Does the Shield Effect of CSR Work in Crises? Evidence in Korea

: This paper investigates the impact of corporate social responsibility (CSR) on shareholders’ wealth during market downturn, focusing on the market crash caused by the COVID-19 pandemic and its aftermaths. We evaluate the relationship between ﬁrms’ CSR and stock returns using a sample of 803 ﬁrms listed on the Korean stock market. The results of our study reveal that ﬁrms’ pre-crisis CSR activities do not protect shareholders’ wealth during the crisis; in fact, they negatively affected stock returns during the COVID-19 crisis. This ﬁnding is consistent across several robustness tests and challenges the prevailing notion that CSR is solely a philanthropic endeavor. This study suggests that ﬁrms need to reconsider their CSR approach in order to better align it with shareholders’ interest.


Introduction
Corporate social responsibility (CSR) has always been a matter of contention among researchers, corporate leaders and other stakeholders. In a competitive economy, CSR is considered a mechanism to achieve firms' goal and sustain competitive advantages [1,2]. However, whether CSR engagement can add value to firms' stock performance, as it pledges, is still a topic of ongoing research and debate in the academic and business communities. In this paper, we discuss whether CSR can protect shareholders' wealth during economic downturn, specifically by investigating the relationship between CSR and firms' stock performance during the COVID-19 crisis.
Broadly, CSR refers to the ethical and responsible actions of a firm to address the social, environmental and economic impacts of its business operations and to improve the well-beings of the society it serves. The primary goal of CSR is to create a positive impact on society and the environment while benefiting the firm through increased reputation and trust [3,4]. On the other hand, how well a firm's stocks are performing in terms of generating a profit or loss for investors is often measured by stock return over a certain period of time. Moreover, stock return gives a comprehensive view of the financial health, perspective and performance of a firm, as it promptly responds to any unexpected change of event or information [5].
Existing theories have conflicting notions when explaining the impact of CSR on firms' performance. Stakeholder theory suggests that firms' CSR engagements enhance their reputation and, in turn, increase firms' value and returns [6,7]. On the other hand, according to agency theory [8], managers overinvest and prefer costly CSR practices to enhance their own reputation at the cost of shareholders' welfare [9,10]. The latter practice is more detrimental during any unexpected economic shock, and the recovery can be greatly hindered. Thus, the notion of "firms can do well by doing good" is still a matter of conflict and leads us to study the relationship between CSR and firms' performance.
Furthermore, the relationship between the firm and its stakeholders, and the impact of these relationships on the firm's performance, are rooted in the Coase theorem and the Alchian-Demsetz property rights theory, respectively. These theories highlight the importance of considering the relationships between the firm and its stakeholders in evaluating the firm's performance and behavior. Furthermore, these studies view the firm as a node of the connections among stakeholders, and the firm's stock performance depends on the strength of those connections [8,11,12]. By considering the impact of these relationships on the allocation of resources, property rights and the alignment of interests, researchers and practitioners can gain a deeper understanding of the factors that influence a firm's performance.
Indeed, the relationship between CSR and firms' performance is context-dependent, and the prior evidence is mixed. Some studies have found a positive relationship between CSR and firms' performance [13][14][15], while others have found a negative [16,17] or insignificant relationship [18,19].
Nevertheless, the COVID-19 pandemic is truly an exogenous shock, and its magnitude is erratic compared to previous financial crises [20]. The pandemic has not only led to a significant loss of life and lifelong trauma but has also instigated extensive alterations in social, political and economic spheres [21]. Long-term lockdowns and social distancing have resulted in disrupted education, work patterns and increased mental health issues (i.e., hoarding behavior, social withdrawal). Moreover, it has sparked debates and divisions over public health measures and government responses [22] and has triggered significant changes in global alliances and geopolitical dynamics. Throughout this period, the world is grappling with the impact of a recession [23], a higher unemployment rate, shifts in consumption, disruptions in supply chain and industry-specific challenges (i.e., tourism, retail and manufacturing) [24,25]. The long-term consequences of the COVID-19 pandemic are still unfolding and will continue to shape the social, political and economic environment in the coming years [26].
Furthermore, financial markets all over the world have been substantially affected by the COVID-19 pandemic [27]. This crisis triggered a global economic downturn [28] and led to a series of shocks in financial markets [29]. Researchers observe a significant increase in volatility, market stagnancy, a significant loss of investors, government intervention and policy adjustments in the global financial market [30,31]. Thus, the COVID-19 pandemic has brought new challenges and opportunities for firms and drawn an increased focus on their CSR practices and its shield-effect.
Moreover, this health pandemic has led to heterogeneous responses from firms and investors, which has drawn attention to the role of CSR during difficult times. The economic crisis emphasizes the significance of CSR as stakeholders demand socially responsible behavior from firms, where stakeholders serve as safeguards for the market system in which firms operate [32]. On the contrary, many investors prioritized firms that have more flexibility in reducing labor costs compared to capital costs when their revenues decline during COVID-19 crisis [33]. During the COVID-19 pandemic, the relationship between CSR and firms' performance remains an open question and continues to be the subject of ongoing research. This growing interest has motivated us to explore whether CSR is valuable enough to achieve firms' goal of maximizing shareholders' wealth during the COVID-19 market crash and subsequent market rebound.
In previous studies, the relationship between CSR and stock returns has been extensively examined in different time horizons and economic conditions, but the outcomes are yet inconclusive. However, it is widely believed that firms that have a strong commitment to CSR have been better able to withstand the economic and operational challenges posed by the pandemic and have generally been more resilient than firms that have not made such a commitment [34].
