Frontiers in Business, Economics and Management ISSN: 2766-824X | Vol. 12, No. 3, 2023 279 Directors' and Officers' Liability Insurance and Audit Fees: Evidence from China Desheng Tang1 and Changting Huang1, * 1School t of Guangxi University, Guangxi, China *Corresponding author email: 1871982146@qq.com Abstract: This paper aims to deeply investigate the relationship between Directors' and Officers' Liability Insurance (hereinafter referred to as D&O insurance) and audit fees. Based on data from China's A-share listed companies from 2009 to 2022, this study empirically analyzes the specific impact of D&O insurance on audit fees. The results show that after purchasing D&O insurance, listed companies experience a significant increase in audit fees. This conclusion remains valid after robustness tests, confirming the positive role of D&O insurance in corporate governance. Further analysis of its mechanism reveals that the increase in audit fees stems from the demand for high audit quality driven by D&O insurance as a governance mechanism, reflecting its supervisory incentive effect rather than resulting from opportunistic and moral hazard behaviors. Additionally, the study finds a significant interactive effect between D&O insurance and audit fees, further validating the existence of its supervisory incentive effect. This paper provides strong evidence for the positive impact of D&O insurance on audit fees. These findings not only deepen our understanding of the governance effects of D&O insurance but also provide important theoretical support and empirical evidence for the future development of D&O insurance in China. Keywords: D&O insurance; Audit fees; Opportunism; Corporate governance. 1. Introduction D&O insurance is designed to provide protection for the management of a company against professional risks. When managers make mistakes in their work decisions that result in losses for the company and face shareholder lawsuits, this insurance can reduce or exempt their liability and financial compensation, thereby enhancing their confidence in risk decision-making. This text conforms to the academic English standard: Since the introduction of D&O insurance into China in 2002, the original intention of the capital market is to transfer the professional risks of directors and executives of listed companies to professional risk managers - insurers. This move not only hopes to improve the external governance level of listed companies and reduce the probability of agency problems[1], but also reduce corporate violations and financial restatements. Research also shows that D&O insurance helps to inhibit managers' short-term behavior, strengthen internal control of enterprises, and thus improve the quality of accounting information[2]. From this perspective, higher regulatory requirements require higher audit quality, which makes companies pay more audit fees. However, some scholars have pointed out that the "shielding" effect of D&O insurance may reduce the cost of violations for managers, weaken the litigation risk of shareholders, and thus induce opportunistic behavior of managers [3]. In addition, the insurance may also encourage managers' self-interested behavior and earnings management, increasing the litigation risk of the enterprise[4,5]. However, higher risks mean greater difficulty in audit work, which will inevitably increase audit fees. Based on the above analysis, it can be found that after the introduction of D&O insurance by listed companies, both the supervision and incentive effects and opportunistic effects may lead to an increase in audit fees for the company. Therefore, in-depth research on which effect leads to the increase in audit fees is of great theoretical and practical significance. Through the research of this article, it can further reveal the specific impact of D&O insurance on audit fees, provide theoretical basis and empirical support for companies to purchase D&O insurance, and also provide a useful reference for the promotion and application of domestic D&O insurance. 2. Mechanism Analysis and Research Hypothesis This text conforms to the norms of academic English: under the assumption of rational economic man, insurers will seek to maximize their own interests and prevent moral hazard by effectively supervising insured companies[6]. At the same time, the insured may also pursue their own interests to maximize their profits, which may lead to opportunistic behavior[7]. In view of this, this article explores the impact of D&O insurance on audit fees from two perspectives. The supervisory effect of D&O insurance helps to increase audit fees. Firstly, listed companies introduce an external supervision mechanism by purchasing D&O insurance. As the ultimate payor, the insurer collects relatively low underwriting fees but needs to bear the uncertainty of huge compensation. Therefore, rational