity	Shake-Out			Decline
Cash Flows	1	2	3	4	5	6	7	8
Operating	−	+	+	−	+	+	−	−
Investing	−	−	−	−	+	+	+	+
Financings	+	+	−	−	+	−	+	−

(Retrieved from: Dickinson, 2011, p. 1974).

).

A variety of variables that could be the drivers of SPCR were controlled in this study (Habib et al., 2018). Profitable and large-sized firms have a higher probability of facing negative reactions from investors; therefore, firm size (FS) and return on asset (ROA) were controlled (Jebran et al., 2019; Jebran et al., 2022). Similarly, the stock price crash for a firm may also be caused by a high debt-to-asset ratio (Zaman et al., 2021), which requires firm leverage (LEV) to be controlled. Average weekly returns (RET) and SIGMA, representing the SD of FSWR of firm i at time t, were controlled as well. FS is measured as the natural log of total assets (Widyaningsih et al., 2017).

3.3. Model Specification

In order to assess the effect of CSR on SPCR, the following models were estimated, taking into account two separate proxies of SPCR:

Model for the 1st proxy of SPCR (NCSKEW):

$${NCSKEW}_{i,t} = {\alpha }_i + {\beta }_1{CSR}_{i,t-1}+{\phi }_{j}Control Variable+{\epsilon }_{i,t}$$

(5)

where ${CSR}_{i,t-1}$ represents social contribution value for firm i in year t − 1.

Model for the 1st proxy of SPCR (NCSKEW), with FLC stages as the moderator:

$${NCSKEW}_{i,t} = {\alpha }_i+{\beta }_1{CSR}_{i,t-1}+{\beta }_2{FLCSTG}_{i,t-1}+{\beta }_3{CSR}_{i,t-1}\ast {FLCSTG}_{i,t-1}+{\phi }_{j}Control Variables +{\epsilon }_{i,t}$$

(6)

where ${FLCSTG}_{i,t-1}$ represents the respective FLC stage for firm i in year t − 1.

Model for the 2nd proxy of SPCR (DUVOL):

$${DUVOL}_{i,t} = {\alpha }_i + {\beta }_1{CSR}_{i,t-1} +{\phi }_{j}Control Variable+ {\epsilon }_{i,t}$$

(7)

Model for the 2nd proxy of SPCR (DUVOL), with FLC stages as the moderator:

$${DUVOL}_{i,t} = {\alpha }_i+{\beta }_1{CSR}_{i,t-1}+{\beta }_2{FLCSTG}_{i,t-1}+{\beta }_3{CSR}_{i,t-1}\ast {FLCSTG}_{i,t-1}+{\phi }_{j}Control Variable+ {\epsilon }_{i,t}$$

(8)

Notably, this study employed five FLC stages, including introduction, growth, maturity, shake-out, and decline. The models were estimated for each stage separately for both proxies of SPCR.

4. Results

4.1. Descriptive Statistics

Table 2. Descriptive statistics.

Variables	Mean	SD	Min	P25	Median	P75	Max	N
NCSKEW	0.19	0.98	−1.46	−0.37	0.03	0.48	4.77	982
DUVOL	0.08	0.57	−1.05	−0.28	0.01	0.38	2.13	982
CSR	0.22	0.16	−0.08	0.10	0.190	0.30	0.82	975
INTRO	0.01	0.12	0.00	0.00	0.00	0.00	1.00	1080
GROWTH	0.18	0.38	0.00	0.00	0.00	0.00	1.00	1080
MATURE	0.34	0.47	0.00	0.00	0.00	1.00	1.00	1080
SHAKEOUT	0.09	0.28	0.00	0.00	0.00	0.00	1.00	1080
DECLINE	0.01	0.10	0.00	0.00	0.00	0.00	1.00	1080
FS	17.35	1.34	14.15	16.52	17.28	18.17	20.61	1011
ROA	9.49	10.55	−23.00	3.00	9.00	15.00	40.00	1013
LEV	0.41	0.16	0.00	0.26	0.37	0.51	0.87	1013
SIGMA	0.05	0.02	0.02	0.03	0.04	0.05	0.15	982
RET	−0.13	0.14	−1.04	−0.14	−0.09	−0.05	−0.01	982

summarizes the descriptive statistics of all study variables for the years 2009 to 2023. The mean values of crash risk proxies, NCSKEW (0.19) and DUVOL (0.08), suggest mild downside risk on average, but the wide ranges indicate substantial variation across firms. CSR, measured as social contribution value per share (SCV), averages 0.22, with values ranging from −0.08 to 0.82, reflecting a broad dispersion in social engagement. Most firms fall into the mature (34%) and growth (18%) stages of the FLC, while very few are in introduction or decline phases. Control variables such as firm size (mean log assets = 17.35), ROA (9.49%), and idiosyncratic volatility (5%) show expected patterns for an emerging market. Average leverage is 41%. Notably, the average weekly return is negative (−0.13), underscoring challenging market conditions during the study period. Overall, the data exhibit strong heterogeneity in CSR activity, risk, and firm characteristics, providing a solid basis for further regression analysis.

