key: cord-0878613-urpjsclx authors: Yi, Yuyang; Zhang, Zongyi; Yan, Youliang title: Kindness is rewarded! The impact of corporate social responsibility on Chinese market reactions to the COVID-19 pandemic date: 2021-09-08 journal: Econ Lett DOI: 10.1016/j.econlet.2021.110066 sha: 0c9f4214098aff2bb8d44bc581413db396639853 doc_id: 878613 cord_uid: urpjsclx This study takes the COVID-19 outbreak as a quasi-natural experiment to investigate whether corporate social responsibility (CSR) performance can help firms mitigate drops in their share prices. The results show that CSR ratings are positively associated with cumulative abnormal return (CAR) during the COVID-19 outbreaks periods. Further, the positive role of CSR is more significant for non-state-owned enterprises (non-SOE) and those located in regions with lower levels of marketization. The COVID-19 outbreak that started spreading globally in early 2020 had a significant negative impact on capital markets. However, did this pandemic affect firms in the stock market differently? Additionally, we investigate whether and how does corporate social responsibility (CSR) affect firms' value? This question is of academic and practical significance. We therefore study whether and how CSR performance affects stock-price responses to negative events with this background. Prior empirical research is controversial on whether CSR is positive or negative for companies. According to the stakeholder theory, CSR contributes to corporate financial performance. In support of this thesis, prior researches point out that CSR performance enhances a firm's reputation and social capital, thereby leading investors to trust the firm, providing an insurance-like effect (Godfrey et al. 2009; Pevzner et al. 2015，Lins, et al 2017 . Cheng (2014) indicated that CSR performance has a signaling effect that reduces information asymmetry. Conversely, the agency theory posits that managers engage in CSR activities, rather than devoting their efforts to shareholder wealth maximization, because they can extract private benefits (Friedman1970). Accordingly, prior studies pointed out that CSR investments may have a negative association with financial performance (Barnett et al. 2006; Surroca et al.2010) . The reason for this controversy is endogeneity. To solve the endogeneity problem behind this controversy, the COVID-19 outbreaks-induced market crash, as a negative exogenous shock, creates a unique opportunity to test the notion that CSR protects firm value during crisis periods. This study finds that CSR ratings have a positive effect on firms' stock returns during the pandemic period. Furthermore, the positive role of CSR is more significant for firms located in regions with lower levels of marketization. Finally, this positive effect of CSR on cumulative abnormal return (CAR) is, in the short-term, greater for non-state-owned enterprises (non-SOEs) than state-owned enterprises (SOEs). The contributions of this study are as follows: First, on the controversial issue of how CSR affects market reaction, we chose the COVID-19 outbreak as a quasi-natural experiment to provide new evidence to address endogeneity. Second, we use CSR ratings provided by Hexun, which is the most comprehensive social responsibility rating database available for listed companies. Third, we explore the mechanism of how CSR affects CAR, and selected the marketization index of the firm's registered location and ownership structure as the moderating variables. Lastly, this study deepens the understanding of the market impact of the COVID-19. The rest of this article is structured as follows: Section2 describes the data and model setting; Section3 reports empirical results, and the last section is conclusion. We collected accounting data and stock returns about on listed companies from CSMAR database. According to the research needs, we deal with our original data as follows:(1) observations with missing values were deleted; (2) firms in the financial industry are also eliminated; (3) companies listed after 2018 were excluded. CAR is the dependent variable. The market models are the most commonly used and have good predictive power (Brenner 1979) . In our study, we used the market model below to calculate firm abnormal returns: where，Ri,t is the return rate of stock i on the trading day t, Rm,t is the market return of the trading market. As shown in Equation (1), we use the estimation window of [-150,-30 ] to estimate market betas. We then use those betas to calculate the firm's abnormal returns (AR) and cumulative abnormal returns (CAR[-n,n]i,t) around the event of the Covid-19 outbreak. We choose January 23, 2020 as the event day, when the Chinese government officially announced the lockdown of Wuhan. Specifically, the calculation of CAR[-n,n]i,t is shown in Equation (2) below: where ARi,t refers to the abnormal returns of firm i , on a given event day; CAR(-n, +n)i,t is the J o u r n a l P r e -p r o o f CAR of firm .