Nevertheless, previous findings may not necessarily be generalized to all countries and cultural context. South Korea was the second country to face a significant outbreak of COVID-19, following China, within a short period of time [35]. The first case of COVID-19 in South Korea was reported on 20 January 2020, and the first death was reported on 20 February 2020. Indeed, the Korean stock market has been significantly affected by the sudden shock of the COVID-19 pandemic, especially after the first cataclysmic outbreak reported on 18 February 2020, in Daegu city. Moreover, the Korean market presents a unique context for examining the correlation between CSR and stock returns, given that a significant number of Korean firms are under the control of chaebols, family-run conglomerates, and operated by family owner-managers [36]. Thus, we intend to observe intensively the role of CSR in protecting shareholders' wealth during economic turmoil in Korean economic, social, cultural and political context. Therefore, in this paper, we investigate the connection between CSR and stock returns of firms listed in the Korean stock market across the COVID-19 market crash and the subsequent recovery.
To achieve a broader idea of the association between CSR and firms' performance, we examine a sample of 803 firms listed in the Korean stock market throughout the COVID-19 crisis and post-crisis recovery periods. Using the environmental, social and governance (ESG) score as a proxy of firms' CSR provided by Korea Corporate Governance Service (KCGS), we find that CSR negatively affects firms' stock return throughout the COVID-19 market crash. Furthermore, the results are ambiguous and do not support the belief that CSR is efficient enough to protect shareholders' wealth during the crisis and its aftermaths.
Later, we conduct a number of robustness tests to validate our core evidence. Firstly, we select a sub-sample of 50 firms with a higher CSR score to examine whether the high-CSR firms performed better throughout the COVID-19 crisis period. The results find no significant evidence that high-CSR firms outperformed during the crisis. For our second additional test, we exclude penny stocks from our sample to analyze the performance of large-cap firms during the sample periods. The results suggest that large-cap firms did not perform differently, which supports the core findings of this study. Next, we investigate the impact of ESG grading on firms' stock return during the sample periods, and the results conclude that ESG grading does not have a positive impact on firms' stock returns. Moreover, it suggests that ESG score and ESG grade have a similar impact on stock returns and are therefore corresponding.
Furthermore, we estimate abnormal stock returns based on the market model during the sample periods as an alternative measure of firms' stock performance. Afterwards, we investigate the association between firms' CSR and abnormal stock returns and conclude that CSR failed to safeguard shareholders' wealth during the crisis period as no significantly positive connection is found between these two.
Finally, we analyze the association between firms' CSR and stock returns over the pre-crisis period, which is defined as "normal time period" in our study. However, we find that CSR was unconnected to stock performance during the pre-crisis normal period, which validates the idea that CSR is not effective enough to protect shareholders' wealth even when the market is stable.
Prior studies have provided conflicting findings regarding the extent to which CSR creates value and safeguards shareholders' interest during a crisis. This study contributes to the literature by investigating the impact of CSR on firms' stock return during COVID-19 crisis and post-crisis recovery periods in the context of the Korean stock market. Empirically, this study addresses the gap in the literature by presenting evidence on whether CSR is effective enough to protect shareholders' wealth during a crisis. In this study, two distinct measures of shareholders' wealth are employed: cumulative stock return and abnormal stock return. Cumulative stock return is considered a market indicator of stock performance, which reflects the overall movement and performance of the stock price over a specific period of time. Abnormal stock return, used the robustness test, is estimated based on the market model, which takes into account the overall market performance, industry trends and stock characteristics. Furthermore, this study incorporates both ESG score and ESG grading as proxies for firms' CSR, providing a more comprehensive understanding of their impact on stock return. Moreover, this study investigates how the environmental, social and governance components of CSR independently affect stock returns and reports their significance on shareholders' wealth during a crisis. Additionally, this study conducts additional tests to validate the consistency of the baseline results by analyzing different sub-sample groups of firms (high-low CSR, large-cap) over the sample period. It also analyzes the relationship between CSR and stock returns during the normal time period to provide a clearer understanding of how CSR influences stock returns in the absence of external shocks. Moreover, this study contributes evidence to the limited empirical research that has explored the connection between CSR and stock returns of Korean firms amid the COVID-19 crisis. Therefore, this study aims to provide insights into the question of whether it is rational to draw an unconditional conclusion about the value of CSR, even in times of crisis in the Korean market context.
Our paper can be connected to [37], who show that high-CSR firms outperform during the COVID-19 crisis period in the Korean stock market. Moreover, the authors also report that high-CSR matters more during unexpected economic shocks when investors' confidence is unstable. For their analysis, the authors use five ESG gradings from KCGS as a proxy of CSR. However, we investigate the relationship between CSR and stock performance by using firms' total ESG score from KCGS as a proxy of CSR. In an additional test, we also use ESG grading as a proxy for firms' CSR. However, in contrast to their findings, we report that firms' CSR negatively affects stock returns during the COVID-19 crisis period. Additionally, the authors report no significant relation between CSR and stock return during the COVID-19 pre-crisis period, supporting the belief that the impact of CSR works well during unexpected economic downturns. Likewise, we find that firms' CSR is unconnected to stock return during the pre-crisis period. However, we choose a longer time period (1 year period prior to the COVID-19 crisis) to examine the effect of CSR on stock return during normal market conditions, compared to the authors' one-month period.
Moreover, our study can be linked to [38], who find no evidence that firms' ESG score is positively associated with stock return during the COVID-19 pandemic. The authors indicate that they have selection bias in their analysis, as not all Japanese firms obtain an ESG score. The situation is quite similar for the Korean stock market, where only a proportion of firms are engaged in CSR activities, and the perceptions of stakeholders are yet unclear. Earlier, ref. [39] report that although CSR performance may be valuable to some participants in the Korean stock market, private sector investors may not consider it an important factor. Therefore, our results may be influenced by various factors, including the nature of Korean business culture, the specific CSR practices of Korean firms, and the regulatory environment in which they operate.
This research focuses on investigating the impact of CSR on stock returns during the COVID-19 crisis and post-crisis periods by addressing the following research questions in the South Korean context:

•
Is the firm's pre-crisis CSR effective enough to shield shareholders' wealth during the crisis?
Overall, our study provides evidence that pre-crisis CSR negatively impact stock returns over the COVID-19 crisis and post-crisis periods. The outcome remains consistent over multiple additional tests and provides strong evidence that the firm's CSR is not effective enough to protect shareholders' wealth in the Korean market.
Our paper is organized as follows: Section 2 provides a review of the existing literature. Section 3 reports the sample structure, descriptive statistics and methodology. Section 4 reports the results of our main empirical analysis and additional robustness tests, where we examine the relationship between firms' CSR and stock performance during the COVID-19 crisis and post-crisis periods. Section 5 concludes the paper.

Literature Review
Existing literature argues that firms that engage in CSR activities are more likely to be perceived positively by stakeholders and may benefit from increased brand image and reputation [3], improved customer loyalty [4], increased employee engagement, better financial performance [40,41], higher stock performance [42] and ensure market competitiveness [15] by mitigating information asymmetry [43] in the long run.
The relationship between CSR and stock return has been explored in previous studies from various perspectives, but the results remains inconclusive. Ref. [14] reports that socially responsible investment (SRI) strategies can potentially lead to improved stock returns. Ref. [13] argue that high CSR portfolios have higher returns than low CSR portfolios, especially during times of market stress, suggesting that CSR can play a stabilizing role in times of crisis. Likewise, ref. [42] report that the relationship between firms and stakeholders built through CSR engagements pays off during economic downturn, and thus high CSR firms enjoy better stock returns.
While the previously mentioned studies have shown a positive relationship between CSR and firms' performance, there is also a significant body of research that reports a negative association between these two. Researchers argue that CSR activities can divert firms' resources from the core business operations and thus negatively impact their financial performance, especially in the short term. Moreover, some stakeholders may view CSR initiatives as costly, irrelevant or distracting from firms' primary goal of maximizing shareholders' wealth [44,45]. Several concerns are reported that can worsen the relationship between CSR and stock returns, including conflicting interests, market perception, economic condition and time horizon. Ref. [16] report that firms with good social performance offer poor financial performance. Moreover, ref. [17] report a negative relationship between CSR and firms' future stock returns, implying that there may be a trade-off between firms' CSR initiatives and their financial performance. By analyzing Korean firms, ref. [46] find that social ties have more negative moderating effects on the relationship between CSR and firms' performance.
The COVID-19 pandemic has compelled firms to re-evaluate their priorities, and while many firms have taken steps to demonstrate their commitment to stakeholders', some have taken contradictory actions. However, it is essential for firms to consider the value of CSR in achieving their goal of maximizing shareholders' wealth in times of economic uncertainty. Firms that continue to prioritize and invest in CSR activities may not only improve their reputation and relationships with stakeholders but also potentially enhance their short-term and long-term financial performance [47][48][49][50]. Moreover, firms with strong CSR performance have been provided downside protection from the potential losses during the COVID-19 pandemic [51]. On the other hand, firms that have neglected their CSR responsibilities may have experienced a significant declines in stock prices during the COVID-19-induced market crash.
According to [52], stocks of firms with higher CSR tended to have higher returns during the early stage of the COVID-19 crisis, attributing this to the positive impact of CSR on customer loyalty, which in turn helps to support the financial performance of firms. Ref. [53] report that firms which engaged in more CSR activities prior to the COVID-19 pandemic enjoyed better stock performance by strengthening their ties with stakeholders. Similarly, ref. [54] conclude that CSR engagement can increase stakeholders' attention, which in turn can lead to improved stock return. Considering firm's ESG as an important factor of industry return, ref. [55] report that firms with higher ESG score outperform during COVID-19.
On the contrary, ref. [56] observe that firms with higher pre-crisis CSR have worse stock return during the crisis period and poor post-crisis operating performance. Nevertheless, few studies find it challenging to establish a clear link between firms' CSR and stock performance. Ref. [57] report that ESG has no explanatory power for stock return over the COVID-19 market crisis period. Ref. [18] find no evidence that CSR affects stock returns and suggest that there may be a potential discrepancy between firms' CSR initiatives and actual actions.
The existing disparity in prior findings indicates that firms may claim to be socially responsible, but their actual actions and practices may not align with those claims. This discrepancy can result in a lack of credibility and trust among stakeholders, which can have a negative impact on the firms' reputation and long-run financial performance. The findings of [19] suggest that there is no evidence that firms with high engagement in ESG initiatives have higher stock performance compared to firms with low ESG engagement. Similarly, ref. [38] find no evidence that firms with high ESG scores are positively associated with higher abnormal returns during the COVID-19 pandemic. The divergent outcomes of previous studies have created an excellent opportunity to study the relationship between CSR and firms' performance and to discover whether firms can effectively balance their dual responsibilities to society and shareholders during the COVID-19 crisis.
However, some studies have reported that the impact of CSR on firms' performance can be asymmetric, meaning that the relationship between CSR and stock performance may differ depending on the stage of the pandemic or the specific aspect of CSR being considered. For instance, the positive impact of CSR on firms' performance was more pronounced during the early stages of the COVID-19 pandemic when the economic and operational challenges posed by the pandemic were most severe. Nevertheless, firms with strong CSR practices may be better positioned to navigate the post-pandemic economic environment, given that consumers and investors are increasingly seeking out socially responsible firms. Ref. [37] demonstrate that stock returns of higher CSR firms decline less during the COVID-19 pandemic period and rebound less during recovery period in Korean stock market. Likewise, ref. [58] report that ESG performance significantly increase firms' cumulative abnormal returns and had asymmetric effects during the pandemic. Moreover, ref. [34] report that CSR firms were less affected by the COVID-19 outbreak, remaining relatively resistant to the fall and recovering faster than non-CSR firms in both the short and long term. Previous findings highlight the importance of considering the timing and specific aspects of CSR when evaluating its impact on firms' performance during times of crisis and uncertainty. Hence, while analyzing the impact of CSR on stock performance, we consider the time horizons of the COVID-19 pandemic and aftermaths separately.
In another context, it is important to note that the relationship between CSR and firms' performance is complex and can be influenced by many factors, including the income level of the countries in which firms operate, market efficiency, firm-specific variables (i.e., size, industry, and financial performance) [59], the type of CSR activities pursued, alignment with the core business and the level of transparency and accountability in reporting. Ref. [60] found that ESG is beneficial during COVID-19, but the impact depends on the income level, firm-specific variables and specific aspects of ESG. In our study, we incorporate several influencing factors while analyzing the relationship between CSR and stock returns to gain a clear understanding.