4.2. Correlation Analysis

Table 3. Correlation analysis.

	[1]	[2]	[3]	[4]	[5]	[6]	[7]	[8]	[9]	[10]	[11]	[12]	[13]
NCSKEW [1]	1
DUVOL [2]	0.895 ***	1
CSR [3]	0.102 ***	0.082 **	1
Intro [4]	−0.033	−0.044	0.006	1
Growth [5]	−0.022	0.002	−0.110 ***	−0.056 *	1
Mature [6]	−0.093 ***	−0.086 ***	0.196 ***	−0.085 ***	−0.337 ***	1
Shake-out [7]	0.008	0.007	0.017	−0.037	−0.145 ***	−0.222 ***	1
Decline [8]	−0.041	−0.022	−0.165 ***	−0.011	−0.045	−0.069 **	−0.030	1
ROA [9]	−0.020	−0.027	−0.330 ***	−0.091 ***	0.127 ***	0.107 ***	0.116 ***	−0.036	1
EPS [10]	0.210 ***	0.183 ***	0.643 ***	−0.228 ***	−0.153 ***	0.144 ***	0.016	−0.226 ***	−0.085 ***	1
FS [11]	0.031	0.01	−0.145 ***	0.109 ***	0.097 ***	−0.148 ***	−0.017	0.069 **	0.178 ***	−0.251 ***	1
SIGMA [12]	0.199 ***	0.160 ***	−0.038	0.100 ***	−0.033	−0.194 ***	−0.059 *	0.126 ***	−0.384 ***	−0.250 ***	0.144 ***	1
RET [13]	−0.216 ***	−0.180 ***	0.045	−0.083 ***	0.040	0.175 ***	0.040	−0.122 ***	0.330 ***	0.235 ***	−0.147 ***	−0.935 ***	1

*** p < 0.01; ** p < 0.05; * p < 0.1.

reports the correlation coefficients among the study variables. The two crash risk measures, NCSKEW and DUVOL, are highly correlated (0.895), confirming that both capture similar downside risk dynamics. Interestingly, CSR shows a positive and significant association with both NCSKEW (0.102) and DUVOL (0.082), suggesting that firms with higher CSR engagement tend to experience higher crash risk, contrary to the conventional view that CSR reduces information asymmetry and stabilizes stock prices. This may indicate that some firms pursue CSR symbolically, perhaps to enhance reputation or mask poor disclosure practices, rather than as a genuine governance mechanism. CSR is also strongly and positively correlated with profitability (ROA = 0.643), consistent with the slack resources theory that financially strong firms can afford greater social investments. However, CSR is negatively associated with firm size (−0.330) and FLC stages such as growth (−0.110) and decline (−0.165), implying that smaller, profitable firms in mid-life stages tend to be more socially active. Leverage shows a weak positive correlation with CSR (0.145), suggesting that indebted firms may use CSR to strengthen stakeholder confidence. Both crash risk proxies are positively correlated with volatility (SIGMA) and negatively with stock returns (RET), as expected. Importantly, aside from the very high correlation between RET and SIGMA (−0.935), all other coefficients remain below the critical 0.7 threshold, indicating that multicollinearity is not a concern. Overall, the correlation patterns reveal substantial variation in firms’ CSR intensity and risk behavior, hinting that CSR’s influence on crash risk may depend on firm-specific and contextual factors explored in later regressions.

4.3. Baseline Effect of CSR on Crash Risk

The regression results for the two measures of SPCR (i.e., NCSKEW and DUVOL) are presented in

Table 4. Regression analysis with SPCR as dependent variable (using NCSKEW as proxy).

