The Effect of Listed Companies’ Social Responsibility on the Market Reaction to Violations

. This study investigates whether the listed company’s social responsibility will impact investors and regulators after corporate frauds. Does the corporate social responsibility (CSR) performance bring about the “compensa-tion” e ﬀ ect or the “reputation collapse” e ﬀ ect? Using the cumulative abnormal return to characterize the market reaction, our results based on the Chinese listed ﬁrms that have committed violations throughout 2010-2020 show that the announcement of violation enforcement will produce signiﬁcantly negative market reaction, and the more serious the type of violation enforcement, the greater the negative market reaction. Further, we ﬁnd that the corporate social responsibility (CSR) performance will not directly a ﬀ ect the market reaction of the announcement of punishment for violations, but it will increase the severity of the company’s punishment for violations, thus indirectly bringing more negative market reaction. Overall, our ﬁndings suggest that the listed company’s social responsibility (CSR) performance before violation cannot play a “compensation” e ﬀ ect to o ﬀ set the negative impact caused by its violation, it may bring more serious administrative punishment to the company due to the “reputation avalanche” e ﬀ ect inversely, thus leading to more negative economic consequences, and is not conducive to the sustainable development of the company.


Introduction
Companies have different attitudes towards corporate social responsibility(CSR) . To maximize shareholder value, most companies neglect their own social responsibility and pay attention to their own economic benefits [4,9,12,14,19,21,23,25,26], which leads to severe environmental pollution problems [21,27], some listed companies even engage in illegal behaviors such as false reports [6,13][, financial fraud and share price control in order to gain high profits, such as "Sanlu milk powder" and "Changchun vaccine scandal", which have caused great harm to consumers and investors. On the other hand, some companies can take social responsibility into account while pursuing profit, such as Baixiang Food Company, which is widely recognized by the society as having been engaged in social welfare for a long time.
Existing studies have shown that corporate social responsibility (CSR) contributes to improve the relationship with consumers, investors and other stakeholders [1, 3,21,22,25,28,29], which supports the view that corporate social responsibility (CSR) activities are committed to enhance shareholder value by benefiting stakeholders [2,4,9,14,17,18,23,24,30,31]. Consequently, researchers have found that the corporate social responsibility can establish a good corporate reputation [16,25,32], cause positive market reaction [19,21], reducing parameter uncertainty in investors' expected return evaluation [6,10,15]. However, little attention has been paid to whether the listed company's social responsibility will impact investors and regulators after corporate frauds. To fill in this gap, this paper explores whether the listed company's social responsibility activities really make impact investors and regulators after corporate frauds. Specifically, our research question is Whether socially responsible investing has the "compensation" effect or the "reputation collapse" effect on the market reaction.
By using regression, our results show that the announcement of violation enforcement will produce significantly negative market reaction, and the more serious the type of violation enforcement, the greater the negative market reaction. Further, we find that the corporate social responsibility (CSR) performance will not directly affect the market reaction of the announcement of punishment for violations, but it will increase the severity of the company's punishment for violations, thus indirectly bringing more negative market reaction.
The contribution of this paper is mainly reflected in the following aspects. Firstly, previous studies mostly focused on the reasons for companies to make socially responsible investment and their economic consequence [2-5, 7, 9, 10, 12-14, 17, 18, 20, 22-24, 31-34], while this study combine corporate social responsibility and market reaction of illegal activity for research. Secondly, it provides a further understanding of CSR from the perspective of regulators and supplements this part of the literature. Besides, it has a certain guiding significance for improving the design of the employee stock ownership plan system and promoting the sustainable development of enterprises.

