Do employee owners deter firms from withholding bad news? Evidence from Korea Open Access

Stock price crash risk has received growing scholarly and regulatory attention because extreme downside events threaten not only individual investors but also broader market stability. Prior research, grounded in the agency framework of Jin and Myers (2006), emphasizes that crashes often arise from managers’ deliberate hoarding of bad news until it accumulates and is released in a concentrated form (Hutton et al., 2009). In this framework, opacity and information asymmetry amplify the likelihood of sudden price collapses, whereas mechanisms that enhance transparency and accountability reduce crash risk. Consequently, a wide range of governance mechanisms—both internal and external—have been studied as determinants of crash risk, including managerial compensation, board oversight, internal controls, institutional monitoring, product-market competition, and stock liquidity. Yet, despite the growing body of work on these factors, the role of employee share ownership (ESO) remains largely unexplored.

The existing literature portrays ESO in ambivalent terms. On the one hand, ESO may entrench management by aligning employees with executives, as employees tend to prioritize wages and benefits over equity value and may side with managers in corporate control contests (Faleye et al., 2006; Park, 2017). Such dynamics can strengthen managerial discretion and exacerbate opacity. On the other hand, a substantial body of work emphasizes the positive impact of ESO on enhancing productivity, reducing agency costs, and improving disclosure (Jones and Kato, 1995; Bova et al., 2015; Blasi et al., 2016). These contrasting views underscore the need to examine ESO not only in terms of performance outcomes but also in relation to extreme risk events such as stock price crashes.

We argue that ESO can mitigate crash risk through three complementary channels that directly address the problem of bad-news hoarding. First, ESO expands the pool of informed insiders and strengthens internal monitoring and whistleblowing incentives. Employee–owners, who are closely involved in daily operations and directly exposed to downside losses, have stronger incentives to detect and escalate concerns about concealed problems, raising the expected cost of hoarding to managers. Second, ESO reduces information asymmetry and alleviates rent extraction pressures. By aligning employees’ interests with long-term firm value rather than short-term wage gains, ESO lessens managers’ need to maintain opacity during labor negotiations, thereby reducing one of the root causes of crash risk. Third, voluntary ESO schemes, which often require employees to invest their own funds, act as credible signals of insider confidence and create reputational pressure on managers to maintain transparency. These mechanisms, supported by prior evidence linking ESO to disclosure quality and transparency (Bova et al., 2015; Li et al., 2019), suggest that ESO can function as an internal governance device that constrains managerial incentives to conceal adverse information and, ultimately, reduces the probability of stock price crashes.

The Korean setting provides a distinctive institutional context in which to test these theoretical predictions. ESO in Korea is organized through employee associations rather than external trustees, giving employees collective voting rights and meaningful influence in corporate governance. Moreover, Korean schemes typically impose short lock-up periods of approximately one year, during which employees are prohibited from selling their shares, thereby reinforcing incentives for near-term monitoring and disclosure. Importantly, the regulatory framework differentiates between mandatory ESO in KOSPI-listed firms, where 20% of new equity issues must be allocated to employees, and voluntary ESO in KOSDAQ-listed firms, where employees must proactively request new share allotments. This heterogeneity creates quasi-experimental variation in employee incentives and managerial responses, offering a unique opportunity to assess when ESO is most effective in mitigating crash risk. Using a dataset of 16,929 firm-years from 2011 to 2020, we conduct firm-year fixed-effect regression analyses. Our findings reveal a negative association between ESO and future stock price crash risk, specifically for firms listed on the KOSDAQ market, which are subject to a voluntary preferential allocation of new shares. Our findings support the argument that ESO mitigates employees' ability to extract above-market rents (Bova, 2013; Jones and Kato, 1995). More broadly, our results suggest that ESO plays a governance role when employees voluntarily purchase shares in their firm.

