Impacts of the China Securities Investor Services Center's Shareholding Exercise on Audit Fees

. Protecting the legitimate rights and interests of small and medium investors is the basis of the sustainable and healthy development of the capital market. The shareholding exercise pilot policy of China Securities Investor Services Center is a major innovation in the protection mechanism of investors' rights and interests. Taking A-share listed companies from 2013 to 2017 as the sample, this paper constructs a difference-in-difference model (DID) to test the impacts of Investor Services Center's shareholding exercise on audit fees of listed companies. The study finds that the Investor Services Center's shareholding exercise significantly reduces the audit fees. The mechanism test results show that Investor Services Center's shareholding exercise can reduce the audit fees by restraining the level of accrued earnings management and agency costs. Further analyses show that the effect of reducing audit fees by Investor Services Center's shareholding exercise is more obvious in larger enterprises with lower proportion of independent directors. This paper enriches the research related to the protection of small and medium-sized investors, clarifies the impact mechanism of investor protection on audit fees, expands the relevant research on the influencing factors of audit fees, and provides theoretical support and suggestions for the relevant policy formulation of the China Securities Investor Services Center.


Introduction
The protection of small and medium investors is an important issue facing China's capital market.  Guojian et al., 2021). As of September 2021, the number of individual investors in China's capital market has exceeded 190 million, accounting for more than 99% of all investors. The proportion of small and medium investors with market value of less than 500,000 yuan is 97%. Protecting the legitimate rights and interests of small and medium investors is an important task for the capital market supervision department and the basis for the healthy operation of the capital market. However, due to the disadvantages of small and medium shareholders in obtaining information and professional knowledge, the high cost of safeguarding rights and the weak awareness of selfprotection, how to effectively protect the interests of small and medium investors is always a challenging issue (Kong Dongmin et al., 2013;Wei Minghai, 2007). In recent years, the construction of the legal system for investor protection has been gradually improved. The public implementation mechanism is mainly based on the administrative supervision of the China Securities Regulatory Commission (CSRC). At the same time, measures such as online voting, proxy voting, independent directors and cumulative voting system have been introduced to regulate the abuse of control power and strengthen the protection of the rights and interests of minority shareholders (Chen Yunsen et  become the focus of theoretical and practical circles (Li Chen et al., 2021). The Investor Services Center's shareholding exercise affects the audit fees. As an innovative "semi-public-semi-private" securities regulatory implementation mechanism, the Investor Services Center's shareholding exercise can regulate the information disclosure behavior of listed companies and improve their information disclosure quality (Chen Yunsen et al., 2021; Xiong Jiacai and Tong Daming, 2022), thus lowering the auditor's assessment of the enterprise risk. At the same time, as a regulatory minority shareholder, the Investor Services Center can restrain the large shareholder from encroaching on the interests of the small and medium investors, improve the level of corporate governance, reduce the level of earnings management of the enterprise, and further reduce the auditor's assessment of the enterprise risk. Therefore, the Investor Services Center's shareholding exercise is beneficial to reduce the audit fees of the enterprise. Hence, this paper studies the investor protection function of the Investor Services Center from the perspective of audit fees.
Since the Investor Services Center began to carry out the pilot exercise in Shanghai, Guangdong (excluding Shenzhen) and Hunan from 2016, this paper selects the listed companies in the above three pilot areas as the treatment group, and selects the listed companies in Zhejiang, Shenzhen and Hubei as the control group taking geographical and economic factors into consideration. This paper establishes a DID model to study the impacts of the Investor Services Center's shareholding exercise on audit fees. The results show that the Investor Services Center's shareholding exercise can significantly reduce the audit fees of the enterprise. The mechanism test results show that the Investor Services Center's shareholding exercise can reduce the audit fees by reducing the accrued earnings management and agency costs. In addition, the impacts of the Investor Services Center's shareholding exercise on audit fees are more significant in listed companies with larger size and a smaller proportion of independent directors.
The contributions of this paper are mainly reflected in the following three aspects: First, this paper uses the DID method to test the impacts of the Investor Services Center's shareholding exercise on audit fees, which enriches the research on the economic consequences of the existing investor protection system. The Investor Services Center is an innovative market supervision mechanism with Chinese characteristics. The existing literature mainly analyzes its nature and impacts from the theoretical level together with case studies, and seldom uses empirical methods to study the economic consequences of the Investor Services Center's shareholding exercise. This study finds that the Investor Services Center's shareholding exercise can reduce the audit fees by reducing the company's earnings management level and agency costs, which provides empirical evidence for the effectiveness of the Investor Services Center's shareholding exercise and improves the relevant research on the protection of small and medium investors in China. Secondly, this paper expands the literature on the influencing factors of audit fees. The existing literature mainly focuses on the impacts of corporate characteristics and governance structure, auditor and firm characteristics and external environment on audit fees. Specifically, corporate characteristics and governance structure include executive characteristics (Cai Chun et  However, little literature explores the impacts on audit fees from the perspective of minority investor protection. This paper regards the Investor Services Center as an external environmental factor, thus supplementing the existing literature on the influencing factors of audit fees. Finally, this paper has certain policy implications and practical significance. As a regulatory minority shareholder, the Investor Services Center demonstrates the advantages of combining government functions with market participation (Chen Yunsen et al., 2021), which is of great importance to the protection of the rights and interests of small and medium investors and the orderly operation of the capital market. The conclusions of this paper prove the effectiveness of the Investor Services Center's shareholding exercise policy at the empirical level, test the effectiveness of the transformation of China's capital market supervision model to flexible governance, and provide empirical evidence for formulating and improving the future policy planning of the Investor Services Center. In addition, it also has reference significance for the supervision of other similar emerging development markets and their investor protection.
The following chapters of this paper are arranged as follows: the second chapter puts forward the theoretical analysis and research hypothesis of this paper, the third chapter is the research design of this paper, the fourth chapter is the analysis of empirical results, the fifth chapter is further analysis, and the sixth chapter is conclusion and prospect.