Institutional Background
Corporate social responsibility (CSR) means that in addition to creating profits and maximizing shareholders' wealth, corporate also need to undertake further responsibilities to other stakeholders [1, 24,30,33], which is the opposite of what the illegal company are doing. As early as the 19th century, some scholars argue that corporate should undertake social responsibility, as time goes by, all social circles pay more attention to the social responsibility performance, and listed companies are no exception [7,13,18,20]. By improving their social responsibility performance, companies can win more investment and consumers' support [1, 14,16,30], and thus achieve their own sustainable development.
China Fourteenth Five-Year Plan highlight the importance of social responsibility. Through the formulation and implementation of relevant laws, regulations and rules, the Chinese government calls on the companies to actively undertake social responsibilities. The Company Law promulgated and implemented in 2006 also indicates that government departments and regulatory agencies have begun to attach importance to the corporate social responsibility, in which the article 5 clearly stipulates that listed companies must undertake corresponding "social responsibility" when making decisions and operations.
In the past 20 years, more and more companies have been willing to disclose CSR reports, and China Securities Regulatory Commission published the penalty announcement information, providing a unique research database with Chinese characteristics for studying CSR from the perspective of regulators and supplements after corporate frauds.

Hypothesis Development
Existing studies have shown that listed companies' violations will lead in the increase of the stock price volatility, rising financing costs and decline in market value through investors' reduced trust in the company's disclosure of information and damage to market reputation [28,[34][35][36][37][38][39]. The punishment imposed by regulators on listed companies for violations includes criticism, warning, condemnation, fine, confiscation of illegal income, market ban, etc. Different types of penalties represent the definition of the severity of listed companies' violations by regulators, and also convey different penalty information to the market based on signaling theory. Through the above analysis, we put forward the following assumption: Hypothesis 1. The announcement of violation enforcement will bring negative impact on the stock price, and the market reaction is related to the type of violation enforcement.
Existing research researchers that the corporate social responsibility can establish a good corporate reputation [1, 5,25,32], cause positive market reaction, reducing parameter uncertainty in investors' expected return evaluation. Based on the previous theoretical analysis, we argue that the CSR performance will have two opposite effects on the market reaction after corporate frauds.
On the one hand, according to the social identity theory and stakeholder theory [1,12], stakeholder groups are tolerant of listed companies' violations due to "value identity" psychology and compensation effect [4,14], for which investors will tolerate corporate infractions, leading to mitigate the negative market reaction to violations. On the other hand, according to the expectancy violation theory, listed companies are engaged in illegal activities after having established a good reputation in stakeholders such as investors, government departments and the public through socially responsible investment, then the stakeholder groups consider the company as "deliberately break the law" or "serious violations" [1, 2,22,25]. The gap results in the "reputation collapse" effect, which exacerbates the negative market reaction to violations. Through the above analysis, we put forward the following two opposite assumptions: Hypothesis 2a. Based on compensation effect, the corporate social responsibility (CSR) performance will mitigate the market reaction of the announcement of punishment for violations.
Hypothesis 2b. Based on "reputation collapse" effect, the corporate social responsibility (CSR) performance will exacerbate the market reaction of the announcement of punishment for violations.
The punishment imposed by regulators on listed companies for violations includes criticism, warning, condemnation, fine, confiscation of illegal income, market ban. Different types of penalties represent the definition of the severity of listed companies' violations by regulators. On the one hand, regulators will also be tolerant of listed companies' violations due to "value identity" psychology, and mitigate punishment; on the other hand, according to the expectancy violation theory, regulators will also consider the company as "deliberately break the law" or "serious violations", resulting in the "reputation collapse" effect, which lead to aggravation of penalty. Through the above analysis, we put forward the following two opposite assumptions: Hypothesis 3a. Based on compensation effect, the corporate social responsibility (CSR) performance will mitigate the severity of the punishment; Hypothesis 3b. Based on "reputation collapse" effect, the corporate social responsibility (CSR) performance will aggravate the severity of the punishment.

Data
We focus on the Chinese listed firms that have frauds throughout 2010-2020. Our main source of data on violation is the China Stock Market & Accounting Research Database (CSMAR). while the financial data and stock return data are also from the CSMAR. Moreover, social responsibility scores data is from Hexun Social Responsibility Rating Database. We first exclude the observations that failed to pass the general meeting of shareholders or stopped to implement. Then we delete observations in which the target firms have missing financial data and missing key variables. In addition, due to the uniqueness of financial and insurance companies in their industries, financial and insurance listed companies are also excluded from the sample. Then the sample with an estimated window period of less than 100 days is eliminated for the reason that we calculate excess return by Market Model Method. We winsorize the first and last 1% of all continuous variables to control for the influence of outliers. The above sampling procedure produces a total of 1578 observations, table 1 provides the summary statistics for the study variables. We found that the mean of cumulative excess return is negative, and the negative impact of violation announcements on market reaction is obtained preliminarily based on the results. The gap between the minimum and maximum values of the variable social responsibility score is huge, and the standard deviation reaches 15.40. It can be seen that the social responsibility performance of different companies varies greatly.