Additional tests conducted to enhance the analytical depth of our study show that ESO reduces stock price crash risk only in volatile industries and those with high information opacity. This indicates that ESO effectively reduces stock price crash risk when information asymmetry is severe. Interestingly, the effectiveness of ESO diminishes when the largest shareholders exert significant influence over management, suggesting that the impact of ESO depends on the dynamics of the firm's ownership structure.

This study makes a significant contribution to the literature on employee share ownership (ESO), disclosure, and stock price crash risk in several important ways. First, we extend prior research that links ESO to disclosure quality (e.g. Bova et al., 2015) by shifting the focus from average levels of transparency to the distributional consequences of information hoarding. While it is well established that greater transparency reduces information asymmetry, relatively little is known about how ESO affects the nonlinear downside tail of stock returns. Building on the theoretical framework of Jin and Myers (2006), who highlight bad-news hoarding as the fundamental driver of crash risk, we theorize and test how ESO constrains managerial incentives to conceal adverse information. By doing so, we move beyond the general “ESO-disclosure” narrative and demonstrate ESO’s role in mitigating the extreme realization of hidden risks.

Second, our study leverages the unique institutional design of ESO in Korea, which allows us to develop sharper theoretical predictions. Korean ESO schemes are characterized by short lock-up periods (approximately one year) and, in the case of KOSDAQ-listed firms, voluntary participation by employees. These features strengthen employees’ monitoring and whistleblowing incentives in the short term, since they cannot exit during the lock-up but will directly bear the losses from any sudden disclosure of hoarded bad news. At the same time, the contrast between mandatory ESO in KOSPI firms and voluntary ESO in KOSDAQ firms provides quasi-experimental variation in the strength of employee incentives and managerial responses. This institutional heterogeneity enables us to theorize and empirically identify contexts in which ESO is more effective in constraining bad-news hoarding, thereby offering a theoretical contribution that goes beyond single-country description.

Third, by documenting that ESO reduces crash risk, particularly in settings with high information opacity, volatile industries, and weaker controlling shareholder dominance, our findings highlight ESO as an alternative governance mechanism that can substitute for weak external monitoring. This insight resonates with and extends the literature on corporate governance and crash risk (Hutton et al., 2009; Kim et al., 2011), underscoring that the benefits of ESO are not uniform but contingent on the surrounding governance environment.

Taken together, these contributions reposition our study’s novelty from a narrow disclosure-quality story to a theory-driven analysis of how ESO affects bad-news hoarding and the extreme downside distribution of stock returns, while exploiting the Korean institutional context to derive and test predictions that are of broader relevance to the governance and crash-risk literature.

The remainder of this paper is organized as follows: Section 2 reviews the relevant literature, explains the unique characteristics of ESO in Korea, and develops our hypotheses. Section 3 describes the study’s data and research design. Section 4 presents the main results. Section 5 provides robustness checks and additional analysis. Finally, Section 6 concludes the article.

Stock price crash risk has been widely attributed to managers’ incentives to withhold firm-specific bad news until it accumulates and is released abruptly, generating extreme negative stock returns (Jin and Myers, 2006; Hutton et al., 2009). In this framework, opaque information environments facilitate bad-news hoarding, whereas governance mechanisms that enhance transparency and constrain managerial discretion mitigate crash risk.

A large body of research documents that both internal and external governance mechanisms shape managers’ disclosure incentives and, in turn, crash risk. Prior studies show that internal governance arrangements—such as executive equity incentives, board oversight, audit committee effectiveness, and internal control quality—affect the extent of information opacity and are systematically related to crash risk (e.g. Kim et al., 2011; Andreou et al., 2016). External monitoring mechanisms, including institutional investor oversight, product market competition, and stock liquidity, similarly discipline managers and reduce the likelihood of extreme price collapses (Callen and Fang, 2013; Chang et al., 2017). Taken together, this literature establishes that crash risk fundamentally reflects the effectiveness of governance mechanisms in limiting managers’ ability to conceal adverse information.