	Model (1)	Model (2)	Model (3)	Model (4)	Model (5)	Model (6)
CSRt−1	−1.141 **	−1.539 ***	−1.098 **	−0.956 **	−1.049 **	−1.152 **
	(0.468)	(0.496)	(0.441)	(0.417)	(0.472)	(0.473)
Intro		0.189
		(0.258)
CSR × Intro		1.518 **
		(0.722)
Growth			0.176
			(0.122)
CSR × Growth			−0.463
			(0.516)
Maturity				−0.198 *
				(0.118)
CSR × Mature				0.045
				(0.463)
Shakeout					0.078
					(0.160)
CSR × Shakeout					−0.737
					(0.647)
Decline						−0.432 *
						(0.244)
CSR × Decline						−6.839 **
						(3.400)
FSt−1	0.088	0.084	0.071	0.129 **	0.097	0.089
	(0.061)	(0.061)	(0.064)	(0.060)	(0.061)	(0.061)
ROAt−1	0.039 ***	0.043 ***	0.039 ***	0.037 ***	0.040 ***	0.039 ***
	(0.009)	(0.009)	(0.009)	(0.009)	(0.009)	(0.010)
LEVt−1	0.054	0.036	0.045	0.025	0.048	0.055
	(0.109)	(0.108)	(0.108)	(0.107)	(0.108)	(0.110)
SIGMAt−1	10.333	9.058	10.382	9.618	10.373	10.547
	(10.442)	(10.111)	(10.393)	(10.318)	(10.390)	(10.469)
RETt−1	−1.515	−1.774	−1.541	−1.588	−1.506	−1.498
	(1.714)	(1.638)	(1.707)	(1.691)	(1.703)	(1.723)
Cons	−2.132 *	−1.995 *	−1.874	−2.772 **	−2.308 *	−2.152 *
	(1.158)	(1.163)	(1.211)	(1.161)	(1.159)	(1.156)
Observations	917	917	917	917	917	917
R-squared	0.171	0.178	0.173	0.178	0.173	0.172
Adj R2	0.165	0.170	0.166	0.171	0.166	0.165
F-stat	6.244	6.551	4.881	5.716	6.162	5.120
p	0.000	0.000	0.000	0.000	0.000	0.000
Year Dummy	Yes	Yes	Yes	Yes	Yes	Yes
Robustness	Yes	Yes	Yes	Yes	Yes	Yes

Note: Where *** p < 0.01, ** p < 0.05, and * p < 0.1, and the figures in parentheses correspond to robust standard errors. Model 1 represents the analysis of CSR; Model 2 represents the 1st stage of FLC (introduction) with an interactive term; Model 3 represents the 2nd stage of FLC (growth) with an interaction term; Model 4 represents the 3rd stage of FLC (maturity) with an interaction term; Model 5 represents the 4th stage of FLC (shake-out) with an interaction term; and, finally, Model 6 represents the 5th and very last stage of FLC (decline) with an interaction term. Appendix A contains the definitions of the variables.

and

Table 5. Regression analysis with SPCR as dependent variable (using DUVOL as proxy).

	Model (1)	Model (2)	Model (3)	Model (4)	Model (5)	Model (6)
CSRt−1	−0.816 ***	−1.066 ***	−0.783 ***	−0.727 ***	−0.763 ***	−0.820 ***
	(0.290)	(0.291)	(0.266)	(0.273)	(0.280)	(0.294)
Intro		−0.068
		(0.155)
CSR × Intro		1.122 ***
		(0.409)
Growth			0.151 *
			(0.079)
CSR × Growth			−0.337
			(0.333)
Maturity				−0.149 **
				(0.067)
CSR × Mature				0.138
				(0.262)
Shakeout					0.049
					(0.088)
CSR × Shakeout					−0.436
					(0.304)
Decline						−0.183
						(0.238)
CSR × Decline						−4.358
						(2.623)
FSt−1	0.042	0.040	0.026	0.068 *	0.047	0.042
	(0.034)	(0.034)	(0.037)	(0.035)	(0.034)	(0.034)
ROAt−1	0.021 ***	0.024 ***	0.021 ***	0.020 ***	0.022 ***	0.022 ***
	(0.005)	(0.005)	(0.004)	(0.004)	(0.004)	(0.005)
LEVt−1	0.049	0.040	0.041	0.031	0.046	0.050
	(0.067)	(0.066)	(0.067)	(0.066)	(0.067)	(0.068)
SIGMAt−1	2.562	2.113	2.609	2.166	2.590	2.621
	(4.555)	(4.515)	(4.494)	(4.496)	(4.538)	(4.580)
RETt−1	−0.894	−0.986	−0.919	−0.931	−0.888	−0.891
	(0.696)	(0.688)	(0.686)	(0.683)	(0.689)	(0.704)
Cons	−0.915	−0.835	−0.662	−1.311 **	−1.016	−0.929
	(0.638)	(0.630)	(0.677)	(0.648)	(0.640)	(0.637)
Observations	917	917	917	917	917	917
R-squared	0.107	0.114	0.112	0.116	0.110	0.109
Adj R2	0.102	0.106	0.104	0.108	0.102	0.101
F-stat	5.838	6.528	5.184	6.937	5.707	4.640
p	0.000	0.000	0.000	0.000	0.000	0.000
Year Dummy	Yes	Yes	Yes	Yes	Yes	Yes
Robustness	Yes	Yes	Yes	Yes	Yes	Yes