Theoretical Analysis and Research Hypothesis
According to the audit risk theory proposed by American Institute of Certified Public Accountants, audit risk consists of inherent risk, control risk and inspection risk. When the audit risk is certain, the higher the inherent risk and control risk of the enterprise, the more time and energy the auditor needs to spend in making rigorous assessment of the inherent risk and control risk of the company in order to reduce the inspection risk and the legal risk and reputation risk that may be faced with litigation as well as ensuring the audit quality. The audit intensity is increased and the audit fees charged by the firm is proportional to the auditor's audit intensity and audit cost, so the audit fees is higher.
For a long time, because the relevant laws and regulations on investor protection in China are not yet perfect, and the supervision of large shareholders and the protection of small and medium investors are relatively weak, many large shareholders encroach on the interests of small shareholders in listed companies , and the exercise of small and medium investors can effectively reduce the agency problem of enterprises, inhibit the tunneling behavior of controlling shareholders, and thus reducing the operating risk of enterprises (Liu Xinming et al., 2021). The Investor Services Center's shareholding exercise can represent the demands of small and medium investors, thus safeguarding their legitimate rights and interests, promoting the listed companies to improve their governance, standardize their operations and maintain the order of the securities market.
Specifically, firstly, the Investor Services Center's shareholding exercise can effectively improve the quality of information disclosure of listed companies, thus reducing the audit fees of enterprises (Xiong Jiacai, Tong Daming, 2022). As a "regulatory minority shareholder", the Investor Services Center can reasonably exercise the rights of proposal, voting rights, litigation rights and other rights, strengthen the external supervision of the enterprise, safeguard the interests of minority shareholders, so as to regulate the information disclosure behavior of listed companies and improve the integrity and accuracy of the information disclosure of listed companies. Meanwhile, the Investor Services Center, as an innovative government supervision model, strengthens the protection of the interests of small and medium investors, thus limiting the encroachment of large shareholders, reducing the earnings management behavior of enterprises, and further improving the quality of information disclosure of the company. Higher quality of information disclosure can reduce the inherent risks of the company, enable the auditor to reduce the assessment of corporate risks, reduce the workload of the auditor, and thus reducing the audit fees.
Secondly, the Investor Services Center's shareholding exercise can improve the level of corporate governance and reduce the audit fees of the enterprise. The existence of the Investor Services Center improves the small and medium investors' right of speech, and the Investor Services Center's shareholding exercise helps more small and medium investors to actively exercise their rights and improve the corporate governance capability. The Investor Services Center's shareholding exercise will not only make the enterprise subject to regulatory penalties, but also the media's extensive coverage and publicity will further expand its influence on public opinion, thus playing a strong external supervisory role on the senior executives and major shareholders of the enterprise (Liu Xinming et al., 2021), urging the management of the enterprise to operate prudently, curbing excessive speculation by major shareholders and improving corporate governance. In addition, the Investor Services Center is not only staffed with professionals, but is also able to cooperate with the regulatory authorities in depth and complement each other with the judicial system, as well as play a BCP Business & Management