Employee share ownership (ESO) represents a distinct internal governance mechanism that has received relatively little attention in the crash risk literature. Unlike traditional governance devices that operate through boards, investors, or markets, ESO directly reshapes internal information flows by extending ownership to employees who are closely involved in day-to-day operations. The literature offers competing views on the governance role of ESO. One strand emphasizes its potential to entrench management by aligning employees with executives, thereby weakening external discipline and exacerbating agency problems (Faleye et al., 2006; Park, 2017). In contrast, a growing body of evidence highlights ESO’s role in improving information transparency and reducing information asymmetry. Studies show that employee ownership is associated with richer voluntary disclosure, improved stock liquidity, and stronger internal monitoring, suggesting that employees with equity stakes have greater incentives to demand timely and accurate information (Bova et al., 2015; Jung and Choi, 2021).

Synthesizing these findings, prior research suggests that ESO can influence crash risk through several related channels. First, by expanding the set of informed insiders, ESO strengthens internal monitoring and increases the expected cost to managers of withholding bad news. Second, ESO can mitigate information asymmetry arising from stakeholder conflicts, as employees’ interests shift toward long-term firm value rather than short-term wage extraction. Third, when employee participation is voluntary, ESO may serve as a credible signal of insider confidence, creating reputational pressure for managers to maintain transparent disclosure. These channels imply that ESO can function as an internal governance mechanism that directly constrains managerial incentives to hoard bad news, thereby reducing the likelihood of stock price crashes.

While the objectives of ESO introduced in Korea in 1958 are similar to those in other countries, the organization and operation of the scheme differ significantly. In Korea, employees usually purchase shares themselves with minimal company contribution (Cin and Smith, 2002). This contrasts with practices in other countries where companies may provide shares at no cost, such as performance-based compensation in the UK or matching shares through retirement plans. In Korea, employees can sell their shares after a one-year lock-up period, whereas in other countries, disposal rights are often tied to retirement ^[1].

The ESO association in Korea is self-organized, legally mandated entity within the firm differs from the external trustee administration seen in the US and UK The internalized approach in Korea gives employees substantial voting rights and incentives to engage in management and influence decision-making processes. With insider knowledge of the company's performance, governance, and future cash flows, ESO associations are better positioned to monitor and enhance firm value, justifying the rationale for ESO.

Additionally, Korea has a preferential allocation rule for ESO when firms issue new shares through a public offering. Article 38 of the Basic Labor Welfare Act stipulates that KOSPI-listed firms must reserve 20% of new shares for ESO association members ^[2]. In contrast, KOSDAQ-listed firms can voluntarily allocate up to 20%. This distinction enables different market types to interpret the same ESO ratio differently in relation to future firm performance (Jang and Yoon, 2023).

Drawing on the crash risk and governance literature, we argue that employee share ownership is expected to reduce the risk of future stock price crashes. By extending ownership to employees who possess firm-specific operational knowledge, ESO reshapes managers’ disclosure incentives in ways that directly affect bad-news hoarding behavior.

First, ESO strengthens internal monitoring by expanding the set of informed insiders. Employees are uniquely positioned to observe early signals of operational and financial problems, and their equity exposure increases the expected personal cost of remaining silent when adverse information is concealed. Unlike external monitors, employee-owners face direct downside risk if accumulated bad news is released abruptly, which raises the likelihood that concerns are escalated internally and disclosed in a timely manner. As a result, ESO increases the expected cost to managers of withholding unfavorable information.

Second, ESO can reduce information asymmetry arising from stakeholder conflicts between managers and employees. In the absence of employee ownership, managers may prefer to maintain opaque information environments to limit rent extraction pressures during labor negotiations. When employees hold equity stakes, however, their interests shift toward long-term firm value rather than short-term wage gains. This alignment weakens managers’ incentives to strategically withhold information and reduces the accumulation of undisclosed bad news. Consistent with this channel, prior studies document that firms with employee ownership exhibit more transparent disclosure environments (Bova et al., 2015), and greater transparency has been shown to mitigate stock price crash risk (Hutton et al., 2009).