̂ is the estimated constant in Equation (1)and ̂ is the estimated coefficient of Rm,t.. Figure 1 shows the daily abnormal returns of the Chinese stock market around the event day. CSR performance is the independent variable, which we selected CSR ratings of listed companies published by Hexun.com, as a proxy variable of CSR performance. Based on the annual reports of listed companies, Hexun measures CSR performance across four dimensions: shareholder, employee, environmental, and social responsibility, and get a total score of 100. The primary reason for choosing the Hexun CSR data is that they include almost all listed companies, which is the most comprehensive third-party social responsibility rating agency in China. CAR(-n, +n)i,t= 0 +1CSRSCORESi+ ∑2CONTROL+∑βIndustry fixed effects+ (3) where, CAR(-n, +n)i,t is the CAR of firmi in the event window period(-n,+n), CSRSCORESi refers to the CSR rating of firm i in 2019 from Hexun. The control variables include : return on equity(ROE), the total liabilities divided by total assets(LEV), the growth rate of the firm's operating income(GROWTH), the natural logarithm of sales revenue(Size), the percentage of corporate shares held by the largest shareholder(OWN1), a dummy variable equal 1 if the CEO is also Chairman of the board of director and 0 otherwise (DUAL), the natural logarithm of the number of directors on the board(BOARD), the percentage of independent directors on the board (INDRATIO), and the natural logarithm of years that the firm was established(lnLISTNF). We also control industry-fixed effects. We choose 3-,5-day event windows around the Wuhan lockdown . Table3 presents the main results. As columns 1 and 3show, the coefficients of the 3-and 5days event window are 0.0014 and 0.0017, respectively, and statistically significant at the 1% level. After adding the control variables, the coefficients in columns2 and 4 are 0.0013 and 0.0016, respectively, and statistically significant at the 1% level. This finding suggests that CSR performance alleviates the negative stock price effect of COVID-19, and therefore may be viewed as providing insurance-like protection and a signaling effect (Godfrey et al. 2009; Cheng, 2014) Table3:The impact of CSR scores on stock market reactions to (1) (2) Previous researches pointed out that CSR performance is related to market characteristics (Jha & Cox 2015) . Furthermore, a region's institutional environment affects listed firms' performance because information asymmetry should be more serious in low marketization regions (McGuire et al. 2012) . In this section, we test whether CSR can provide a signaling effect to alleviate information asymmetry with different level of marketization. We use Fan et al. (2011)'s marketization index as a proxy, which represent regional difference in the market development level (Jian and Wong 2010) . Then, we divide the sample into high and low groups based on the median marketization index of a firm's registered location. Table 4 reports the results. Columns 1 and 3 report that the coefficients of firms in low marketization regions, which are 0.0011 and 0.0013, respectively, and statistically significant at the 1 %level. Columns 2 and 4 show that the coefficients of firms in high marketization regions are positive, but statistically insignificant at the 1 %level. These results support our hypothesis that CSR can alleviate information asymmetry. One unique feature of Chinese listed firms is that non-SOEs have a different regulatory environment, management philosophy, and financial performance compare with SOEs. (Du et al.2014, Jian and Wong 2010 ) . Non-SOEs have more pressure on financing and more serious regulatory environment than SOEs during the COVID-19. Furthermore, we hypothesize that non-SOEs needed signaling effect and insurance-like effect of CSR more to alleviate a negative shock than did SOEs during the pandemic period. To compare the different effect of CSR on CAR for SOEs and non-SOEs, we divided sample into two groups and conducted regression. Table 5 reports the results. Column 1 and 3 reported the results for SOEs, and Column 2 and 4 report the J o u r n a l P r e -p r o o f results for non-SOEs. The coefficients of non-SOEs are 0.0010 and 0.0012, and statistically significant. The coefficients of SOEs are 0.0008 and 0.0010, but statistically insignificant. The coefficients of SOEs are lower than the coefficients of non-SOE in 3-and 5-event windows. The results can be explained that CSR had more significant effect on the stock prices of non-SOEs than those of SOEs (Qiao et al.2021 We alternate event window of the independent variable to conduct robustness test. The regression results are shown in Table 6 . CSR scores are positively associated with CAR over the 2-, 4-,and 6-day event windows-around the event day. These results are consistent with the previous regression. Therefore, the positive effect of CSR scores on CAR are still valid. Furthermore, we also alternated the dependent variable by calculating the average CSR rating from 2017 to 2019 to conduct a