FEMS 2022
Volume 19 (2022) 166 linkage role with the internal governance mechanism of listed companies to make up for the deficiencies of the current internal corporate governance and administrative supervision. Effective corporate governance can reduce earnings management and financial manipulation of listed companies, reduce the risk of material misstatement in the financial statements, reduce the inherent risk or control risk of companies, and thus reduce the audit risk and ultimately reduce audit fees.
Finally, the Investor Services Center's shareholding exercise can improve the quality of internal control, resulting in lower audit fees. The Investor Services Center represents a large number of small and medium investors to exercise their rights in a centralized way. It encourages them to exercise their rights actively and protect their rights in accordance with the law. It can balance the influence of the controlling shareholders in the board of directors, thereby constraining the management and the controlling shareholders from encroaching on the rights and interests of minority shareholders in pursuit of their personal interests, as well as improving the effectiveness of internal control in enterprises (Sun Guangguo and Li Binghui, 2021). Effective internal control can improve the authenticity and reliability of accounting information and can become an important institutional guarantee to improve the quality of accounting information (Zheng Chun et al., 2021). At the same time, effective internal control can reduce the opportunism of the management and controlling shareholders, reduce agency costs, ensure the effectiveness and timeliness of internal and external information transmission, reduce information asymmetry, improve earnings quality, enable auditors to narrow the scope of substantive testing, thus reducing audit fees. Based on this, the following research hypothesis is proposed.
Hypothesis 1a: The Investor Services Center's shareholding exercise will reduce the company's audit fees.
However, some studies have found that small and medium investors' participation in business decisions is limited, which may have a negative impact on corporate governance. There may be opportunistic behaviors in their exercise process, which neglect the long-term development of enterprises for self-interest (Belloc, 2013). In the meantime, small and medium investors have limitations in obtaining, identifying and judging corporate information and professional skills. Excessive participation in corporate decisions may lead to ineffective decisions (Xin Yu, 2020). Zheng Guojian et al. (2016) took "Konka Group Co., Ltd" as an example to analyze the vulnerability of the alliance of small and medium-sized investors, and concluded that small and medium shareholders have limitations in corporate governance, which may cause confusion in corporate governance and operation management. Moreover, higher minority shareholders' participation will put pressure on corporate decision-making, increase corporate pandering behavior, and tend to manipulate earnings (Kong Dongming and Liu Shasha, 2017). Therefore, the exercise of small and medium investors may lower the level of corporate governance, increase the level of corporate earnings management and increase corporate operating risks, which will increase audit risk and audit investment, thus increasing audit fees. Based on this, the following research hypothesis is proposed.
Hypothesis 1b: The Investor Services Center's shareholding exercise will increase the company's audit fees.

Sample selection and data sources
Based on the initial samples of listed companies in Shanghai and Shenzhen A shares from 2013 to 2017, this paper excludes financial companies, ST companies and samples missing major variables, and finally obtains 4855 sample observations. The financial data used in this paper are all from CSMAR database. In order to eliminate the influence of outliers on the regression results, this paper performs Winsorize processing of ±1% on continuous variables, and uses Stata 16.0 for statistical analysis.

Model setting and variable definition
In order to study the impacts of The Investor Services Center's shareholding exercise on audit fees, this paper uses the following DID model to carry out tests. When the listed company is in the pilot provinces and cities of the Investor Services Center (Guangdong excluding Shenzhen, Shanghai and Hunan), the value is 1, and the value is 0 for enterprises located in Zhejiang, Shenzhen and Hubei.