Third, the governance effect of ESO is likely to be stronger when employee participation is voluntary. Voluntary ESO requires employees to invest their own funds, which serves as a credible signal of insider confidence and intensifies reputational pressure on managers to honor that confidence through transparent disclosure. In contrast, mandatory ESO participation may reflect regulatory compliance rather than genuine employee engagement, resulting in weaker monitoring incentives. This distinction is particularly salient in the Korean setting, where voluntary ESO participation is concentrated among KOSDAQ-listed firms. Taken together, these arguments lead to the following hypotheses:

The initial data includes firms listed on the KOSPI and KOSDAQ markets during the 2011–2020 period ^[3], ^[4]. To measure the variables, we exclude firms in the financial industry and those with insufficient data on stock prices, trading volumes, and accounting. We obtain financial and employee ownership data from the Fn-guide. After this selection process, we obtain a final sample of 16,929 firm-year observations. In our final sample, we obtain 6,536 observations listed on KOSPI and 10,393 observations listed on KOSDAQ. Appendix A1 summarizes the composition of the sample.

Table 1 reports the sample distributions of ESO by year. Columns (1) and (2) report the total number of firms in the sample and the number of firms with employee ownership, respectively. Column (3) provides the percentage of firms with employee ownership in the sample. Column (4) shows the average ESO of firms in the sample. The proportion of firms with ESO remains stable over the years, from 35.1% to 36.6%, and the average ESO is around 1%, with the highest of 1.214 in 2011 and the lowest of 0.975 in 2016.

Following previous studies (Callen and Fang, 2015; Hutton et al., 2009; Kim et al., 2019), we employ three measures of firm-specific crash risk based on Jin and Myers’ (2006) market model for each firm-year observation. We estimate firm-specific weekly returns using the following market and industry index model regression for each firm and year (Hutton et al., 2009):

where $ri,t$ is the return on stock i in week t, $rm,t$ is the return on the value-weighted market index in week t, and $rk,t$ is the return on the value-weighted industry index based on the two-digit Korean Standard Industrial Classification Code in week t. By including the lead and lag terms for the value-weighted market index returns, and those for the industry index returns in Equation (1), we correct for nonsynchronous trading (Dimson, 1979). We denote $Wi,τ=ln⁡(1+εi,τ)$⁠, the firm-specific weekly return for firm j in week t, measured by the natural logarithm of one plus the residual return from Equation (1).

The first measure of stock price crash risk NCSKEW is the negative conditional skewness of firm-specific weekly returns over the fiscal year. NCSKEW is computed by taking the negative of the third moment of firm-specific weekly returns for each year and normalizing it by the standard deviation of firm-specific returns raised to the third power. NCSKEW is calculated as

where n is the number of firm-specific weekly returns for firm j during year A higher value of NCSKEW indicates higher crash risk.

The second measure, DUVOL, is the down-to-up volatility of negative versus positive returns. To calculate DUVOL, for each firm i over the fiscal year, firm-specific weekly returns are separated into two groups: “down” weeks when the returns are below the annual mean, and “up” weeks when the returns are above the annual mean. Then, the standard deviation of each group is calculated separately. As such, DUVOL is the natural logarithm of the ratio of the standard deviation on the “down” weeks to the standard deviation on the “up” weeks. DUVOL is calculated as

where, $nu$ and $nd$ are the number of up and down weeks in year t, respectively. A higher value of DUVOL indicates greater crash risk.

The third measure, VCRASH, suggested by Bradshaw et al. (2010) and Zhu (2016) is the absolute value of the difference between minimum firm-specific weekly return and its mean value over the fiscal year, divided by its standard deviation for the year. VCRASH measures the probability of extremely negative returns and is calculated as

To investigate whether ESO is associated with future stock price crash risk, we estimate the following regression model:

where the dependent variable $CARSHt+1$ is proxied by $NCSKEWt+1$⁠, $DUVOLt+1$⁠, or $VCRASHt+1$ as defined above. To mitigate the concern of reverse causality, we use the dependent variable in year t+1, which is a one-year lead from the independent and control variables. Accordingly, we test whether a firm’s ESO in year t can predict stock price crash risk in year t+1. Our variable of interest, $ESOt$⁠, is the percentage of ESO calculated by dividing the amount by the shares outstanding in year t.