Sample
The aim of this study is to examine the association between firms' corporate social responsibility (CSR) and stock returns during the COVID-19 crisis and post-crisis periods. The environmental, social and governance (ESG) score of each firm is obtained from the Korea Corporate Governance Service (KCGS) database, which serves as a measurement of the firm's CSR. The KCGS database provides the total ESG score and grading for domestic listed firms. Moreover, stock return and financial data are retrieved from FnGuide for the firms available in the KCGS database. Throughout this study, we use the terms CSR, pre-crisis CSR and ESG interchangeably.
In this study, we investigate two sample periods separately to document the effect of CSR on stock returns during the COVID-19 crisis and post-crisis recovery periods. We define the COVID-19 crisis period from 18 February to 20 March 2020, while post-crisis period from 23 March to 5 June 2020. We use the available ESG total score, prior to the COVID-19 crisis as a measure of firms' CSR, estimated at the year-end of 2019 for our study. For the two sample periods, we collect daily firms' stock return data from FnGuide for the available firms. Moreover, annualized financial data are retrieved for the year-end of 2018, as the fiscal year for all firms documented in KCGS dataset is April-May during the year of 2019.
As a measure of the stock performance, we calculate the cumulative stock return for all available firms over the sample periods using daily stock returns, which are labeled as crisis return and post-crisis return in our analysis. To measure a firm's CSR, we collect data on the firm's total environmental, social and governance (ESG) score from KCGS. Furthermore, several variables are calculated to proxy firm characteristics that may affect stock performance. To control the size effect of a firm, we use log (ME) which is the natural logarithm of the firm's market capitalization. To capture a firm's financial strength, leverage and cash ratio are calculated based on accounting data. Leverage is the firm's debt-to-assets ratio, calculated as total debt over total assets. Cash ratio is the firm's cash and cash equivalent divided by total assets, used as a proxy for firm liquidity. Profitability is calculated as the firm's operating income scaled by total assets. The firm's book-tomarket ratio, labeled as B/M, is calculated as the book value of the firm's equity scaled by market capitalization. Moreover, we add a dummy variable, negative B/M, which takes the value of one if the firm's B/M is negative and zero otherwise. We include Negative B/M in our study as the returns of the firms with negative book-to-market ratios may behave similarly to those of firms with higher book-to-market ratios [61]. Momentum is the firm's cumulative stock return over the 1-year period prior to the COVID-19 crisis (January-December 2019). Idiosyncratic risk is estimated as the residual standard deviation from market-adjusted returns estimated based on 20 years of monthly data.
Financial firms are excluded from the dataset due to their higher leverage. We combine all non-financial firms with available data in the FnGuide stock return and financial databases with the available firms in the KCGS database. Therefore, we construct a crosssectional sample of 803 (802) firms for which all predictor variables are accessible during the COVID-19 crisis (post-crisis) period.