insurers will take measures to reduce the possibility of moral hazard and opportunistic behavior to maximize their own interests. Secondly, after taking on the D&O insurance contract, the insurance company will require the enterprise to disclose more high- quality information and choose a more professional audit firm in order to avoid huge compensation and improve the enterprise's risk-taking ability. Therefore, the insurer expects the moral hazard and opportunistic behavior of the insured, and will strengthen supervision, thereby increasing the audit fees of listed companies. From the perspective of opportunism, the insured may pursue the maximization of their own interests. However, 280 such behavioral motives are suppressed due to various constraints such as shareholders, laws, and regulations, resulting in excessively high opportunism costs. Nevertheless, the "shielding" effect of D&O insurance weakens the constraints of shareholders and laws on managers [8], which may lead managers to engage in opportunistic behavior under the "shield" of D&O insurance. This can result in the company incurring excessive agency costs, such as perks of office and actively promoting mergers and acquisitions to obtain benefits, which may be deliberately concealed. Consequently, such behavior not only increases the risk and difficulty of material misstatements for the company but also harms the interests of shareholders and insurers. To reduce potential risks, audit firms may increase their investment to ensure audit quality. However, due to the prevalence of opportunistic behavior among company management, it may be difficult to improve audit quality[3]. In summary, after purchasing D&O insurance, rational insurers will take effective supervisory measures to reduce the probability of opportunistic behavior and company risks, as well as the possibility and complexity of material misstatements, in order to maximize their own interests. However, D&O insurance may also have an opportunistic "shielding" effect that can induce managerial opportunism. In such cases, managers may deliberately conceal financial information and manipulate earnings, leading to increased company risk. Therefore, if the introduction of D&O insurance in a listed company results in significant improvements in audit quality and company value, along with a significant reduction in agency costs, it suggests that D&O insurance mainly exerts a supervisory and incentive effect. Conversely, if there is no significant change or even a significant decrease in audit quality, a decrease in company value, and an increase in agency costs, it indicates that D&O insurance is more likely to exhibit an opportunistic effect. By observing changes in these indicators, we can determine whether D&O insurance primarily exerts a supervisory and incentive effect or an opportunistic effect in listed companies. Based on the above analysis, this paper proposes Hypothesis 1 and its opposing hypotheses 2a and 2b. Hypothesis 1: After purchasing D&O insurance, the audit fees of listed companies will increase. Hypothesis 2a: After a listed company purchases D&O insurance, the insurer will fulfill its supervisory obligations and impose stricter audit requirements, resulting in a decrease in agency costs and an increase in corporate value. Hypothesis 2b: After purchasing D&O insurance, listed companies may engage in opportunistic behavior, leading to a decline in corporate value and an increase in agency costs. 3. Data Sources and Research Design 3.1. Data Sources In 2009, the China Insurance Regulatory Commission issued the "Notice on Further Regulating the Order of the Insurance Market and Cleaning Up Illegal and Irregular Behaviors in the Insurance Intermediary Market," which required insurance companies to clean up and correct illegal and irregular behaviors involving shareholders, related parties, customers, employees, and other stakeholders. This measure provided a strong impetus for the development of D&O insurance. Based on this background, this paper selects annual data from 2009 to 2022 for listed companies on the Shanghai and Shenzhen A-share markets as the research sample, and screens them according to the following criteria: firstly, excluding samples of financial and insurance companies; secondly, excluding samples of companies with discontinuous or missing key indicators; thirdly, excluding samples of ST companies; and finally, excluding samples of companies with missing cross-sectional data for the current year. To ensure robustness of the results, this paper also applies a Winsorize treatment to continuous variables at the upper and lower 1% levels. The raw data used in this paper is sourced from the CSMAR (China Stock Market & Accounting Research) database, while the relevant data for D&O insurance is obtained from the China Research Data Service Platform (CNRDS). 