The set of control variables include $CRASHt$⁠, $dTURNt$⁠, $SIZEt$⁠, $MBt$⁠, $LEVt$⁠, $ROEt$⁠, $SIGMAt$⁠, $RETt$⁠, $ADACCt$⁠, and $HHIt$ based on previous studies (Chen et al., 2001; Cho et al., 2023; Hutton et al., 2009). All linear variables are winsorized at the top and bottom 1% to mitigate the impact of outliers. We use firm-year fixed effects to address potential endogeneity concerns and heteroskedasticity-robust standard errors to account for firms’ potential dependence over time. Detailed variable definitions are provided in Appendix A2.

Panel A of Table 2 reports the descriptive statistics of all variables in the research model for the full sample. The mean values of NCSKEW_t+1, DUVOL_t+1, and VCRASH_t+1 are −0.411, −0.250, and 2.350, respectively, consistent with prior studies (Zhu, 2016; Cho et al., 2023). The mean value of ESO is 0.357%. Panel B shows the results of the mean difference tests (t-test) conducted between the groups; firms with and without ESO and firms in KOSPI- and KOSDAQ-listed firms. Columns (1) and (2) present the mean values of the variables for firms with and without ESO. Column (3) indicates that the differences between the mean values for most of the variables are significant, except for $MBt$ and $HHIt$⁠. In particular, the mean values of $NCSKEWt+1$ and $DUVOLt+1$ of firms without ESO are lower than those for firms with ESO. The mean values of $ROEt$ and $RETt$ are less negative for firms with ESO, and the mean value of $ADACCt$ is lower for firms with ESO, implying that earning quality is higher for firms with ESO. Columns (4) and (5) show the mean values of the variables for KOSPI- and KOSDAQ-listed firms. These mean values differ significantly, except for $HHIt$⁠. Specifically, the mean values of $NCSKEWt+1$⁠, $DUVOLt+1$ and $VCARSHt+1$ for KOSDAQ-listed firms are statistically lower than those for KOSPI-listed firms. The mean value of $ESOt$ for firms listed on KOSPI is 0.500%, whereas that for firms listed on KOSDAQ is 0.267%. Combined with the results in Panel A of Table 1, we observe that KOSPI-listed firms have not only a higher proportion of firms adopting ESO but also a higher ratio of ESO compared to KOSDAQ.

Hypothesis 1 predicts that ESO is negatively associated with future stock price crash risk. Table 3 presents the coefficient estimates for Equation (4) in firm-year fixed effect specification to test H1. Across all columns, the coefficients of $ESOt$ are negative and significant at the 5% level, which is consistent with H1. On average, firms adopting ESO are associated with a 0.027 decrease in $NCSKEWt+1$⁠, 0.013 decrease in $DUVOLt+1$⁠, and 0.021 decrease in $VCRASHt+1$⁠. Given that the mean values of $NCSKEWt+1$⁠, $DUVOLt+1$⁠, and $VCARSHt+1$ are −0.411, −0.250, and 2.350, respectively, the economic significance of the mitigating the effect of ESO on future crash risk is nontrivial.

The findings are consistent with the view of the proponents of ESO: ESO benefits the firm by playing an effective governance role in mitigating managers’ information withholding. In particular, given the one-year lock-up period for employee owners in Korea, managers with greater ESO would have greater incentive to keep information transparent and communicate clearly and in a timely manner (Bova et al., 2015; Bova et al., 2016).

Results of control variables are generally consistent with our predictions and findings of prior studies. In particular, the current stock price crash risk, $CRASHt$⁠, is negatively related to the future stock price crash risk, consistent with prior literature (Hutton et al., 2009; Jin and Myers, 2006). The coefficients on $SIZEt$ and $MBt$ are positive and highly significant, which is consistent with previous studies (Chen et al., 2001; Kim et al., 2011; Kim et al., 2014). This finding suggests that large and growing firms are more likely to experience crash events. Consistent with Hutton et al. (2009) and in line with previous studies, $ROEt$ are negatively related with future stock crash risk.