Note: Where *** p < 0.01, ** p < 0.05, and * p < 0.1 and the figures in parentheses correspond to robust standard errors. Model 1 represents the analysis of CSR; Model 2 represents the 1st stage of FLC (introduction) with an interaction term; Model 3 represents the 2nd stage of FLC (growth) with an interaction term; Model 4 represents the 3rd stage of FLC (maturity) with an interaction term; Model 5 represents the 4th stage of FLC (shake-out) with an interaction term; and, finally, Model 6 represents the 5th and very last stage of FLC (decline) with an interaction term. Appendix A contains the definitions of the variables.

, respectively. Across all six fixed-effect panel regression models, the coefficient on lagged corporate social responsibility (CSR_t−1) is negative and statistically significant, indicating a robust inverse relationship between CSR performance and future SPCR. Consistence with the prior studies (Hunjra et al., 2020; J.-B. Kim et al., 2014), the results in both tables depict that a firm’s CSR causes a negative and significant impact on SPCR. This suggests that the firms with higher CSR performance are likely to have a lower risk of crash in their stock prices. This means that when companies are much more operative and energetic in their CSR activities and contribute towards society, they help reduce SPCR. In other words, “doing good” appears to help firms avoid extreme negative stock price crashes. This finding aligns with the idea that strong CSR practices foster greater transparency, ethical behavior, and trust with stakeholders, which in turn mitigates the hoarding of bad news that often precipitates crashes.

This result supports Hypothesis 1, which proposed that CSR reduces the likelihood of stock price crashes. The findings are aligned with stakeholder theory, which posits that CSR enhances transparency, builds reputational capital, and fosters long-term trust with stakeholders, mechanisms that collectively reduce the need for managers to conceal adverse information. Consistent with the findings of Y. Kim et al. (2014), the current study asserts that CSR firms are less likely to conceal negative news and have a greater level of financial disclosure transparency, resulting in lower SPCR. The evidence from this study reaffirms that CSR functions not only as a normative commitment but also as a risk mitigation tool that contributes to financial stability in emerging markets.

4.4. Firm Life Cycle as Moderator

Models 02-06 were estimated to assess the effect of CSR on SPCR in the presence of FLC. FLC was categorized in five stages (i.e., introduction, growth, maturity, shake-out, and decline). Each model considered a single stage as the moderating variable. Models 02-06 in Table 4 and Table 5 depict the moderating effect of various stages of FLC on the connection linking CSR performance of the firm and its SPCR, for both NCSKEW and DUVOL measures, respectively. The most striking result is for the introduction stage: the coefficient on CSR × Intro is positive and highly significant in both of the tables. This implies that for firms in the nascent (introductory) stage of their life cycle, the usual risk-reducing effect of CSR is significantly weakened—indeed, the sum of the coefficients (baseline CSR + CSR × Intro) is approximately zero or slightly positive, suggesting little to no crash risk reduction from CSR in the introduction phase. In practical terms, a new or start-up firm’s engagement in CSR does not translate into the same protective effect against crashes that more mature firms enjoy. One explanation is that young firms use CSR primarily as a reputation-building or signaling tool to gain legitimacy with investors and stakeholders, rather than as a deeply ingrained aspect of transparent corporate culture. Early-stage companies lack established track records, so while they may invest in CSR to signal quality and build goodwill, stakeholders could view these efforts as less credible or as marketing ploys until the firm’s performance catches up. This finding resonates with the idea that reputational capital from CSR must be accumulated over time to effectively insure against downside risk.

