Earnings Management and Internal Control in Bank-dominated Corporate. Governance: Evidence from Japan

Transcription

1 1 Earnings Management and Internal Control in Bank-dominated Corporate Governance: Evidence from Japan ABSTRACT We examine the relation between internal governance and earnings management in Japanese listed firms. Post the recent accounting fraud in large companies such as Olympus. Corp. and Toshiba Corp., Japanese internal governance systems are also widely criticized. Different from US and UK, Japan is known as bank-dominated corporate governance system. We predict that bank-client relation are expected for mitigating opportunistic earnings management by mitigating the degree of information asymmetry that is a main cause of agency problems arising from debt-contract. Our results show that bank-appointed audit board members mitigate managerial earnings management. Furthermore, neither outside directors nor ACs are helpful to decrease opportunistic managerial earnings management. Our findings imply that lender monitoring system through audit board members can contribute to substitute for monitoring role of outside directors and ACs. Keywords: Auditing; Japan; Agency Theory; Corporate Governance

2 2 I. INTRODUCTION Accounting earnings are more reliable and more informative when managers opportunistic behavior is inhibited through various monitoring systems (Dechow, Sloan, & Sweeney, 1996; Wild, 1996). Nevertheless, managers can manipulate earnings to maximize their own interests and influence the informativeness of earnings (Chung, Firth, & Kim, 2002; Gul, Chen, & Tsui, 2003; Healy, 1985; Holthausen, Larcker, & Sloan, 1995). After several financial scandals such as those involving Enron or Worldcom, an international trend has prevailed for development and implementation of corporate governance mechanisms to tackle the opportunistic behaviors that have assumed investors credibility in financial information. Because of recent accounting scandals such as Kanebo Co., Olympus Corp., and Toshiba Corp., internal corporate governance systems in Japan have also been widely criticized by Western media and practitioners (Financial Times, 2007; Financial Times, 2012a; Financial Times, 2015; etc.). We address the empirical question of whether or not Japanese internal control system featured as board structure and bank monitoring can mitigate opportunistic earnings management, which helps to provide a clue as to how effective internal control system can design among different nations. Under separation of ownership and control in public companies, corporate governance provides important mechanisms that affect managerial decision making. In market-oriented or shareholder-oriented economies such as the US, for some monitoring mechanisms such as board independence and audit committees are expected to control public firms.

3 3 Within countries having strong legal enforcement, the relations of bank-dominated or stakeholderoriented economies such as those of Japan and Germany differ from market-oriented or shareholderoriented economies such as those of the US and UK (Aoki, 1990; Yoshimori, 1995). These remarkable differences might affect the association between monitoring systems and earnings management. Information asymmetry between lenders and borrowers matters in agency problem arising from debtcontract. Monitoring systems that help to mitigate the information asymmetry can mitigate this type of agency problems. In fact, previous studies have indicated that commercial banks are expected to be delegated as monitors for other creditors and shareholders (Aoki et al., 1994; Morck & Nakamura, 1999). Corporate governance is expected to inhibit opportunistic earnings management, which often results in harmful corporate scandals. Regarding earnings management, Leuz, Nanda, and Wysocki (2003) point out a relation between corporate governance features and earnings management, which implies that firms with weaker monitoring systems might leave much room for managers to engage in accounting fraud 1. In the US, a board structure with higher independence of the audit committee (AC) can be used to monitor corporate accounting processes (Klein, 2002). Results of earlier studies (Garcia- 1 Leuz et al. (2003) defined the following three distinct country clusters: (1) outsider economies with strong legal enforcement, like that prevailing in the US, (2) insider economies with strong legal enforcement like that of Japan, and (3) insider economies with weak legal enforcement like that of India.