Post
Dummy variable for the time of policy implementation After the Investor Services Center's shareholding exercise pilot policy is implemented, the value is 1. Before the policy is implemented, the value is 0.

Size
Company size Natural logarithm of the total assets of the listed company at the end of the year.

Lev
Asset-liability ratio Total debt divided by total assets.

PPE
Fixed assets ratio Ratio of fixed assets to total assets at the end of the period.

Cash flow from operations
Net cash flows from operations divided by total assets.

ROA
Rate of return on total assets Net profit divided by total assets.

Loss
Loss status An indicator that equals one if a firm's net income is less than zero, and zero otherwise.

Grow
Corporate growth The increase in operating income at the end of this year divided by the total operating income at the end of last year.

Soe
State-owned enterprise An indicator that equals one if the company is stateowned property, and zero otherwise.

Proportion of independent directors
Percentage of independent directors on the board.

BDsize
Board size Natural logarithm of the number of board members.

Dual
The combination of two duties An indicator that equals one if the duties of chairman and general manager are combined, and zero otherwise.

Top1
Equity structure The largest shareholder's shareholding ratio.

Big4
Audit quality An indicator that equals one if a firm is audited by a Big 4 auditor, and zero otherwise.
In this model, Auditfee is the enterprise audit fees, which is measured by the natural logarithm of the current audit fees of the listed company (Hu Guoqiang et al., 2020). The variable Treat indicates whether or not it is a sample of the experimental group. In 2016, the Investor Services Center first conducted a pilot project in Shanghai, Guangdong (excluding Shenzhen) and Hunan. In this paper, listed companies from three regions with similar economic and geographical conditions,

Analysis of empirical results
In order to test the impacts of the Investor Services Center's shareholding exercise on audit fees, model (1) is used for empirical test. The regression results of the impacts of the Investor Services Center's shareholding exercise on audit fees of listed companies are shown in Table 3. Columns (1) and (2) of Table 3 show the regression results of the model without or with control variables, respectively. The results in column (1) show that, with absence of control variables, the coefficient of Treat × Post is -0.080, which is significantly negative at 5%. After adding the control variables, the empirical result is shown in column (2) of Center's shareholding exercise can effectively reduce the audit fees of listed companies, which verifies the hypothesis 1a in this paper. At the same time, the regression results of control variables show that company size (Size), asset-liability ratio (Lev), cash flow from operating activities (CFO), Loss status of the company (Loss) and Big4 are significantly positively correlated with audit fees, while fixed assets ratio (PPE), rate of return on total assets (ROA), nature of property rights (Soe), the combination of two duties(Dual), the largest shareholder shareholding ratio (Top1) are significantly negatively correlated with audit fees. Note: ***, **, * are significant at the statistical level of 1%, 5% and 10% respectively, with t value in brackets, the following tables will follow the above illustration.

Adjust the sample time window
The stock market dropped dramatically in 2015, but the year 2015 was not excluded from the above regression analysis. Therefore, in order to avoid the impacts of this event on audit fees, in the robustness test, the year 2015 sample is removed and Post is redefined. When the year is 2013 or 2014, the value of Post is 0; when the year is 2016 or 2017, the value of Post is 1. In this case, the model (1) is regressed again, and columns (1) and (2) of Table 4 show the regression results of the model without and with control variables, respectively. The results in column (1) show that, with absence of control variables, the coefficient of Treat × Post is -0.092, which is significantly negative at 5%. After adding the control variables, the empirical result is shown in column (2) of Table 4, with a coefficient of Treat × Post at -0.067, which is significantly negative at 1%. At this point, it can be concluded that the audit fees of listed companies in the pilot area decrease significantly at 1% level after the Investor Services Center's shareholding exercise pilot policy is implemented. The conclusion of hypothesis 1a in this paper is robust.
As the Investor Services Center's shareholding exercise was extended from the pilot area to the whole country in May 2017, the situation of the control group changed, which may lead to inaccurate results. Therefore, this paper excludes the sample of 2017, and only uses the sample of 2016 to represent the implementation of the Investor Services Center's shareholding exercise and redefine Post. When the year is 2013, 2014 or 2015, Post equals zero; when the current year is 2016, Post equals one. In this way, the model (1) is regressed again, and columns (3) and (4) of Table 4 show the regression results of the model controlling for year fixed effect only and for both industry and year fixed effects, respectively. The results in column (3) show that without controlling for industry fixed effect, the coefficient of Treat × Post is -0.028, which is significantly negative at the 5% level. After controlling for both industry and year fixed effects, the empirical result is shown in column (4) of Table 4, with a coefficient of Treat × Post at -0.053, which is significantly negative at the 10% level. At this point, it can be concluded that the audit fees of listed companies in the pilot area decrease significantly at the 10% level after the Investor Services Center's shareholding exercise pilot policy is implemented. The conclusion of Hypothesis 1a in this paper is robust.