3.2. Research Design First, in order to examine the impact of D&O insurance on audit fees, this article sets up the following multiple regression model for identification. 𝐴𝑒𝑑𝑖𝑑_𝐹𝑒𝑒 , 𝛽 𝛽 πΌπ‘›π‘ π‘’π‘Ÿπ‘Žπ‘›π‘π‘’ , βˆ‘ 𝜎 πΆπ‘œπ‘›π‘‘π‘Ÿπ‘œπ‘™ πœ€ (1) The dependent variable, Audit_Fee, represents the audit fees paid by listed companies to audit firms for both domestic and international audits each year, which is expressed as the natural logarithm of the sum. The independent variable, Insurance, indicates whether a company has purchased D&O insurance and is represented by a dummy variable. If a company has purchased the insurance, it is assigned a value of 1; if not, it is assigned a value of 0. The remaining variables are control variables, industry, and year, which are presented in Table 1. Secondly, in order to examine whether D&O insurance plays a supervisory role in the influencing process or triggers opportunistic behavior, this article sets up the following model for testing. πΌπ‘›π‘“π‘œπ‘Ÿπ‘šπ‘Žπ‘‘π‘–π‘œπ‘› , 𝛽 𝛽 πΌπ‘›π‘ π‘’π‘Ÿπ‘Žπ‘›π‘π‘’ , 𝛽 πΌπ‘›π‘ π‘’π‘Ÿπ‘Žπ‘›π‘π‘’ βˆ— 𝐴𝑒𝑑𝑖𝑑_𝐹𝑒𝑒 , 𝛽 𝐴𝑒𝑑𝑖𝑑_𝐹𝑒𝑒 , βˆ‘ 𝜎 πΆπ‘œπ‘›π‘‘π‘Ÿπ‘œπ‘™ , , πœ€ , (2) 𝐴𝑔𝑒𝑛𝑐𝑦_πΆπ‘œπ‘ π‘‘π‘  , 𝛽 𝛽 πΌπ‘›π‘ π‘’π‘Ÿπ‘Žπ‘›π‘π‘’ , 𝛽 πΌπ‘›π‘ π‘’π‘Ÿπ‘Žπ‘›π‘π‘’ βˆ— 𝐴𝑒𝑑𝑖𝑑_𝐹𝑒𝑒 , 𝛽 𝐴𝑒𝑑𝑖𝑑_𝐹𝑒𝑒 , βˆ‘ 𝜎 πΆπ‘œπ‘›π‘‘π‘Ÿπ‘œπ‘™ , , πœ€ , (3) Among them, the explained variables Information and Agency_Costs measure analyst attention and agency costs, respectively, while audit fees serve as a moderating effect. By introducing interaction terms, we can test the moderating effect of audit fees on agency costs or analyst attention, and thus detect the improvement in internal control indicators as D&O insurance is introduced and audit fees increase. 4. Empirical Analysis 4.1. Descriptive statistics Table 2 reports the descriptive statistics of the main variables. Through data comparison, it can be found that companies that have purchased D&O insurance pay significantly higher audit fees, while the difference between companies that have not purchased D&O insurance and the full sample is not significant, indicating that the introduction of D&O insurance has a real impact on audit fees. Secondly, 281 from the data point of view, listed companies that have introduced D&O insurance have chosen more top ten audit firms with higher audit technology, indicating that they have higher requirements for audit quality. Therefore, this article controls it to avoid bias in the results. Table 1. Variable Definitions Variable Variable Name Variable Code Definition Dependent Variable Audit Fees Audit_Fee The sum of domestic and international audit fees paid annually by listed companies to audit firms, expressed as the natural logarithm. Independent Variable D&O Insurance Insured Dummy variable. Takes the value of 1 if the listed company has purchased D&O insurance, and 0 otherwise. Control Variables Corporate Leverage Lev The ratio of total liabilities at the end of the year to total assets at the end of the year. Net Profit Margin ROE Return on equity. Company Size Size The natural logarithm of total assets at the end of the year. Company Growth Rate Growth The growth rate of total operating income. Nature of Property Rights SOE Dummy variable. Takes the value of 1 if it is a state-owned enterprise, and 0 otherwise. Proportion of Independent Directors Indep The proportion of independent directors to the total number of board members. Company Value TobinQ The ratio of enterprise value to its replacement cost. Duality Dual Dummy variable. Takes the value of 1 if the chairman and CEO positions are held by the same person, and 0 otherwise. Time Since IPO Age The total time since the company's initial public offering. Audit Firm Choice Big10 Dummy variable. Takes the value of 1 if it is one of the top 10 audit firms, and 0 otherwise. Audit Opinion Opinion Dummy variable. Takes the value of 1 if the previous year's opinion was a standard unqualified opinion, and 0 otherwise. Year Year Year dummy variables. Set according to different years. Industry Ind Industry dummy variables. Covers multiple industries in the research sample, with different industry dummy variables set accordingly. Table 2. Descriptive statistics Variable All Sample (N=34737) Insured=1 (N=3388) Insured=0 (N=31349) Mean SD Median Mean SD Median Mean SD Median Audit_Fee 13.67 0.665 13.58 14.19 0.880 14.05 13.61 0.616 13.53 Insured 0.000 0.000 0.000 1.000 0.000 1.000 0.098 0.297 0.000 Lev 0.418 0.202 0.410 0.499 0.199 0.514 0.426 0.203 0.420 ROE 0.067 0.124 0.073 0.059 0.146 0.071 0.066 0.126 0.073 Size 22.12 1.204 21.95 23.26 1.589 23.14 22.23 1.292 22.03 Growth 0.170 0.396 0.109 0.149 0.395 0.088 0.168 0.396 0.107 SOE 0.352 0.478 0.000 0.610 0.488 1.000 0.377 0.485 0.000 Indep 0.375 0.053 0.357 0.379 0.056 0.364 0.376 0.0530 0.364 TobinQ 2.069 1.296 1.653 1.786 1.242 1.386 2.041 1.294 1.625 Dual 0.287 0.452 0.000 