4 4 Meca & Sanchez-Ballesta, 2009; Ghosh, Marra, & Moon, 2010; Jaggi & Leung, 2007) have implied that fairness of financial reporting depends on the effectiveness of internal control mechanisms. This paper examines the role of both board independence and bank-appointed directors or bankappointed audit board members who are expected to be effective monitors in relationship-oriented systems (Aoki, 1990). While board and audit committees independence can reduce opportunistic earnings management in US (Klein, 2002), Japanese traditional internal governance systems differ from those of the US. Japanese corporate governance has features of a consensus-oriented system. Board members are occupied mainly with inside directors who have been promoted from roles of firm employees and rarely includes outside directors. Moreover, bank relations are expected to function as effective monitoring mechanisms, different from Anglo-American internal control systems (Aoki, 1990). In Japan, bank-appointed board members (or Torishimariyaku in Japanese) might play an important monitoring role. In addition, the role of audit board members (or Kansayaku in Japanese) is important because they participate in board meetings and engage in checking the accounting processes of directors on the board. This paper empirically investigates the relation between internal governance and earnings management in Japanese listed firms during We find that bank-appointed audit board members are effective for preventing opportunistic earnings management. In addition, both outside directors and ACs do not contribute to decrease opportunistic managerial earnings management. These results suggest that Japanese bank-dominated corporate governance takes an important monitoring role

5 5 of mitigating earnings management. We contribute to the literature related to opportunistic earnings management. Recent accounting fraud uncovered in large companies have revived the debate related to desirable monitoring mechanisms under bank-dominated corporate governance systems like Japan (Financial Times, 2006; Financial Times, 2011). Our findings suggest that monitoring mechanisms that mitigate information asymmetry between lenders and borrowers could reduce managerial opportunistic earnings management. We make important contribution to provide empirical evidences related to the debate on desirable monitoring systems in Japan. In addition, this study contributes to the growing finance literature related to lenders monitoring. Previous studies have examined financial performance or executive turnover related to bank monitoring in Japan (Kaplan & Minton, 1994; Morck & Nakamura, 1999; Morck, Nakamura, & Shivdasani, 2000; etc). We extend the literature related to monitoring of lenders that have not examined relation between bank-appointed audit board members and earnings management in Japan. The remainder of our manuscript is organized as follows. Next section presents the background of Japanese corporate governance. This paper subsequently discusses our empirical predictions. Next, we explain our empirical strategies, data, and descriptive statistics. We then present a summary of our empirical results. Finally, we conclude this paper with a discussion of our findings.

6 6 II. JAPANESE CORPORATE GOVERNANCE Japanese corporate governance mechanisms were known as bank-dominated and relationship-oriented systems during the 1990s (Aoki, 1990). Hoshi and Kashyap (2001) summarize the features of these corporate governance mechanisms such as bank monitoring system and Keiretsu known as Japanese business groups. The Japanese corporate governance system is in a transition era. Especially after the financial deregulation of the 2000s, commercial bank cross-shareholding has been limited by the Bank Shareholding Restriction Law of 2001 (Hoshi & Kashyap, 2010). We introduce Japanese internal audit systems of two types as portrayed in Figure 1. Before establishment of the Commercial Code (2003), firms have had to adopt a dual system, which comprises a board of directors and audit board members (or traditionally designated as statutory auditors) as shown in (1) of Figure 1. Audit board members can participate in board meetings and can monitor the financial reporting of firms. Although audit board members express their opinions at board meetings, they have no voting rights related to management decisions. Amendments made in 2001 were undertaken to strengthen the independence of audit board members 2. After 2005, firms that adopted a dual system must appoint audit board members, of whom most are outside audit board members. 2 For example, firms are encouraged to increase the number of outside audit board members. In addition, the terms of office and responsibility of audit board members are extended by law.