Variables
Using event study methodology, this paper calculates the cumulative abnormal return (CAR) near the event window to measure the market reaction. Specifically, we select the violation enforcement announcement date as the event point, that is, the violation enforcement announcement date is set to day 0, by calculating the cumulative abnormal return to study the market reaction after the penalty announcement date. In the robust test, we also use the cumulative abnormal return rate from different time interval.
Regulators will issue different penalties according to the severity of the company violations, so we use PT to measure severity of listed companies' violations by regulators, and the specific setting of PT is shown in the table 2.

Research Design
We introduce the following OLS regression model (1) to test the relationship between corporate social responsibility performance and the market reaction of the announcement of punishment for violations.
Following prior literature, we also control for ROA, Size, Turn and Lev. All measures are constructed at the end of year t-1.

Market Reaction to Violations
We examine the impact of violations on market reaction first. To verify impact of violations on market reaction, we mainly use the method of calculating the CAR mean of the overall violation sample and performing the T test. Then, we verify the relationship between the market reaction and the type of punishment by comparing the market reaction results of different types of punishment. Table 3 shows the CAR mean and the results of T test results of CAR mean in different time intervals, which shows that the average CAR of different intervals is significantly negative at the 1% confidence interval, that is, the negative impact of the violation announcement on the stock price of the listed company is very significant, confirming that listed companies' violations will lead in the decline in stock prices through investors' reduced trust in the company's disclosure of information, and this also proves the first half of Hypothesis1. Table 4 shows the CAR mean of different type of punishment and their T test results, columns 1 to columns 5 show the test results for each of the five penalty types. In column (3) Significance at the 10%, 5%, and 1% levels is indicated by *, **, and ***, respectively.  Note: This table presents the mean CAR of different type of punishment and their T test results. Significance at the 10%, 5%, and 1% levels is indicated by *, **, and ***, respectively.
to column (5), the CAR means for 'condemnation', 'fine' and 'confiscation of illegal income' these three types of penalties are significantly negative in most intervals, while the mean t test results for 'criticism' and 'warning' are not significant.
In summary, among the punishment methods, the punishment effect of criticism and warning measures is weak, which means that investors in the market respond very weakly to these two punishment methods, while the methods of 'condemnation', 'fine' and 'confiscation of illegal income' are relatively effective. In particular, the punishment methods of 'condemnation' and 'confiscation of illegal income' have a strong market reaction, which also proves the second half of the Hypothesis1 that the market reaction is related to the type of violation enforcement.
The above findings, taken together, indicate that the announcement of violation enforcement will bring negative impact on the stock price, and the market reaction is related to the type of violation enforcement, which supports Hypothesis1. Note: This table presents the results of correlation coefficient matrix. Significance at the 10%, 5%, and 1% levels is indicated by *, **, and ***, respectively. Columns 1 to 4 show the regression results for whole sample group, criticism sample group, condemnation sample group and fine sample group. Significance at the 10%, 5%, and 1% levels is indicated by *, **, and ***, respectively.

Impact on Market Reaction of CSR
After proving the existence of market reaction to violations, we further explore the relationship between CSR and market reaction to violations. Table 5 shows the results of correlation coefficient matrix. We can see that there is no direct and significant correlation between the CAR of listed companies and the CSR performance, that is, the CSR of listed companies cannot directly affect the company's stock price, and cannot directly affect the market reaction when the company violates the rules. Significance at the 10%, 5%, and 1% levels is indicated by *, **, and ***, respectively.
From the regression results of table 6, it can be seen that whether it is full sample regression or group regression, there is no direct and significant correlation between the CAR and the CSR performance, that is, the CSR does not directly affect the company's stock price, and can't directly affect the company's market reaction when the company violates the rules.
In summary, there is no significant correlation between CSR and the market reaction to violations, which may be the result of the offset of 'compensation effect' and 'reputation collapse' in the Hypothesis2, that is, Hypothesis2 does not hold.