0.174 0.380 0.000 0.276 0.447 0.000 Age 2.887 0.337 2.944 3.035 0.309 3.091 2.901 0.338 2.944 Big10 0.570 0.495 1.000 0.686 0.464 1.000 0.562 0.496 1.000 Opinion 0.976 0.152 1.000 0.975 0.156 1.000 0.976 0.152 1.000 4.2. Regression analysis Table 3 reports the multiple regression results of D&O insurance on audit fees. Column (1) shows the results of the single regression. Column (2) shows the multiple regression results after adding control variables. Column (3) is the result after adding control variables and controlling for year and industry. The regression coefficient of D&O insurance on audit fees is 0.147, showing a positive correlation at the 1% significance level. The results indicate that the introduction of D&O insurance by companies will lead to a significant increase in audit fees. Table 3. Multiple regression results Variable (1) (2) (3) Audit_Fee Audit_Fee Audit_Fee Insurance 0.524*** (34.82) 0.151*** (14.39) 0.147*** (12.53) N 34373 34373 34373 Year No No Yes Industry No No Yes Control No Yes Yes Adj_R2 0.047 0.576 0.607 282 4.3. Analysis of the impact mechanism of D&O insurance on audit fees The introduction of D&O insurance may have multiple impacts on audit fees. On the one hand, the introduction of D&O insurance may promote companies to adopt stricter external audit standards through the supervision and incentive effects of the insurer, resulting in an increase in audit fees. In this case, D&O insurance actually improves corporate governance and strengthens the supervision and restraint of directors and executives' behavior. On the other hand, the bottom-up role of D&O insurance may stimulate opportunistic behavior among directors and executives, which may lead audit firms to increase audit fees in order to avoid potential audit risks. In this case, D&O insurance may become an "umbrella" for opportunistic behavior, making some directors and executives more inclined to take risky decision-making behaviors. Therefore, the increase in audit fees may be due to the monitoring and incentive effects of D&O insurance, or it may be caused by opportunistic behavior. In order to more accurately determine the reasons for the increase in audit fees, further research and analysis of the specific circumstances of the company and the actual operation of D&O insurance are needed. At the same time, other factors that may affect audit fees, such as the company's governance structure and internal control quality, also need to be addressed. After the introduction of D&O insurance, if the insurance transfers the liability risk of directors and executives due to its "shield" function, thereby reducing their liability costs, this may induce opportunistic behavior or moral hazard among directors and executives. When the expected risk increases, in order to reduce the risk borne by itself, the audit firm may increase the input of audit work, resulting in the increase of audit fees. However, in the context of opportunistic behavior of managers, even if the auditor enhances audit work, the information transparency of the company may remain unchanged or even decline. In addition, opportunistic behavior of managers may also lead to the increase of agency costs and the decrease of company value. 4.3.1. D&O Insurance, Audit Fees, and Audit Quality From the perspective of audit quality, if the introduction of D&O insurance leads to an increase in audit fees for public companies through its supervisory effect, it generally implies that after the introduction of D&O insurance, indicators reflecting audit quality such as earnings management of public companies will decrease with the introduction of D&O insurance and the increase in audit fees. This is because the supervisory effect of D&O insurance helps to strengthen corporate governance and improve internal control, thereby enhancing audit quality. However, if the introduction of D&O insurance stimulates opportunistic motives among directors and executives, leading to the emergence of moral hazard behavior, the internal control quality and accounting information transparency of public companies may decline due to the presence of opportunistic behavior. This is because opportunistic behavior can lead to the weakening of corporate governance and the failure of internal control, which in turn has a negative impact on audit quality. Therefore, the impact of the introduction of D&O insurance on audit quality depends on whether it can exert a supervisory effect and whether it stimulates opportunistic behavior among directors and executives. Although audit fees are often seen as a direct reflection of audit quality, with higher audit fees generally interpreted as an improvement in audit quality, to more deeply explore the mechanism of D&O insurance from the perspective of audit quality, we have used the discretionary accruals calculated using the modified Jones model as a proxy