7 Figure 1 goes about here After the 2000s, Japanese corporate governance mechanisms have been reformed to adopt Anglo- Saxon corporate governance mechanisms, inducing changes of board systems and traditional bankdominated corporate governance features (Chizema & Shinozawa, 2012). First, the introduction of the US-style committee system has been permitted after the establishment of Commercial Code (2003). Therefore, adopting the committee system depends on voluntary choice of firms, as shown in (2) of Figure 1. Furthermore, the principle of Corporate Governance in Japan, which recommends that firms with at least one outside director be established (Tokyo Stock Exchange, 2009). However, appointing outside directors is not mandated under audit board member systems. Audit quality also depends on the audit firm size (Becker, DeFond, Jiambalvo, & Subramanyam, 1998; Francis, Maydew, & Sparks, 1999). Regarding the change of accounting audits, large Japanese audit firms were restructured after Kanebo s earnings fraud was uncovered in The large audit firms were known as the Big 4 audit firms: Deloitte Touche Tohmatsu, Shin-Nihon Ernst & Young, KPMG Azsan LLC, and Chuo-Aoyama PWC. In 2006, PWC was structured as Arata PWC and Chuo- Aoyama was restructured as Misuzu, which became defunct in July Thereafter, the Big 5 audit firms were Deloitte Touche Tohmatsu, Shin-Nihon Ernst & Young, KPMG Azsan LLC, Arata PWC, and Chuo-Aoyama during From 2008 to the present, the Big 4 audit firms excluding Chuo-

8 8 Aoyama have been large audit firms in Japan. III. EMPIRICAL PREDICTIONS Previous literatures suggest two types of views related to opportunistic earnings management. In a dispersed ownership structure, the monitoring roles of the board of directors are useful to monitor managers activities (Fama & Jensen, 1983; Jensen & Meckling, 1976; Shleifer & Vishny, 1997). Corporate governance mechanisms contribute to the alignment of the interests of managers with the interests of shareholders by enhancing the reliability of financial information (Watts & Zimmerman, 1986). On the other hands, Japanese traditional corporate governance system is bank-dominated, which is different from market-oriented systems like Anglo-Saxon countries (Aoki, 1990; Aoki, Patrick, & Sheard, 1994; Sheard, 1994). We examine to assess whether internal control affects the effectiveness of preventing earnings management or not under bank-dominated corporate governance. Internal Control Systems We examine three internal control systems to analyze Japanese firms. First, Jensen (1993) points out that board of director monitoring roles can become less effective as the board size increases. This effect is attributable to problems of coordination and communication. However, mixed evidence exists about the relation between board size and earnings management. For Hong Kong, Chin, Firth, and Rui (2006) find a positive association between board size and earnings management. This finding suggests

9 9 that effective monitoring is expected for firms with smaller boards. On the other hand, Xie et al. (2003) find a negative relation between board size and earnings management in US firms. Second, the board ability to function as an effective monitor depends on its independence from management (Beasley, 1996). Therefore, board independence is also expected for effective internal control mechanisms. Mixed evidence also exists for the relation between earnings management and board independence. In Australia, Davidson, Goodwin-Stewart, and Kent (2005) find a negative relation between earnings management and board independence. On the other hand, Bradbury Mak, and Tan (2006) find no significant association between board independence and earnings management in Singapore. Third, Japanese firms are permitted to introduce the US-style committee system that are intended to strengthen the internal controls in Japan (Commercial Code, 2003). Effective ACs can be expected to help improve financial reporting quality and reduce opportunistic earnings management. Regarding the role of ACs, Klein (2002) finds that independent audit committees constrain managerial earnings manipulation and enhance the quality of financial reporting in the US. Empirical studies of several countries outside the US also show the effective monitoring role of audit committees (Garcia-Meca & Sanchez-Ballesta, 2009; Ghosh et al., 2010; Jaggi & Leung, 2007; Kang, Kilgore, & Wright, 2011). Bank Monitoring System Monitoring system that eliminates information asymmetry between lenders and borrowers are