Impact on Types of Penalties of CSR
To test Hypothesis3, we introduce a new dummy variable SP to describe the severity of penalties imposed on the offending company. Specifically, SP=0 when the type of penalty is criticism or warning, and SP=1 when the type of penalty is condemnation, fine and confiscation of illegal income. Then, replace CAR in model (1) with SP to get Model (2), and Model (2) is as follows.
First of all, group according to the severity of penalties, and conduct the mean difference test. The results of the differences test between groups are shown in table 7. Table 7 shows the results of mean difference test by group according to the severity of penalties. It can be seen from the results that the ROA of the two groups is significantly different at the 5% confidence interval, and the CSR performance of the two groups is significantly different at the 1% confidence interval, that is, the CSR performance of the second group with more severity of penalties is higher, which also preliminarily confirms the Hypothesis3 that there is a relationship between social responsibility performance and the severity of penalties. Table 8 presents the results of correlation coefficient matrix. As can be seen from the results, there is a significantly positive correlation between corporate social responsibility performance and the severity of violation penalties at the 10% confidence interval, indicating that good CSR performance will bring more serious penalties when a company violates the rules. This is consistent with the "reputation collapse" effect in Hypothesis3b. To further test this relationship, we tested by logistic regression based on model (2), and the regression results are as follows.
From the regression results of table 9, it can be seen that CSR is significantly positive correlated with the severity of violation punishment at a 1% confidence interval, that is, reporting that the severity of penalties of listed companies by regulators has a significantly positive relationship with its social responsibility performance CSR. Note: This table presents the results of correlation coefficient matrix. Significance at the 10%, 5%, and 1% levels is indicated by *, **, and ***, respectively.  (2). Significance at the 10%, 5%, and 1% levels is indicated by *, **, and ***, respectively.
According to social responsibility theory and organizational identification theory, a company's good social responsibility performance will affect the company's position in the eyes of stakeholders, investors, government and the public, and so that establish a good reputation.
However, when the company who has great CSR performance violates the law, according to the expectancy violation theory, regulators will also consider the company as "deliberately break the law" or "serious violations", resulting in the "reputation collapse" effect, which lead to aggravation of penalty.

Conclusion
Based on the data of China's Shanghai and Shenzhen A-share listed companies from 2010-2020, we provide that the announcement of violation enforcement will produce significantly negative market reaction, and the more serious the type of violation enforcement, the greater the negative market reaction. Further, we find that the corporate social responsibility (CSR) performance will not directly affect the market reaction of the announcement of punishment for violations, but it will increase the severity of the company's punishment for violations, thus indirectly bringing more negative market reaction. Overall, our findings suggest that the listed company's social responsibility (CSR) performance before violation cannot play a "compensation" effect to offset the negative impact caused by its violation, it may bring more serious administrative punishment to the company due to the "reputation avalanche" effect inversely. Our findings have implication in practice as follows.
First, the violation penalty announcement has a significant negative impact on the stock price of listed companies, and the impact is related to the type of violation penalty. This shows that the violation penalty announcement can convey a more obvious signal to the capital market, and has a strong deterrent effect on the violation company, which indicates that the operation of China's securities market is more and more standardized and healthy, and it is of great significance to the sustainable development of the company.
Second, the social responsibility performance of listed companies is related to the severity of penalties for corporate violations. From the analysis, we find that the corporate social responsibility (CSR) does not directly affect the market reaction when they violate the rules, but affects the market reaction by affecting the severity of the penalties imposed on listed companies when they violate the rules. The corporate social responsibility (CSR) will become the judgment basis for government departments, investors and the public to evaluate their reputation and status. However, when listed companies violate the rules, their good social responsibility performance will become a 'catalyst' for them to be more severely punished, which may be the result of the collapse effect caused by the performance gap before and after the company. For companies, the idea of using CSR to whitewash their own violations is not feasible. In order to achieve long-term sustainable development, the company should take into account CSR on the basis of law-abiding, so as to achieve the harmonious unity of interests and responsibilities.