Time-varying DID
In 2017, the Investor Services Center's shareholding exercise policy implemented by the China Securities Investors Services Center was extended from some pilot areas to the whole country. Therefore, for different regions in China, the implementation time of the Investor Services Center policy is different. Under such circumstances, the time-varying DID model can better deal with different time points. Hence, this paper constructs a time-varying DID model.
The model is selected for a sample period of 2013-2020. We redefine the Post variable so that when the firm is located in Shanghai, Guangdong (except Shenzhen), and Hunan: Post equals 0 when the time is before 2016, and Post equals 1 for 2016 and subsequent years; when the firm is located in Zhejiang Province, Shenzhen, and Hubei Province: Post equals 0 when the time is before 2017, and Post equals 1 for 2017 and subsequent years. In this way, the model (2) is regressed, and columns (5) and (6) of Table 4 show the regression results of the model without and with control variables, respectively. The results in column (5) show that, with absence of control variables, the coefficient of Treat × Post is -0.063, which is significantly negative at 1%. After adding the control variables, the empirical result is shown in column (6) of Table 4, with a coefficient of Treat × Post at -0.034, which is significantly negative at 1%. We can conclude that the audit fees of listed companies in the pilot areas have significantly decreased at 1% level after the Investor Services Center's shareholding exercise. The conclusion of hypothesis 1a in this paper is robust.

Parallel trend test
Being consistent with the parallel trend hypothesis test is an important premise of the DID model, that is, it is necessary to test that the audit fees of listed companies in the treatment group and the control group have the same variation trend before Investor Services Center's shareholding exercise. In order to test the parallel trend, we add the cross-product term of the dummy variable and the variable Treat at each time point in the regression. With reference to the research of Ke and Zhang (2021), the model is constructed as follows: Specifically, when the observed value is the data of enterprises in Shanghai, Guangdong (excluding Shenzhen) and Hunan in 2016, Current equals one; otherwise, it is equal to zero. When the observed value is the data of the second year before the implementation of the investment service center shareholding exercise policy for enterprises in Shanghai, Guangdong (except Shenzhen) and Hunan, Before2 is equal to one; otherwise, it equals zero. When the observed value is the data of the first year before the implementation of the investment service center shareholding exercise policy for enterprises in Shanghai, Guangdong (except Shenzhen) and Hunan, Before1 is equal to one ; otherwise, it equals zero. When the observed value is the data of the first year after the implementation of the policy in Shanghai, Guangdong (excluding Shenzhen) and Hunan, After1 is equal to one; otherwise, it equals zero. In order to avoid the problem of complete collinearity, the third year before the implementation of the policy ( Before3 ) is taken as the benchmark group. If the estimated coefficients α 1 , α 2 of Before2, Before1 are not significantly different from 0, it proves that there is no significant difference in audit fees of listed companies in the control group and the experimental group, and there is a parallel trend in the first and second years before the Investor Services Center's shareholding exercise.
The regression results are shown in Table 5. α 1 , α 2 are not significantly different from 0 and have no specific change rule. The parallel trend test results are shown in Figure 1. In Figure 1, the vertical axis represents the magnitude of the estimated coefficients of the impacts of the Investor Services Center's shareholding exercise on the audit fees at different time points, and the horizontal axis represents the relative time before and after the implementation of Investor Services Center's shareholding exercise policy. Current represents the initial period of policy implementation. This means that when the horizontal axis is 0, the dotted line above and below the solid point represents a 95% confidence interval. It can be seen from the figure that 95% of the confidence intervals of α 1 , α 2 all contain 0, which verifies the hypothesis of parallel trend and proves that the model in this paper is effective. The estimated coefficients α 3 , α 4 of Current, After1 are all significantly negative, indicating that the Investor Services Center's shareholding exercise will reduce the audit fees of listed companies, which further proves the conclusion of this paper.