10 10 important issues to decrease agency problems arising from debt contract. In a dispersed ownership structure, the monitoring roles of the board of directors are useful to monitor managers activities (Fama & Jensen, 1983; Jensen & Meckling, 1976; Shleifer & Vishny, 1997). In addition, board directors which are appointed by commercial banks can provide their private information and participate in managerial decision making such as capital investment related to the clients firms (Hoshi, Kashyap, & Scharfestein, 1990, 1991). Bank monitoring roles are represented as a function of client firms connected by bank relations (Aoki et al., 1994). In fact, lending activities help to reduce information asymmetry and increase of market liquidity because private information might be gathered form bank relationships (Sakawa, Ubukara, and Watanabel, 2014) In addition, Commercial banks serve as a disciplinary mechanism by appointing directors who had been previously employed by banks, and by forcing executive turnover at firms when firms face poor performance (Kaplan & Minton, 1994; Kang & Shivdasani, 1995). Bankappointed directors enhance executive incentives in Japanese firms (Colpan & Yoshikawa, 2012; Sakawa & Watanabel, 2008). In addition, Commercial banks have an incentive to participate in management decision makings of their clients firms through appointing a member of board of directors or audit board. Owing to their monitoring activities, the banks have information advantage related to capital investment of their clients firms. Even recently, appointing directors or audit board members (or audit and supervisory board

11 11 members) from commercial banks is a typical means of supporting firms that have poor performance 3. If commercial banks fail to monitor their clients firms, they have to bear economic loss such as a decline of the banks stock price 4. Therefore, findings of these previous studies imply that the monitoring roles of members who attend board meetings are important for mitigating earnings management. Japanese internal governance systems are based mainly on dual audit systems that consist of boards of directors and audit board members. Under Japanese bank-dominated corporate governance, bank relationships represented as bank-appointed directors or audit board members can be expected for effective monitors by reducing information asymmetry degree. Appointment of bank directors or audit board members are expected to help to mitigate opportunistic earnings management in Japanese firms through the appointment of directors. IV. EMPIRICAL METHOD AND DATA Empirical Methodology We construct a model examining the association between earnings management and Japanese internal control mechanisms. This study adopts discretionary accruals as a proxy of earnings management, 3 The bank ties present benefits for shareholders of client firms because banks appoint directors for client firms when they are in a bad condition (Kaplan & Minton, 1994). However, such practices have raised concern by shareholders, particularly foreign investors (Financial Times, 2012b). 4 In case of Toshiba, stock price of lenders which have relationship with Toshiba declined after announcement of the accounting scandals (Financial Times, 2016)

12 12 consistent with previous studies (Bartov, Mohanram, & Seethamraju, 2001; Klein 2002; Warfield, Wild, & Wild, 1995). We adopt the absolute value of discretionary accruals because managers adjust the earnings report upward or downward using positive or negative accruals. According to the Jones and the cross-sectional modified Jones models, nondiscretionary accruals (DA) are calculated using the following equation (1), which represents a difference between total accruals (TA) and non-discretionary accruals (NDA), divided by total assets for the beginning period. The total accruals (TA) are the difference between net income and cash flows from operations. 1 Total accruals (TA) are calculated as the Jones, the cross-sectional modified Jones, and the CFOmodified Jones model (Jones, 1991; Dechow, Sloan, & Sweeney, 1995; Kasznik, 1999). Assets (A) denote total assets for firm i in year t. Each model is estimated using the following equation (2). Each model s non-discretionary accruals (NDA) are estimated by each of (3), (4), and (5). α β β ε ND α β β ND α Δ β β ND α Δ β β β 5 Therein, TA stands for total accruals, measured as the difference between net income (earnings before extraordinary items and discontinued operations) and operating cash flows for firm i in year t: ΔREV signifies the change in net revenue for firm i in year t; ΔAR is the change in accounts receivable for firm