robustness test. The regression results are shown in Table 7 . CSR mean scores are positively associated with CAR. These results are still robust. The Shanghai Stock Exchange (SSE) mandates the constituent firms of the SSE corporate governance index, firms cross-listed overseas, and financial firms to disclose annual CSR reports. Meanwhile, the Shenzhen Stock Exchange (SZSE) mandates the constituent firms of the SZSE 100 index to disclose such reports. According to these requirements, a total of 335 firms in our sample are mandated to make CSR disclosures. As suggested by the reviewer, we use as observations the above mentioned firms only to check the robustness of our empirical findings in the previous manuscript. We report the related regression results in Table 7 below. Please refer to the manuscript for the definitions of the variables. As shown in Table 7 We measure CSR activities of Chinese A-share firms with CSR scores evaluated by the Hexun.com. The Hexun.com constructs CSR scores with three subcategories including shareholder responsibility, employee responsibility and social responsibility. That is, the Hexun.com does not provide scores on firms' environmental awareness. A possible reason is that the China Securities Regulatory Commission (CSRC) did not require listed firms to add an "Environment and Social Responsibility" section in their semi-annual and annual reports until 2021. Nevertheless, we studied the impact of each CSR subcategory on stock returns as suggested by the reviewer. We regress CAR on the three CSR rating subcategories in Table 8 . As column (1) and column (4) show, the coefficients of shareholder responsibility are positive and significant. Higher shareholder responsibility ratings imply consistent performance well in profit growth, debt J o u r n a l P r e -p r o o f 1 1 repayment, and innovation investment. Columns (3) and column (6) report that the coefficients of social responsibility are positive and significant. To score high social responsibility, a firm should have performed well in income tax to total profit ratio and Charitable donation. The coefficients of employee responsibility are statistically insignificant. Strong employee responsibility scores imply consistent performance in employee benefits, employee training, and safety training. The results in Table 8 suggest that shareholder responsibility and social responsibility of CSR rating have significantly positive effects on CAR during the crisis. However, employee responsibility has an insignificant effect on CAR. Therefore, shareholder responsibility and social responsibility of CSR are most important for protecting a firm's value during crisis breakouts. Higher scores in shareholder responsibility and social responsibility should help ensure overall financial stability and good reputation of a firm and offer a heightened resilience to shock events, including COVID-19. Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Beyond dichotomy: the curvilinear relationship between social responsibility and financial performance The sensitivity of the efficient market hypothesis to alternative specifications of the market model Corporate social responsibility and access to finance Does Religion Mitigate Earnings Management? Evidence from China The social responsibility of business is to increase its profits The relationship between Corporate Social Responsibility and Shareholder Value: An Empirical Test of the Risk Management Hypothesis Corporate social responsibility and social capital Propping through related party transactions Social Capital, Trust, And Firm Performance: The Value Of Corporate Social Responsibility During The Financial Crisis The impact of religion on financial reporting irregularities When Firms Talk, Do Investors Listen? The Role of Trust in Stock Market Reactions to Corporate Earnings Announcements Does targeted poverty alleviation disclosure improve stock performance? Corporate social responsibility and financial performance: the role of intangible resources This study provides new evidence that CSR did affect market reactions to the COVID-19 pandemic. We found that CSR ratings are positively associated with stock returns in short-term. These results support that the insurance-like effects and signaling effects of CSR during negative shocks. Furthermore, we compare the role of CSR in regions with different level of marketization. The results show that the positive role of CSR is more significant in low marketization region and support the signaling effect of CSR. Further, we compare the role of CSR according to ownership structure. The results show that the insurance effect and signaling effect of non-SOE's CSR is more significant than the SOE in the short-term. This paper extends the literature on the effect of CSR on companies' performance and the effect of the COVID-19 breakouts on economics and finance. Our findings also deepen the understanding of the relationship between CSR and market reactions during dramatic public health shocks.