for audit quality[9], because higher audit quality implies less tolerance for earnings management behavior. [10]. The results of (1), (2), and (3) in Table 4 report the regression results of of D&O insurance, audit fees, and audit quality. Both columns (1) and (2) are significant at the 1% significance level. Column (1) indicates that the introduction of D&O insurance can effectively improve the quality of internal control and reduce the occurrence of earnings management behavior. Column (2) shows that audit fees can also effectively reduce the occurrence of earnings management behavior. The results of column (3) reveal that the interaction term between D&O insurance and audit fees has a significant negative correlation with accruals, indicating that the company has exerted a sufficient supervisory effect after introducing D&O insurance. 4.3.2. D&O Insurance, Audit Fees, and Enterprise Value If the introduction of D&O insurance leads to an increase in agency costs and a decrease in enterprise value for public companies, it can be seen as evidence that D&O insurance has induced opportunistic behavior by transferring risk from directors and executives. In this case, D&O insurance may act as a "shield" for opportunistic behavior, making some directors and executives more inclined to take riskier decision-making actions, thereby increasing agency costs and reducing enterprise value. However, if there is a noticeable improvement in agency costs and enterprise value after the introduction of D&O insurance, it suggests that D&O insurance is exerting a supervisory and incentive effect. The supervisory and incentive effect of D&O insurance may enhance corporate governance by strengthening the monitoring and restraint of directors' and executives' behavior, thereby reducing agency costs and increasing enterprise value. Therefore, by observing the changes in agency costs and enterprise value after the introduction of D&O insurance, we can determine whether D&O insurance has exerted a supervisory and incentive effect or induced opportunistic behavior. This requires further research and analysis of the specific situation of the company and the actual operation of D&O insurance. To examine the impact of D&O insurance on enterprise value, we have chosen Tobin's Q as a metric for measuring enterprise value. The results of (4), (5), and (6) in Table 4 report the regression results of D&O insurance, audit fees, and corporate value. From the results of columns (4) and (5), it can be seen that not only the introduction of D&O insurance can improve corporate value, but also the increase in audit fees can also enhance corporate value. The results of column (6) show that the interaction term between D&O insurance and audit fees is significantly positively correlated with corporate value, indicating that the introduction of D&O insurance into the company has played a full supervisory role, reduced agency costs, and improved corporate value. 283 Table 4. The mechanism test of D&O insurance Variable Audit quality Corporate Value (1) (2) (3) (4) (5) (6) DA DA DA Tobin-Q Tobin-Q Tobin-Q Insurance -0.008*** (-3.73) -0.043** (-2. 18) 0.168*** (5.78) 0.637* (1.89) Audit_Fee -0.013*** (-10.98) -0.012*** (-10.63) 0.165*** (9.04) 0.149*** (7.08) Insurance* Audit_Fee -0.003** (-2.25) 0.058** (2.21) N 34737 34737 34737 34737 34737 34737 Year Yes Yes Yes Yes Yes Yes Industry Yes Yes Yes Yes Yes Yes Control Yes Yes Yes Yes Yes Yes Adj_R2 0.113 0.109 0.104 0.349 0.305 0.127 4.4. Robustness test 4.4.1. PSM method test Due to the fact that only about 10% of the companies in the sample have purchased D&O insurance, there may be sample selection bias, which may affect the accuracy of the regression results. In order to eliminate this bias, this article uses a 1:1 propensity score matching method (PSM) to match the samples to ensure the reliability of the regression analysis. Through this method, we can more accurately estimate the impact of D&O insurance on company performance and obtain more credible conclusions. Column (1) of Table 5 is the regression result tested by PSM, which shows that companies purchasing D&O insurance can effectively increase audit fees. 4.4.2. Method IV test Considering that companies with higher audit quality usually have sound governance mechanisms, D&O insurance, as an external governance tool, may be introduced by these companies to further improve governance efficiency. In order to control this potential reverse causality effect and enhance the credibility of causal inference in empirical research, this article uses a two-stage least squares method for regression analysis again. This method helps to more accurately identify the impact of D&O insurance on corporate governance efficiency and provides more reliable empirical results. Therefore, this article chooses to use the annual average of purchasing D&O insurance as