13 13 i in year t; PPE represents property, plant, and equipment for firm i in year t; ΔCFO is the change in operational cash flow for firm i in year t; and ε it represents an error term for firm i in year t. The models are estimated separately for each Nikkei Industry Classification Code and year to produce industryspecific estimates of the coefficients, following Teshima and Shuto (2008). The change in accounts receivable (ΔAR) is not included in estimating the parameters of TA in equation (2), but it is included in estimation of non-discretionary accruals (NDA) in equation (4). To investigate our empirical predictions, we examine the following OLS regression equation. DA β β β Σβ Control ε 6 In that equation, DA stands for the absolute value of discretionary accruals (DA) using the Jones and the modified Jones model. We measure both internal control systems and bank monitoring mechanisms. We include three internal control variables such as board size, outside directors, and audit committees. Board Size represents the number of directors on the board. Outside Directors is the percentage of directors from outside the firm 5. Committee represents the dummy of the AC, which is equal to 1 if a firm establishes 5 Followed as Japanese previous studies, we define board independence as proportion of outside directors which are directors who have never served as executive director, executive officer, employee of the company or any of its subsidiaries, as reported in the companies annual report. We also check the robustness of results using of strictly defined board independence in the latter section.

14 14 AC; otherwise it is 0. In addition, we adopt two bank monitoring variables such as the number of bankappointed directors (Bank Directors) and bank-appointed audit board members (Bank Audit Board). Regarding audit board members, firms with committees have no board of audit members (Kansayaku in Japanese). Therefore, we drop firms with committees to analyze the role of Bank Audit Members. The control variables were selected as in previous studies. Management Shareholdings signifies the proportion of directors shareholdings, following Morck, Nakamura, and Shivdasani (2000) and Teshima and Shuto (2008). We adopt Stock return volatility (Volatility) to control for firm risk. We adopt the Big N dummy variable: Big N is equal to 1 if a firm is audited by a Big 4 or Big 5 audit firm; otherwise it is 0. Firm size is controlled by the logarithm of market value (MV). We control for growth as a Market to Book ratio because growing firms are more likely to be associated with earnings management (Matsumoto, 2002). Return on assets (ROA) is adopted to control for firm profit. We also include financial leverage (Leverage) to control for correlation with discretionary accruals (DeFond & Jiambalvo, 1994; Frankel, Johnson, & Nelson, 2002). Earnings management is more prevalent at firms whose executive compensation is closely tied to stock value via stock options (Bergstresser & Philippon, 2006; Cornett, Marcus, & Tehranian, 2008). We control for stock based incentives by the adoption of executive stock option (Stock Option) (Sakawa, Moriyama, & Watanabel, 2012). Finally, we also use the Topix 500 dummy (Topix 500) as a control variable to control firm size. Topix 500 index consists of the top 500 listed firms on Tokyo stock Exchange and accounts for about 85 percent of the Exchange related to market capitalization.

15 15 Data and Descriptive Statistics We select data from non-financial firms listed on the first section of Tokyo Stock Exchange during Financial accounting data were collected from the Nikkei Needs Corporate database. Corporate governance data including the ownership structure and board composition were obtained from the Nikkei Needs Corporate governance evaluation system (CGES) database. The audit firms information was hand-collected by Toyo Keizai ( ). The big N audit firms consist of Big 4 or Big 5 audit firms during our sample periods 6. Our selected sample consists of 11,689 firm year observations Table 1 and Table 2 go about here Table 1 and Table 2 present descriptive statistics and the correlation matrix. The mean of the 6 An earlier report presented by Yoshida (2006) explained these methods to gather audit firm information in Japan. 7 We restrict our sample firms based on the following criteria: (i) Financial statements are available in the sample period. (ii) Observations include at least 15 firms in the same industry classified by Nikkei Industry Classification Code.

16 16 absolute value of discretionary accruals (DA) using the Jones, modified Jones, and modified CFO Jones are about 3% of total assets. The respective averages of Bank Directors and Bank Audit Board Members is 0.67 and The average of Board Size is about 9. That of Outside Directors is about 10%, which implies that the board size is the same level in US and that board independence is less. The average of Committee occupies only 2% of samples, which implies that almost no firms have introduced ACs. V. EMPIRICAL RESULTS OLS Results We investigate our empirical predictions related to Japanese internal control mechanisms and earnings management by estimating the OLS results of equation (6). In Table 3, we can find Board Size is significant and negatively related with earnings management. This result implies that larger boards function to mitigate earnings management. Both Outside Directors and Committee are not significantly correlated with earnings management. Table 3 shows that that Bank Director is not significantly, but negatively associated with earnings management Table 3 goes about here The results of control variables show that Management shareholdings are significant and positive, which implies that agency conflicts can be expected to occur in firms with higher managerial ownership.