To control for the potential omitted variable bias, we employ a two-stage least squares regression (2SLS) approach. Based on prior studies (Pendleton, 2005; Paterson and Welbourn, 2020; Pierce et al., 2001; Bova et al., 2015), we adopt employee tenure and compensation as instrumental variables in our 2SLS analyses. Table 4 presents the 2SLS regression results ^[5]. In the first stage result presented in column (1), both $TENUREt$ and $PAYROLLt$ show significant and expected associations with $ESOt$⁠. That is, longer tenure is likely to enhance participation in the stock ownership program (Pierce et al., 2001). Using the fitted value of ESO (⁠$ESOtˆ$⁠) from the first-stage regression, we perform a second-stage analysis on the relationship between ESO and future stock price crashes and the results are displayed in Columns (2) to (4). The coefficients for $ESOtˆ$ is negative for all three measures of crash risk measures and statistically significant at the 1% level. Theses 2SLS results align with our primary findings, supporting Hypothesis 1.

Next, to control for potential sample selection bias, we conduct propensity score matching (PSM) analyses. Table 5 shows the PSM results. In Panel A, we find no significant differences between the two groups, indicating the successful mitigation of sample selection bias. Using this matched sample, we re-estimate the regression model and report the outcomes in Panel B of Table 4. $ESOt$ exhibits a significantly negative coefficient for $NCSKEWt+1$⁠, $DUVOLt+1$⁠, and $VCRASHt+1$⁠, consistent with the finding in Table 2. However, it is worth noting that the results regarding the control variables differ somewhat from the main analyses. In particular, $CARSHt$ and $SIZEt$ remain statistically significant, whereas the other variables lose their statistical significance. Taken together, the results obtained through the PSM approach confirm that our findings on the effect of employee stock on alleviating stock price crash risk are not driven by sample selection bias.

Our second hypothesis is the pronounced association with ESO and stock price crash risk. To test Hypothesis 2, we partition the sample into KOSPI and KOSDAQ-listed firms and conduct separate regression analyses for each group. Columns (1)–(3) of the Table 6 show the regression results for KOSPI-listed firms, where the coefficient of $ESOt$ is insignificant for all three measures of $CRASHt+1$⁠. In contrast, columns (4) and (6) present the results for KOSDAQ-listed firms, where the coefficient of $ESOt$ is statistically significant and negative across all crash risk measures. These findings suggest that the results in Table 3 are primarily driven by KOSDAQ-listed firms and that proactive ESO can enhance the timeliness of information disclosure, thereby deterring crashes. In summary, voluntary ESO is more effective in improving information timeliness, supporting Hypothesis 2.

Kim and Patel (2020) suggest that the industry environment partially impacts ESO outcomes, indicating industry characteristics, such as industry sales volatility are likely to be associated with market returns or idiosyncratic risk (Carline et al., 2023; Peters and Wagner, 2014). Therefore, we analyze whether the impact of ESO varies with industry volatility, measured by the standard deviation of industry-level sales growth over the past three fiscal years (⁠$VOLt$⁠). Table 7 shows for KOSPI-listed firms, $ESOt$ is insignificant, but the interaction term $ESOt×VOLt$, has a significantly negative coefficient. This suggests that ESO reduces crash risk of KOSPI-listed firms in volatile industries. For KOSDAQ-listed sample, $ESOt$ has a negative coefficient, while the interaction term is insignificant. These results suggest that industry volatility does not impact the association between ESO and crash risk in KOSDAQ-listed firms, but it does in KOSPI-listed ones. We interpret these findings to mean that for KOSPI-listed firms, where new share allotments to employees are mandatory, ESO effectively reduces stock price crash risks in a specific context (i.e. industry volatility). In contrast, in KOSDAQ-listed firms, ESOs reduce crash risk in general circumstances, regardless of industry volatility.