For firms in the growth, mature, and shake-out stages, the results are all statistically insignificant. In these stages, the effect of CSR on crash risk does not differ significantly from the baseline, implying that once past the introduction phase, firms generally reap similar crash mitigation benefits from CSR. This makes intuitive sense: in growth and maturity phases, firms have more established operations and stakeholder relationships, so genuine CSR engagement likely enhances transparency and trust uniformly, reducing crash risk in line with the overall trend. The lack of a significant difference suggests that the risk-reducing impact of CSR is broadly consistent across the middle life-cycle stages, and that a mature firm’s CSR has about the same proportional effect on crash risk as that of a growing firm, all else equal.

The decline stage tells a more nuanced story. The interaction CSR × Decline is negative (–4.358) and economically large in magnitude, hinting that CSR might have an especially strong crash risk-reducing effect for firms in decline. However, this coefficient is statistically insignificant in the DUVOL regression. Interestingly, when NCSKEW is used in parallel analysis, CSR remains significant in the decline phase. Why would CSR’s effect “lose significance” in the decline stage for DUVOL but not for NCSKEW? One potential reason is the difference in what these metrics capture. NCSKEW focuses on the asymmetry of extreme negative returns, making it very sensitive to rare crash events. DUVOL, in contrast, measures the relative volatility of downside vs. upside returns over a period, which may be more diffuse and less sensitive to one-off extreme drops. If CSR provides a true “insurance-like” buffer that chiefly reduces the severity of extreme crashes (e.g., by prompting stakeholder support or more candid disclosure when a firm is in crisis), that effect might be more readily detected in the skewness of returns (NCSKEW) than in the volatility ratio (DUVOL). In a declining firm, strong CSR commitment could, for example, temper panic selling by investors or assuage creditors during bad news events, thereby truncating the worst downside outcomes, a pattern NCSKEW would capture. DUVOL, however, might not fully reflect this if the day-to-day return volatility remains high during both ups and downs. Another consideration is statistical power; firms in the decline stage are relatively few, often distressed, and volatile, which can lead to noisier estimates. The large negative coefficient on CSR × Decline in the DUVOL model suggests a substantial effect in theory, but the lack of significance implies high variance or insufficient samples to confirm it. By contrast, the same effect registers as significant in the NCSKEW model, possibly due to the metric’s focus on tail events. Overall, the divergence between DUVOL and NCSKEW in the decline stage underlines that results should be interpreted cautiously; nevertheless, across both measures, the direction of the CSR effect in decline is negative (risk-reducing), consistent with an insurance-like benefit of CSR even for embattled firms, albeit one that is harder to statistically pinpoint with the DUVOL proxy.

These results partially support Hypothesis 2, which posited that the effect of CSR on crash risk is moderated by FLC stage, and echo findings by Dickinson (2011) and Godfrey (2005), both of whom emphasize the role of context and timing in determining the efficacy of governance mechanisms like CSR. The overall results suggest CSR as a major determinant of crash risk, consistent with the prior literature (Y. Kim et al., 2014; Lee, 2016). However, most of the stages of FLC do not have a moderating effect on this relationship, except for firms in the introduction and decline stages.

4.5. Control Variables and Other Considerations

Among the control variables, return on assets (ROA) is positively and significantly related to crash risk in all model specifications, indicating that more profitable firms may face greater risk of sudden stock price collapses. This counterintuitive pattern aligns with the bad news-hoarding hypothesis, wherein firms with strong recent performance are more inclined to manage earnings or delay disclosure of unfavorable developments, thereby increasing the probability of abrupt corrections. Firm size (FS) is marginally significant in only one model, while leverage (LEV), stock return volatility (SIGMA), and average stock returns (RET) are statistically insignificant across the board. These results suggest that traditional firm characteristics such as scale or financial structure are less predictive of crash risk once CSR engagement and firm fixed effects are considered. The explanatory power of the models, as indicated by adjusted R² values (approximately 0.17 for NCSKEW and 0.11 for DUVOL), is modest but reasonable given the rare and idiosyncratic nature of crash events. Importantly, the robustness of CSR’s effect despite the inclusion of these controls strengthens the validity of the main findings and supports the view that CSR operates as a unique informational and reputational safeguard against extreme downside market events. Further, the inclusion of firm fixed effects means that the model effectively controls for all time-invariant differences across firms (such as industry, baseline risk appetite, or corporate culture). This makes the results more robust by ensuring that the negative CSR–crash link is not driven by some omitted firm-level trait correlated with CSR. The use of robust (heteroskedasticity-consistent) standard errors adds confidence that the statistical significance is not due to outlier influence or heteroskedastic variance. It is important to note that the results were qualitatively similar when using an alternative crash risk measure (NCSKEW), reinforcing that the observed relationships, particularly the baseline negative effect of CSR, are not an artifact of one specific metric.