17 17 The variable Volatility is significant and positive. Big N is not significant. ln (MV) is negative and not significantly related to earnings management. Market to Book is positively associated with earnings management, suggesting that earnings management is greater for firms with high growth opportunities, consistent with Matsumoto (2002). ROA and Leverage are not significant. Stock Option is positively associated with earnings management, consistent with Bergstresser and Philippon (2006) and Cornett et al. (2008). Finally, TOPIX 500 is negatively associated with earnings management. We next report the OLS results including those for Bank Audit Member in Table 4 8. As we described, samples with ACs are omitted because audit board members are not appointed in firms with ACs. First, Board Size is negatively associated with earnings management. Both Outside Directors and Audit Board Size are not significant. We find that Bank Audit Board is significant and negatively associated with earnings management. This result implies that bank client relations function effectively through appointment of audit board members Table 4 goes about here We also summarize the results of control variables here. Management shareholdings and Volatility 8 For consideration of the multicollinearity of Bank Directors and Bank Audit Members, we include these two variables separately.

18 18 are significant and positive, similar to the results presented in Table 3. ln (MV) is negative and not significant. Market to Book is positively related to earnings management, as presented in Table 3. ROA and Leverage are not significant. Stock Option is positively associated and TOPIX 500 is negatively associated with earnings management, as described in the results of Table 3. Additional Analysis We examine the relation between internal governance system and earnings management in Japan. In this section, we check the robustness of our findings. Firstly, we adopt strict definition of board independence followed as previous study like Klein (2002) 9. After the robustness estimation using of strictly independent board independence, we next consider endogeneity among internal control and earnings management. Firstly, we check the robustness using of alternative definition of board independence in un-tabulated results. Using of the strictly defined board independence, we also do not find the significant results of both board independence and Committees. Therefore, we can confirm the 9 Previous studies like Klein (2002) defines board independence by several criterions. In Japan, the guideline of TSE (Tokyo Stock Exchange) has changed to define board independence strictly. Followed as the guideline of TSE, we define five categories of outside directors who might not be truly independent directors : (1) directors related to the parent company, (2) directors related to other affiliated companies, (3) directors who come from commercial banks, (4) directors who are mutually dispatched, and (5) directors who are mutually dispatched.

19 19 robustness of our results by un-tabulated estimations. Second, we investigate the possible endogenous relations between monitoring activity and earnings management. To explore endogenous relations between internal control and earnings management, we use two-stage least-squares regression. We adopt instrument variables as lagged variables of instrumented variables, foreign shareholdings, and a dummy variable of American depositary receipts (ADR) Table 5 goes about here Using Table 5, the coefficient of Bank Audit Board is negatively associated with earnings management for all three models. Board Size is significant and negative in Model (3) after considering endogeneity. Outside Directors is significant and positive in Model (1), which implies that they are ineffective monitors for preventing opportunistic earnings management. As for the other firm characteristic variables, Audit Board Size is negative and not significant. Management Shareholdings and Volatility are significant and positive. Big N is not significant, which implies that large audit firms with good reputations are not helpful to mitigate opportunistic earnings 10 ADRs are equal to 1 if a firm has American depositary receipt (ADR) programs, otherwise they are 0. Kang and Stultz (1997) find that firms with ADR programs have greater foreign ownership. Therefore, we adopt the ADR dummy as an instrument variable.