Descriptive Statistics
We report the descriptive statistics of all variables in our cross-sectional sample dataset in Table 1. The average return is strongly negative, with a mean of −34% and a median of −36% (25th percentile value of −45%, unreported) during the crisis period. This indicates a concerning situation for the investors and the other stakeholders regarding the survival of the many firms they are associated with. Conversely, the average return throughout the post-crisis period is 59% with a median of 55%. The average post-crisis return provides evidence of market recovery after the COVID-19 financial market crash. The average CSR score for all the non-financial firms in this dataset is 2.62 with a median of 2.55, indicating that many non-financial Korean firms were unable to obtain higher ESG scores in 2019 (on a scale of seven). The extreme values of firms' return and CSR demonstrate the presence of outliers in our sample dataset. However, the sample size is small, and these outliers are believed to be random. Henceforward, subsequent analyses are conducted with this dataset with confidence to capture the prominent information these outliers hold.
We further conduct a correlation analysis to perceive the relative associations between the variables in our dataset, and the results are presented in Table 2. We observe that the crisis period return is negatively correlated with the post-crisis period return (−0.39), which is reasonable enough to explain the scenarios of a market crash due to the COVID-19 pandemic and market recovery afterwards. However, CSR shows insignificant and negative correlation with the stock return during the crisis period. On the other hand, during the post-crisis period, the association between CSR and return is significantly negative. This may add fuel to the debate on whether firm's CSR is genuine to protect the shareholders' wealth during financial turmoil.

Methodology
The purpose of this study is to estimate the association between the firm's CSR and stock performance during a crisis. The ambiguous outcomes from previous studies motivate us to explore the relationship between pre-crisis CSR and the firm's stock return during the COVID-19 crisis and post-crisis periods in the South Korean stock market.
If firms are consistently socially responsible, we expect to observe a significant and positive association between CSR and stock performance, even during the crisis period. Specifically, we hypothesized that a firm's pre-crisis CSR has a positive impact on stock returns during the COVID-19 crisis and post-crisis periods. Therefore, we conduct several cross-sectional analyses over the sample periods to test the hypothesis in our study.
The baseline regression models contain cumulative stock return as the dependent variable for two sample periods and the firm's CSR score as the variable of interest. We include several firm-level variables in our models to address the possibility that these characteristics may affect stock returns. CSR investments and the impacts of the financial crisis may differ across industries. To account for this effect, industry dummy is included (defined in the 10th Korea Standard Industrial Classification) in all regression models. Furthermore, we also control for the Fama-French three factors model and the momentum factor in our regression analyses. All regression models are adjusted for heteroscedasticity, and t-statistics are presented in parenthesis. Statistical significance is reported at the 10%, 5% and 1% levels denoted by *, ** and ***, respectively. We use the following model specification for our analysis: Here, R i is the firm's cumulative stock return over the sample periods, CSR i is the firm's ESG score for the year-end of 2019, Control Variables indicates the firm-level variables for the fiscal year-end of 2018, Factor Loadings are incorporated to control for the Fama-French three factors and the momentum factor for the year 2019, and Industry Fixed Effects are the dummy variables for the 16 industries based on the 10th Korea Standard Industrial Classification, excluding the financial industry.
In our analysis, our focus is on the β 1 specified in (1) to quantify the association between firms' CSR and stock returns during the COVID-19 crisis and post-crisis periods. The results of our empirical analysis are reported in Section 4.

Empirical Results
To test the association between firms' stock return and CSR, we estimate several cross-sectional regression models following the specification (1). Table 3 reports the results of the baseline regression models of firms' stock returns as a function of firms' pre-crisis CSR score and a number of control variables during the COVID-19 crisis and post-crisis periods. The dependent variables are the crisis period return and the post-crisis period return, respectively. The variable of interest of our regression analysis is the firms' ESG score measured at the year-end of 2019. Furthermore, firm-level controls, Fama-French three factors and momentum factors as market-level controls, and industry dummies as industry-level controls are included in different models. We confirm that the firms' CSR negatively affects the respective stock returns during the COVID-19 crisis period. In the presence of firm-level control variables in the crisis period models, the relationship between CSR and stock return becomes significantly negative; otherwise, it is insignificant.

CSR and Firms' Stock Return
However, while investigating the post-crisis period, contradictory outcomes are identified using similar models. Coefficient estimates are significantly negative in the absence of firm-level controls, but insignificant otherwise.
Overall, the outcomes confirm that CSR-firms had lower stock returns throughout the COVID-19 crisis and firms' CSR was unsuccessful to shield the shareholders' wealth during crisis period despite of having high expectations.
Moreover, large and undervalued firms that entered the crisis period with lower debt and lower idiosyncratic risk enjoyed comparatively higher crisis-period returns or at least faced less turmoil due to the market crash. On the other hand, small firms with higher debt and risk experienced higher stock returns and a quicker recovery during post-crisis period.