This study investigated the impact of CSR on the SPCR. The overall results suggest CSR as a major determinant of crash risk, consistent with the prior literature (Y. Kim et al., 2014; Lee, 2016). However, most of the stages of FLC do not have a moderating effect on this relationship, except for firms in the introduction and decline stage.

5. Conclusions

Firms in emerging economies are becoming increasingly conscious of their reputation, leading to greater engagement in CSR activities. Many businesses pursue responsible business conduct (RBC) to build trust-based relationships with stakeholders, including customers, suppliers, employees, and the broader community (Borghesi et al., 2014; Ferrell et al., 2016). As SPCR becomes an increasingly critical issue in financial decision-making and risk management, recent episodes such as the Global Financial Crisis (GFC) have renewed attention among regulators, investors, and researchers regarding the causes and consequences of SPCR. Prior research has linked SPCR to managerial tendencies to hoard bad news (Hutton et al., 2009), driven by incentives to preserve career prospects and secure short-term benefits (Kothari et al., 2009). Against this backdrop, CSR is expected to play a pivotal role in discouraging such opportunistic behaviors by fostering transparency and moral legitimacy. This study, therefore, hypothesized an inverse relationship between CSR and SPCR, with a focus on Pakistan, a representative emerging market. Furthermore, we examined whether this relationship is moderated by the FLC, which reflects a firm’s strategic evolution in response to internal and external forces (Dickinson, 2011).

Our empirical findings support the view that CSR serves as a significant determinant of crash risk. Firms with stronger CSR performance are less likely to experience extreme stock price drops, consistent with Hypothesis 1 and the prior literature (e.g., Y. Kim et al., 2014; Lee, 2016). While the overall relationship between CSR and SPCR is negative and significant, the moderating role of FLC (Hypothesis 2) reveals more nuanced effects. Specifically, the CSR–SPCR link weakens or even reverses in the introduction stage, potentially due to the symbolic use of CSR by early-stage firms. In contrast, CSR appears to offer stronger crash risk protection in the maturity and decline phases, particularly under the NCSKEW proxy. Descriptive and diagnostic statistics confirmed the validity of our panel dataset, supporting the robustness of these results.

This research contributes in several important ways. First, it is one of the few studies to empirically examine the moderating effect of FLC on the CSR–SPCR relationship. Second, it adds to the literature by exploring these dynamics in the context of an emerging market. Third, the study offers practical value to corporate managers by highlighting the importance of CSR as a strategic tool in risk mitigation and resource allocation. Fourth, investors and portfolio managers may find this research useful when evaluating firm-level risk, particularly in identifying CSR performance as an indicator of crash vulnerability. Finally, the findings offer useful insights for policymakers aiming to reduce economic uncertainty and foster a regulatory environment that promotes CSR engagement.

As Pakistan continues to navigate uncertain macroeconomic conditions, the results offer valuable guidance for corporate leaders and capital market participants. Strong CSR practices, embedded within a sound governance framework, can reduce crash risk and enhance market confidence. The evidence also suggests that managerial ownership and robust governance structures may reinforce effective risk management and limit opportunistic behaviors. In this regard, CSR and corporate governance complement one another in safeguarding long-term shareholder value.

From a policy perspective, encouraging CSR, especially in areas related to environmental sustainability and stakeholder engagement, may reduce firm-level financial risk and enhance overall market stability. Our findings also suggest that charitable contributions, while beneficial for public relations, should be strategically aligned with transparency and accountability to avoid reputational risk. Furthermore, long-term institutional investors, who are generally drawn to firms with high CSR ratings, may play a stabilizing role by supporting firms with strong social and governance commitments. This offers an avenue for regulators and firms seeking to build resilient investor bases.

While this study addressed the moderating effect of FLC and validated key assumptions, there is scope for deeper analysis. Future research could examine how various dimensions of CSR, such as environmental initiatives, employee welfare, and governance disclosures, differentially impact crash risk. Similarly, studies might explore whether firms that excel in specific CSR domains (e.g., environmental stewardship or community engagement) experience distinct risk profiles in terms of stock liquidity or volatility. Other promising directions include assessing the role of board composition, ownership concentration, or institutional investor presence in shaping the CSR–SPCR relationship. These questions merit further inquiry as researchers and practitioners seek to better understand how CSR contributes to sustainable financial performance and market stability.