an instrumental variable[7]. The reason is that the correlation between the annual average of purchasing D&O insurance and audit quality is low, and the higher the average, the greater the probability that the company will purchase D&O insurance. Columns (2) and (3) of Table 5 report the regression results of the instrumental variable two-stage least squares method. The regression results show that D&O insurance can significantly increase audit fees, which strongly proves the correctness of the core conclusion of this article. Table 5. Robustness test Variable (1) (2) (3) Audit_Fee Insurance Audit_Fee Insurance 0.133*** (8.36) -0.119*** (-4.22) mean_Ins 0.034*** (11.37) N 3488 34737 34737 Year Yes Yes Yes Industry Yes Yes Yes Control Yes Yes Yes Adj_R2 0.688 0.101 5. Conclusion Based on the data of listed companies in Shanghai and Shenzhen from 2009 to 2022, this article empirically examines the impact of D&O insurance on audit fees and deeply explores the mechanism of D&O insurance. The research results show that the introduction of D&O insurance significantly increases the audit fees of listed companies. Further analysis of its mechanism of action reveals that the increase in audit fees stems from the demand for high audit quality by D&O insurance as a governance mechanism, which reflects its supervision and incentive effects, rather than due to opportunistic and moral hazard behaviors. In addition, the study also found that there is a significant interactive effect between D&O insurance and audit fees, which once again verifies the existence of its supervision and incentive effects. Based on the research results of this article, we draw the following implications: Firstly, D&O insurance has the dual nature of supervision and incentive effects and may lead to opportunistic behavior, so we need to fully understand its true governance effect and action path, so as to fully exert its positive governance function and avoid possible negative effects. Secondly, as an effective external governance 284 mechanism, D&O insurance can not only supervise the behavior of directors and executives, but also motivate them to work hard, innovate boldly, and inhibit opportunistic behavior. Therefore, the introduction of this mechanism is of positive significance for improving the construction of modern enterprise system, enriching the external governance mechanism of listed companies, and enhancing the level of corporate governance. Acknowledgment This paper is supported by the Guangxi Graduate Education Innovation Project, with project number A3430022003. Thank you for the financial support provided by this project, which has enabled the smooth conduct of research work. At the same time, I also thank the relevant institutions and individuals for their support and assistance during the project application and implementation process. Finally, I would like to thank the reviewers and editors for their careful review and valuable comments. Their suggestions have played an important role in improving and enhancing this article. References [1] Noel O'Sullivan. The impact of directors’ and officers’ insurance on audit pricing: Evidence from UK companies[J]. Accounting Forum, 2009, 33(2): 146-161. [2] YW Wang, CW Chen. Directors' and officers' liability insurance and the sensitivity of; directors' compensation to firm performance [J]. INTERNATIONAL REVIEW OF ECONOMICS & FINANCE, 2016, 45286-297. [3] Zhihong Chen, Oliver-Zhen Li, Hong Zou. DirectorsΧ³ and officersΧ³ liability insurance and the cost of equity[J]. Journal of Accounting and Economics, 2016, 61(1): 100-120. [4] H ZOU, S WONG, C SHUM, et al. Controlling-minority shareholder incentive conflicts and directors’ and officers’ liability insurance: Evidence from China[J]. Journal of Banking & Finance, 2008, 32(12): 2636-2645. [5] Chen Lin, Micah-S. Officer, Hong Zou. Directors' and officers' liability insurance and acquisition outcomes[J]. Journal of Financial Economics, 2011, 102(3): 507-525. [6] T Baker, SJ Griffith. Predicting corporate governance risk: Evidence from the directors' & officers' liability insurance market[J]. UNIVERSITY OF CHICAGO LAW REVIEW, 2007, 74(2): 487-544. [7] Rongli Yuan, Jian Sun, Feng Cao. Directors' and officers' liability insurance and stock price crash risk[J]. Journal of Corporate Finance, 2016, 37173-192. [8] Hyeesoo-(Sally)-H. Chung, Stephen-A. Hillegeist, Jinyoung-P. Wynn. Directors’ and officers’ legal liability insurance and audit pricing[J]. Journal of Accounting and Public Policy, 2015, 34(6): 551-577. [9] Patricia-M. Dechow, Richard-G. Sloan, Amy-P. Sweeney. Detecting Earnings Management[J]. The Accounting Review, 1995, 70(2): 193-225. [10] ML DeFond, K Raghunandan, KR Subramanyam. Do non- audit service fees impair auditor independence? Evidence from going concern audit opinions[J]. JOURNAL OF ACCOUNTING RESEARCH, 2002, 40(4): 1247-1274.