ESG Components and Firms' Stock Return
In Table 4, we examine the relationships between the components of ESG and firms' stock return during the COVID-19 crisis and post-crisis periods. Hereby, we re-analyze the previous models for the firms' ESG components separately. The ESG score has three major components: environmental, social and governance. The E-score (Environmental) consists of three categories-environmental management, environmental performance and shareholder response. The S-score (Social) consists of four categories-employees, partners and competitors, consumers and community. The G-score (Governance) for non-financial firms consists of four categories-protection of shareholders' rights, board of directors, disclosures and audit body. We acquire the total score for each of these three components from the KCGS database. The coefficient estimates on the E-score are negatively associated with firms' stock return, and during the crisis period, it is statistically significant. The S-score shows a negative (positive) impact on crisis (post-crisis) period stock return. G-score also displays mixed associations with COVID-19 crisis and post-crisis period stock return.
ESG components give us mixed outcomes about their associations with the COVID-19 crisis and post-crisis period stock returns. However, it is reasonable to conclude that firms' CSR failed to protect shareholders' wealth during COVID-19 crisis and post-crisis periods.

Robustness Tests
The empirical results from the previous chapter suggest that firms' pre-crisis CSR was unsuccessful in protecting the shareholders' wealth during the COVID-19 crisis and pre-crisis periods. Therefore, we conduct several additional tests to support our baseline regression results in this section.

Performance of High-CSR Firms
One may argue that firms with excellent CSR score successfully protected their shareholders' well-being during the COVID-19 crisis, while other firms failed to do so. To address this dispute, we repeat the specification (1) for a sub-sample of firms that have comparatively higher CSR score than that of the other firms in our main data sample.
For constructing the sub-sample, we selected 50 firms with a CSR score higher or equivalent to 4.5 (on a scale of 7). A CSR score equivalent or above 4.5 (95th percentile) is defined as "Excellent" CSR score [62] in our sample. We designate this high-ranked CSR group of firms as our sample of interest for this section and explore their stock performance during the COVID-19 crisis and post-crisis periods.

Descriptive Statistics and Industry Composition of High-CSR Firms
We report the descriptive statistics for the high-CSR firms, as well as for the 753 low-CSR firms in Table 5 (Panel A). The average crisis period and post-crisis period returns do not vary significantly between the two groups, raising doubts about the commitment of the high-CSR firms. The average CSR score for the high-CSR firms is significantly higher (4.92) than that of the other firms (2.47). However, the firms' characteristics do not portray any significant difference between these two groups prior to the COVID-19 crisis. In this section, we report the industries that comprise the majority of firms in our sample dataset. Both groups have a higher percentage in manufacturing, retail and wholesale and in professional, scientific and technical services industries. Moreover, low-CSR firms have a higher percentage in the publishing, video, broadcasting, communication and information service business (7.4%) industry. This outcome suggests that there is no severe imbalance between industries for both groups.

The Relationship between High-CSR Score and Firms' Stock Performance
In this section, we examine whether firms with excellent CSR score perform differently during the COVID-19 crisis and post-crisis periods. The results in Table 6 suggest that firms with comparatively higher CSR scores did not perform differently during the COVID-19 crisis, although they are believed to be more socially responsible and would protect shareholders' wealth during a crisis.
This table reports the results of the additional regression analysis for firms' stock return on respective CSR and other control variables during the COVID-19 crisis and post-crisis periods. All regression models are adjusted for heteroscedasticity and t-statistics are presented in parenthesis. *, ** and *** indicate the significance at 10%, 5% and 1% level, respectively.
For the high-CSR firms, CSR is positively (negatively) associated with stock return during the COVID-19 crisis (post-crisis) period, but none of these are statistically significant. Meanwhile, firms with low-CSR show significantly negative or insignificant association between firms' CSR and stock performance during the crisis and post-crisis periods. Our findings contribute to the ongoing debate regarding whether firms' CSR commitment is basically a rhetorical move to gain public interests.

Excellent CSR and Stock Return
We extend the previous analysis to test whether the firms with excellent CSR scores perform better over the sample periods. At this point, we include an additional control variable, "Excellent Dummy × CSR," which is a dummy variable that takes the value of 1 for firms with excellent CSR score (≥4.5) and 0 otherwise, multiplied by the respective CSR score. We test the following specification for the analysis: Table 6 reports the results of the regression model, where firms' stock return is regressed on their pre-crisis CSR score and several control variables, including the aforementioned "Excellent Dummy × CSR." The coefficient estimates on CSR are negative or insignificant during both the COVID-19 crisis and post-crisis periods. However, the intriguing part is that the coefficients on the Excellent Dummy × CSR during the crisis period are also negative and follow the same pattern as CSR. Moreover, the coefficient estimates for the Excellent Dummy × CSR show ambiguous significance during the post-crisis period.
These results illustrate a similar conclusion to our baseline findings and suggest that firms with comparatively higher CSR scores did not perform any better during the COVID-19 crisis period and firms' CSR engagement may possibly be all-talk.