Hutton et al. (2009) claim that the risk of a stock price crash increases with higher accounting information opacity. Given that ESO diminishes managers’ incentives to conceal news, we expect the influence of ESO on stock price crash risk to be more pronounced in firms with high information opacity. To investigate whether the effect of ESO varies with accounting information opacity, we partition the samples into two groups: firms with low opacity (LOW) and firms with high opacity (HIGH), based on the median value of accounting information opacity. Table 8 presents the results for KOSDAQ-listed firms ^[6]. The coefficient of $ESOt$ achieves statistical significance for NCSKEW_t+1 and DUVOL _t+1 in high opacity firms. This suggests that ESO helps mitigate stock price crash risk, particularly in firms where accounting information opacity is high.

We examine the role of ESO in mitigating the agency problem between the largest and minor shareholders, a key source of information asymmetry (Amstrong et al., 2010; Boubaker et al., 2014; Khlif et al., 2017). When the largest shareholders possess a level of ownership that grants them substantial control over management decisions, the impact of ESO may be less pronounced. Conversely, if the interests of the largest shareholders align with employee shareholders, ESO’s impact is accentuated, particularly in firms with significantly larger shareholders’ ownership. Lee (2018) demonstrates that stock price crash risk decreases with increasing largest shareholder ownership. Lee et al. (2021) suggest that incentives to disclose firm-specific information vary with the largest shareholder’s ownership, identifying a critical point at 39%.^[7] To explore this relationship, we segment the samples into firms with high (HIGH) and low percentages of largest shareholder ownership (LOW). A firm is classified as HIGH when the largest shareholder’s ownership exceeds 40%, as LOW when it is below 40%, considering ownership below 40% as typical ^[8]. Table 9 presents the regression results for KOSDAQ firms ^[9]. The coefficient of $ESOt$ exhibits a significantly negative for NCSKEW _t+1 and DUVOL_t+1. These results suggest that ESO effectively reduces crash risk when the largest shareholders do not exert significant control over management decisions. This aligns with prior findings (Lee, 2018; Lee et al., 2021) that high largest shareholder ownership increases information transparency, while low ownership decreases it. Thus, ESO can reduce information asymmetry in less transparent environments.

We investigate the impact of employee share ownership (ESO) on future stock price crash risk in Korean firms and find robust evidence of a significant negative association. The mitigating effect of ESO is particularly pronounced in settings where employee participation is voluntary, consistent with the view that ESO constrains employees’ incentives to extract above-market rents. Additional analyses show that the effect of ESO is stronger in firms operating in volatile industries and in firms characterized by high accounting information opacity, suggesting that ESO is especially effective in reducing information asymmetry in weaker information environments. We also find that the governance role of ESO diminishes when controlling shareholders exert substantial influence over managerial decision-making, highlighting that the effectiveness of ESO is contingent on the firm’s ownership structure.

This study contributes to the literature by identifying ESO as an internal governance mechanism that mitigates stock price crash risk, thereby extending prior work that has primarily focused on disclosure quality and average transparency outcomes. By exploiting the distinctive institutional features of the Korean ESO system, our analysis offers insights that go beyond the Anglo-American context and deepens our understanding of how employee ownership shapes information environments and extreme downside risk in capital markets. These findings provide relevant implications for both regulators and practitioners concerned with the design of governance mechanisms that enhance transparency and limit managerial opportunism.

At the same time, our study is subject to several limitations. Our analysis focuses on a single institutional setting and a specific form of employee ownership, which may limit the generalizability of the results to other countries or ownership arrangements. Future research could extend our framework by examining ESO designs across different institutional environments, exploring alternative ownership models, or analyzing longer-term outcomes beyond stock price crash risk. Such extensions would further clarify the conditions under which employee ownership serves as an effective governance mechanism and help refine its role in corporate governance design.

The supplementary material for this article can be found online.