Accrued earnings management
According to the analysis of audit risk model, the exercise of regulatory minority shareholders can reduce the inherent risk and control risk of audit faced by certified public accountants by reducing the level of corporate earnings management, and reduce time and effort spent on audits, thus reducing the audit costs of enterprises. Healy (1985) found a significant positive correlation between accrued earnings management and audit fees. In this paper, the discretionary accruals calculated by the modified Jones model (Dechow et al., 1995) are used to measure the accrued earnings management (DA), and the absolute value of DA is used to examine the overall degree of accrued earnings management (Wei Zhen, 2019). The regression results in column (1) of Table 5 show that the regression coefficient of the Investor Services Center's exercise (Treat×Post) corresponding to accrual earnings management is -0.009, which is significant at the statistical level of 10%, indicating that the Investor Services Center's shareholding exercise has a significant inhibitory effect on the accrual earnings management behavior of the company. It can be seen that the exercise of minority shareholders can indeed reduce the audit fees of the company by reducing the degree of accrued earnings management.   Conflicts of interest between large shareholders and small and medium shareholders will increase the second-type agency costs and the audit risk of auditors, resulting in an increase in audit fees, that is, the higher the second-type agency costs, the higher the audit fees. The exercise and supervision of the Investor Services Center will reduce the degree of information asymmetry between the large shareholders and the small and medium shareholders as well as the possibility of the large shareholders infringing the interests of the small and medium shareholders; thus, inhibiting the agency conflict between the large shareholders and the small and medium shareholders, and reducing the second type of agency costs. Therefore, the audit risk and costs are reduced, ultimately reducing the audit fees. Referring to the existing research (Ma Dongshan and Han Liangliang, 2018), this paper uses the ratio of other receivables to the total assets of the company to measure the second-type agency costs (SC). The regression results in column (2) of Table 5 show that the regression coefficient of the Investor Services Center's shareholding exercise (Treat×Post) to agency cost is -0.004, and it is significantly negative at the statistical level of 1%, which indicates that the Investor Services Center's shareholding exercise can significantly reduce the second-type agency costs of a company, thus further reducing the audit fees of a company. Companies with different sizes differ in many aspects, such as organizational structure and business volume. Therefore, the impacts on audit fees are different after the implementation of the Investor Services Center's shareholding exercise policy. In larger companies, the organization structure and business are usually more complex, and the business volume and the industry span are larger, which make the audit sampling size larger and the audit risk faced by auditors is higher (Jiang Yihong et al., 2004). Audit fees will increase with the size of the company (Zhang Shuhui and Luo Mengmao, 2016).

Impacts of company size
In order to study whether there is any difference in the effect of Investor Services Center's shareholding exercise in reducing audit fees under different firm sizes, consistent with the existing studies (Cai Chun et al., 2015;Chemmanur et al., 2014), this paper selects the natural logarithm of the total assets of listed companies at the end of the year to measure the size of the company, and divides the samples into large-size companies and small-size companies based on the annual median of all company sizes. The results are shown in Table 7. The coefficient of Treat×Post is -0.105 in the group with larger company size, which is significant at the statistical level of 1%; while the coefficient of Treat×Post is not significant in the group with smaller company size. The Fisher's Permutation test (Lian Yujun and Liao Junping, 2017) on the coefficient between the groups proves that there is a significant difference between the intergroup coefficients of the two groups, with the corresponding p-value of 0.1636. It shows that the effect of reducing audit fees by the Investor Services Center's shareholding exercise is more obvious in larger companies.