CSR and Large-Cap Firms
In this segment, we investigate the performance of large-cap firms during the COVID-19 crisis and post-crisis periods. Therefore, we exclude penny stock firms (closing stock price less than 1000 Korean won) from our dataset. We exclude penny stocks since they are highly risky, less liquid and manipulative in nature. Moreover, these firms are subject to more price pressure, and the effects can be more pronounced during a crisis period.
The sub-sample consists of 786 non-financial firms. We report the results of the regression analysis of the large-cap firms' stock return on their respective CSR and the other control variables in Table 6. The coefficient estimates of CSR are significantly negative or insignificant during both the COVID-19 crisis and post-crisis periods.
In short, even if we exclude penny stocks from our dataset, the results do not reveal any different consequences. The overall result suggests that the CSR score of large-cap firms, which is believed to safeguard their shareholders' wealth during a crisis, failed to do so during the COVID-19 crisis period.

ESG Grading and Stock Performance
The ESG database of KCGS provides ESG grading along with firms' ESG scores. ESG grading is discreet in nature and takes on values within seven letter grades (S, A+, A, B+, B, C, D). Addressing the idea that firms' ESG grading may affect stock performance, unlike our baseline empirical results, we examine the association between firms' ESG grading and stock return during the COVID-19 crisis and post-crisis periods. After obtaining firms' ESG grade from KCGS for the year-end of 2019, we assign a subsequent series of numbers ranging from 1 to 7 to all firms based on their ESG grading (S = 7, A+ = 6, A = 5, B+ = 4, B = 3, C = 2, D = 1).
The findings of the regression of ESG grading on stock return are reported in Table 6. The coefficient estimates of ESG grading are negatively or insignificantly associated with stock return during the COVID-19 crisis and post-crisis periods. Hence, we suggest that firms' ESG grading fails to explain the shield effect of CSR that commits to protecting shareholders' wealth during economic turmoil.

Excellent ESG Grading and Stock Return
Additionally, we perform an extended analysis to examine whether firms' excellent ESG grading affects stock performance over the sample periods. We include an additional control variable "Excellent Dummy × ESG Grade," which takes the value of 1 for firms with excellent ESG grade (equivalent to 5 or higher) and 0 otherwise, multiplied by the firms' corresponding ESG grade. Table 6 reports the regression results. The coefficient estimates on ESG grade are insignificant or negative during the COVID-19 crisis and post-crisis periods. However, the coefficients of "Excellent Dummy × ESG Grade" report significantly negative estimates during the COVID-19 crisis period and weakly positive estimates during the post-crisis period.
The results suggest that firms with excellent ESG grading were unable to protect shareholders' wealth during the COVID-19 crisis period. Henceforth, this outcome provides us with strong evidence that CSR is somehow a misleading act of firms that is still far from providing protection to shareholders' well-being.

CSR and Firms' Abnormal Return
Furthermore, we consider an alternative measure for firms' stock performance, "Abnormal return" and rerun the baseline regression model to examine the relationship between firms' CSR and abnormal stock return during the COVID-19 crisis and post-crisis periods.
Abnormal return is the firms' stock return minus the expected return calculated based on the market model to address the firms' responsiveness towards the market portfolio. In Table 6, we report the outcomes of the tests whether firms' CSR affects stock performance over the COVID-19 crisis and post-crisis periods, considering the abnormal stock return as a proxy for stock performance. We determine the market model using five years daily stock returns (2015-2019) and the KOSPI composite index as the market return to estimate the abnormal returns for firms over the sample periods.
Therefore, the outcomes illustrate that the coefficient estimates of CSR are negative over the COVID-19 crisis period and mixed association during the post-crisis period. By analyzing the results, we can suggest that firms' CSR failed to defend shareholders' wealth during the crisis.

CSR and Firms' Pre-Crisis Stock Performance
The baseline empirical results of our study show that firms' CSR did not protect the shareholders' wealth during the COVID-19 crisis period. One may argue that this evidence is documented because the COVID-19 health crisis is truly an exogenous shock. To prove the consistency of our main evidence, we need to check if firms' CSR is associated with better stock returns during "Normal" time.
Therefore, we examine the association between firms' pre-crisis period stock return and their respective CSR score, as reported in Table 6. For the "Normal" time period, we obtain the stock return of 797 non-financial firms over the one-year period prior to the COVID-19 crisis, from January to December 2019. The dependent variable for this analysis is the "Pre-crisis Stock Return," calculated as the cumulative stock return over the sample period. Additionally, we acquire firms' CSR score for the year-end of 2018 from the KCGS database.
The results show that the coefficient estimates of CSR are insignificant for both models, suggesting that firms' CSR is not related to stock return even during the pre-crisis period. It is evident from our analysis that firms' CSR is unrelated to stock return even without any external shock, thus validating our baseline results.