Impacts of the independent director's supervisory role
Independent directors play an important supervisory role in business decisions and play a positive role in the quality of accounting information (Zheng Chunmei et al., 2021;Cai Jifu, 2007). Due to the different degrees of supervision by independent directors of different enterprises, the effects of the Investor Services Center's shareholding exercise are different among different enterprises. When the supervision of the independent directors of the enterprise is stronger, the behavior of the majority shareholders to gain private interests by manipulating the operation decisions will be more strongly curbed, and the audit risk becomes relatively small. In this way, the Investor Services Center, as a regulatory minority shareholder, plays little role. In contrast, the enterprises with weak independent directors' supervision have weak internal supervision on the majority shareholders, less disincentive to opportunistic behavior of controlling shareholders, and the standardization and quality of information disclosure are lower. In this way, the Investor Services Center's shareholding exercise can play a linkage role in the internal corporate governance mechanism, thus reducing audit risk and achieving a more significant effect on reducing the audit fees of enterprises.
Therefore, in order to test whether there is any difference in audit fees reduction effect of the Investor Services Center's shareholding exercise in the enterprises with different independent directors' supervisory roles, this study measures the independent directors' supervisory efforts based on the ratio of the number of independent directors to the total number of the board of directors (Li Wengui and Yan Han, 2020). According to the annual median proportion of independent directors, the sample is divided into two groups for grouping regression. The test results are shown in Table 8. The results show that the coefficient of Treat×Post is -0.058 in enterprises with relatively low proportion of independent directors, which is significant at the statistical level of 5%; while the coefficient of Treat×Post is -0.046 in enterprises with relatively high proportion of independent directors, which is significant at the statistical level of 10%. It can be seen that the absolute value and significance of the regression coefficient of the sample with less supervision by independent directors are higher than those of the sample with greater supervision by independent directors. Through the Fisher's Permutation test (Lian Yujun and Liao Junping, 2017), it is proved that there is a significant difference between the intergroup coefficients of the two groups, with the corresponding p-value of 0.043, which further indicates that in the enterprises with weak independent director supervision, the Investor Services Center's shareholding exercise can more effectively reduce the operating risk of the company, thus helping to reduce the litigation risk that the auditor may face and the possibility of significant errors in the financial statements. In this way, it significantly reduces the audit risk and audit cost, ultimately leading to lower audit fees.

Conclusion and Prospects
Due to the shareholdings of listed companies in China are relatively concentrated and the legal system related to the protection of small and medium sized investors is not yet sound, the problem of large shareholders encroaching on the interests of small and medium shareholders has long existed. As an important innovation in the small and medium investors' protection system of the regulatory authorities in China, the China Securities Investors Service Center can help small and medium investors to exercise their rights actively, thus effectively reducing business risks. As an important operating expense of an enterprise, audit fees can reflect the effect of the Investors Service Center's exercise. Therefore, this paper constructs a DID model to explore the impacts of Investors Service Center's shareholding exercise on audit fees. The study finds that after the Investors Service Center's shareholding exercise, audit fees of the enterprise can be significantly reduced, and the reliability of the conclusion is verified by adjusting the sample time window and the time-varying DID model. In the meantime, this paper finds that the Investors Service Center's shareholding exercise can play its role by reducing the accrued earnings management and agency costs, thus reducing the audit fees. In addition, further analyses finds that when the size of enterprise is larger and the proportion of independent directors is smaller, the effectiveness of Investors Service Center's shareholding exercise in reducing audit fees is more significant.
The empirical research in this paper verifies the effectiveness of the Investors Service Center's shareholding exercise on investor protection, and provides a new perspective for the existing research on the protection of small and medium investors. In this research, the following policy implications can be drawn: Investors Service Center can cooperate with other external supervision institutions and strengthen the cooperation with exchanges, auditors, institutional investors and independent directors, etc. so as to form a synergy for investor rights protection, thus strengthening its own supervisory function, highlighting the demonstration function of shareholding exercise, and making up for the deficiencies in the current internal governance and administrative supervision of listed companies in China in a gentle but powerful way. At the same time, small and medium investors should make full use of the Investors Service Center's functions to actively exercise their own rights, raise the awareness of safeguarding their rights according to the law, protect their own interests and improve the corporate governance environment. In addition, the managements of the enterprise should give full play to the supervisory role of independent directors, reduce the information asymmetry between large shareholders and small and medium shareholders, reduce the agency costs and accrual earnings management level of the enterprise, actively improve the level of corporate governance, and promote the sustainable and healthy development of China's capital market.