Conclusions
Is firms' CSR valuable enough to preserve the shareholders' wealth during a crisis? In our study, we try to address this buzzing question that many researchers are actively investigating. In this paper, we provide several empirical pieces of evidence and conclude that CSR is not effective enough to protect shareholders' wealth during a crisis.
The COVID-19 health crisis is different from previous market crashes [25] and represents a unique challenge for the global economy, which may have different implications for firm performance. This economic turmoil provides us with a distinct opportunity to investigate the shielding effect of CSR on shareholders' wealth. Therefore, we investigate the association between firms' pre-crisis CSR and stock return during the COVID-19 crisis and post-crisis periods to provide valuable insights for investors, policymakers and firms alike.
According to the traditional perspective, CSR has a positive impact on firms' performance and contributes to the long-term well-beings of stakeholders [63], while also aligning with the goal of maximizing shareholders' wealth [64]. It suggests that being socially responsible can help firms avoid conflicts with stakeholders, but the ultimate goal of protecting shareholders' wealth should not be compromised [65]. This perspective is supported by previous literature and theoretical frameworks on CSR and business strategy [1].
Prior studies report mixed evidence on the effects of CSR on stock return during the COVID-19 crisis. Some find that firms with greater corporate social involvement outperformed during the COVID-19 stock market crash [66]. Additionally, investors prefer funds that exhibit strong sustainability performance, and sustainable funds outperform during the COVID-19 pandemic [67]. Ref. [68] report that firms with higher ESG ratings are associated with higher abnormal return during and lower stock volatility during the crisis. Moreover, the social component of CSR is found to be more prominent in major findings [53], where family-owned and large firms performed better during the COVID-19 crisis.
Nevertheless, some studies report disconnected findings [57] or a negative association between CSR and stock return and claim that ESG scores are outdated and relying solely on these scores is no longer adequate [69]. In addition to that, some consider CSR as a risk factor that leads to a lower expected return, although it offers multiple social benefits [70].
In our study, we conducted several empirical analyses to investigate the hypothetical belief that CSR protects shareholders' wealth by observing the relationship between firms' CSR and stock returns. We estimated the baseline regression model of stock return as a function of firms' CSR and several control variables. Our sample consisted of 803 non-financial Korean firms with ESG scores in the KCGS database. We examined the cross-sectional variation in CSR and stock returns among these firms while controlling for several firm and market-level factors, along with industry controls that may affect stock returns during a crisis.
Henceforth, we found that firms' CSR had a negative impact on stock return during the COVID-19 crisis and post-crisis recovery periods, which calls into question the protective effect of CSR. Moreover, we make a few interesting discoveries in our study. Firstly, we found that none of the ESG components had a substantial positive impact on stock return during the crisis period, contrary to previous studies that highlighted the significant influence of the social score [68]. Secondly, firms with excellent CSR scores did not perform any better during the crisis period. Thirdly, CSR of large-cap firms was negatively associated with stock return, which contradicts the popular belief that large firms may exhibit greater resilience during a crisis [53]. Finally, we did not find significant evidence that CSR affects stock return even during the pre-crisis "normal" time period. Our findings suggest that the relationship between CSR and stock return may be context-dependent, and CSR may not be an effective defense mechanism during the times of crisis and its aftermath.
Our study sheds light on some broader issues related to corporate ethics, management and the agency problem, which are vital considerations in achieving the goal of maximizing shareholders' wealth while also being socially responsible [44,45]. In another context, firms may engage in CSR activities not only for ethical reasons but also as a strategy to mitigate potential legal risks and improve their public image by creating a halo effect [71]. Hence, the true intention of CSR engagements and whether it is enough to benefit shareholders during a crisis is still a subject of debate.
The findings of this study provide some valuable insights for the researchers, policymakers and practitioners. Firstly, our study highlights the need for Korean firms to reassess their CSR approach and modify it to better protect shareholders' wealth and mitigate negative impacts on stock returns during crises, rather than solely focusing on gain legitimacy [72] in the eyes of outside stakeholders. Secondly, as the Korean market is dominated by family-owned firms [73], there should be a focus on establishing strong corporate ethical policies and ensuring compliance with them by identifying and mitigating agency issues. Thirdly, our findings suggest a need for balancing the social and financial performance of Korean firms, as a higher CSR score failed to protect shareholders wealth during the COVID-19 crisis. Finally, policymakers and regulatory authorities may consider revisiting and refining guidelines related to CSR reporting and disclosure to promote a more balanced approach that incorporates both societal goals and shareholders' interests, particularly in times of crisis.
Over the past few years, many Korean firms have taken initiatives to incorporate CSR into their management strategies, driven by increasing global awareness and updated regulatory guidelines [74]. In addition to that, the South Korean government and regulatory bodies have implemented measures to enhance the standard of CSR reporting in accordance with global standards [75]. The current circumstances present an opportunity for Korean firms to enhance the transparency and quality of their CSR investments and disclosure, fostering a more sustainable and value-driven business environment. Nevertheless, further research is needed to explore the underlying mechanisms and factors that influence the relationship between CSR and stock returns during crises, contributing to a deeper understanding of the complex dynamics involved.
Overall, we conclude that firms' CSR is not sufficiently valuable in preserving shareholders' wealth during a crisis, rejecting the unconditional belief that CSR has a positive effect on stock performance. This highlights the rising demand for further research to better understand firms' intentions for engaging in CSR and the potential impact on shareholders' wealth, particularly during times of crisis.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
No new data were created or analyzed in this study. Data sharing is not applicable to this article.