20 20 management in Japan. The results of financial variables are almost identical to the OLS results. VI. CONCLUSIONS Japanese audit quality has been discussed frequently since several cases of corporate fraud surfaced recently: Kanebo, Olympus, and Toshiba. The principle of corporate governance in Japan recommends outside directors as effective monitors. These principles are expected to transform the prevailing insider-dominated board system in Japan. We investigated whether outside directors or ACs mitigate managerial opportunistic earnings management or not. In addition, Japanese corporate governance systems are known as bank-dominated system (Aoki 1990; Aoki et al., 1994). Consistent with the perspectives of Aoki (1990) and Aoki et al. (1994), we examine monitoring role of bank client relations, which are expected to include bank-appointed directors and bank-appointed audit board members. We provide empirical evidences related to the association between earnings management and monitoring mechanisms. First, we find that bank monitoring mechanisms function through bankappointed audit board members. This result implies that bank-appointed audit board members mitigate earnings management by mitigating information asymmetry degree between lenders and borrowers. Second, the OLS suggest larger boards are effective for mitigating management opportunistic earnings management. After considering endogeneity, empirical results weakly support the relation between board size and earnings management. Finally, neither outside directors nor ACs are helpful to decrease opportunistic managerial earnings management. US-style committees have not been introduced into

21 21 most firms, and they do not affect managerial opportunistic earnings management under Japanese bankdominated systems. Under Japan s unique internal control system, we reveal the monitoring roles of audit board members under the dual audit system. This study has several limitations. First, our findings focus on monitoring roles of bank-dominated corporate governance systems related to opportunistic earnings management. Therefore, it is a future task to examine whether or not bank-dominated corporate governance systems are fully functioning effectively after corporate governance reforms such as the introduction of US-style committees. Second, we focus on Japanese firms to address serious concerns related to the accounting scandals. Therefore, it is a valuable research to provide insights related to the effective internal control mechanisms in any country where internal control mechanism is reformed to introduce US style internal control systems. Overall, our study provide an invaluable insight for both investors and regulators to ascertain effectiveness of bank client relations as effective gatekeepers in Japanese corporate governance.

22 22 APPENDIX A Definition of Variables Variable Definition Dependent Variables Jones Absolute value of discretionary accruals (DA) using Jones model (%) Modified Jones Absolute value of discretionary accruals (DA) using a modified Jones model (%) Modified CFO Jones Absolute value of discretionary accruals (DA) using a modified Jones model with cash flow (%) Board Variables Board Size Outside Directors Committee Bank Directors Bank Audit Board Audit Board Size Number of directors on the board Proportion of outside directors Committee is equal to 1 if a firm adopts the Audit Committee (AC), otherwise it is 0. Proportion of outside directors who come from commercial banks. Number of outside audit board members who come from commercial bank Number of audit board members Other Variables Management Proportion of board of directors' shareholdings Shareholdings Volatility Stock price volatility during three years Big N Big N audit firms consist of big 4 or big 5 audit firms ln (MV) The logarithm of market value is adopted Market to Book Market to Book = Market Value/ Book Value of Capital ROA Return on Assets Leverage Financial Leverage = Debt/Total Assets Stock Option Stock Option is equal to 1 if a firm adopt stock option; otherwise it is 0. Topix 500 Topix 500 is equal to 1 if a firm include Topix 500 index; otherwise it is 0.

23 23 REFERENCES Aoki, M Toward an economic model of the Japanese firm. Journal of Economic Literature, 28: Aoki, M., Patrick, H., & Sheard, P The Japanese main bank system: An introductory review. In Aoki, M. & H. Patrick (Eds.), The Japanese main bank system: Its relevancy for developing and transforming economies: Oxford: Oxford University Press. Bartov, E., Mohanram, P. S., & Seethamraju, C Valuation of internet stocks An IPO perspective. Journal of Accounting Research, 40: Beasley, M. S An empirical analysis of the relation between the board of director composition and financial statement fraud. The Accounting Review, 71: Becker, C. L., DeFond, M. L., Jiambalvo, J., & Subramanyam, K. R The effect of audit quality on earnings management. Contemporary Accounting Research, 15: Bergstresser, D., & Philippon, T CEO incentives and earnings management, Journal of Financial Economics, 80: Bradbury, M., Mak, Y., & Tan, S Board characteristics, audit committee characteristics, and abnormal accruals. Pacific Accounting Review, 18: Chin, K., Firth, M., & Rui, O Earnings management, corporate governance, and the market performance of seasoned equity offerings. Journal of Contemporary Accounting and Economics, 2:

24 24 Chizema, A., & Shinozawa, Y The company with committees : Change or continuity in Japanese corporate governance? Journal of Management Studies, 49: Chung, R., Firth, M., & Kim, J Institutional monitoring and opportunistic earnings management. Journal of Corporate Finance, 8: Colpan, A., & Yoshikawa, T Performance sensitivity of executive pay: The role of foreign investors and affiliated directors in Japan. Corporate Governance: An International Review, 20: Commercial Code Revision on Boards Ministry of Justice, Japan. Cornett, M. M., Marcus, A. J., & Tehranian, H Corporate governance and pay-for-performance: The impact of earnings management. Journal of Financial Economics, 87: Davidson, R., Goodwin-Stewart, J., & Kent, P Internal governance structures and earnings management. Accounting and Finance, 45: Dechow, P. M., Sloan, R. G., & Sweeney, A. P Detecting earnings management. The Accounting Review, 70: Dechow, P. M., Sloan, R. G., & Sweeney, A. P Causes and consequences of earnings manipulation: An analysis of firms subject to enforcement actions by the SEC. Contemporary Accounting Review, 13: DeFond, M. L., & Jiambalvo, J Debt covenant violation and manipulation of accruals. Journal of Accounting and Economics, 77:

25 25 Fama, E., & Jensen, M Separation of ownership and control. Journal of Law and Economics, 26: Financial Times Kanebo s history of alleged cooking the books. May 12. Financial Times An accountability gap is holding back Japan's economy. March 15. Financial Times Ex-Olympus chief questioned payments. October 14. Financial Times. 2012a. Former Olympus chief warns on governance. April 19. Financial Times. 2012b. Olympus unveils new management. February 27. Financial Times Toshiba chief promises overhaul amid calls for him to quit. June 25. Financial Times Toshiba shares sink for a third day as financial worries deepen. December 29. Francis, J. R., Maydew, E. L., & Sparks, H. C The role of big 6 auditors in the credible reporting of accruals. Auditing: A Journal of Practice & Theory, 18: Frankel, R. M., Johnson, M. F, & Nelson, K. K The relation between auditors' fees for nonaudit services and earnings management. The Accounting Review, 77: Garcia-Meca, E., & Sanchez-Ballesta, J. P Corporate governance and earnings management: A meta-analysis. Corporate Governance: An International Review, 17: Ghosh, A., Marra, A., & Moon, D Corporate boards, audit committees, and earnings management: Pre- and post-sox evidence. Journal of Business Finance & Accounting, 37: Gul, F. A., Chen, C. J. P., & Tsui, J. S. L Discretionary accounting accruals, managers incentives,

26 26 and audit fees. Contemporary Accounting Research, 20: Healy, J The effect of bonus schemes on accounting decisions. Journal of Accounting and Economics, 7: Holthausen, R., Larcker, D., & Sloan, R Annual bonus schemes and the manipulation of earnings. Journal of Accounting and Economics, 19: Hoshi, T., Kashyap, A. & Scharfstein, D The role of banks in reducing the costs of financial distress in Japan. Journal of Financial Economics, 27: Hoshi, T., Kashyap, A. & Scharfstein, D Corporate structure, liquidity, and investment: Evidence from Japanese industrial groups. Quarterly Journal of Economics, 106: Hoshi, T., & Kashyap, A Corporate financing and governance in Japan: The road to the future. Cambridge, MA: MIT Press. Hoshi, T., & Kashyap, A Will the U.S. bank recapitalization succeed? Eight lessons from Japan. Journal of Financial Economics, 97: Jaggi, B., & Leung, S Impact of family dominance on monitoring of earnings management by audit committees: Evidence from Hong Kong. Journal of International Accounting, Auditing and Taxation, 16: Jensen, M The modern industrial revolution, exit, and the failure of internal control system. Journal of Finance, 48: Jensen, M., & Meckling, W Theory of the firm: Managerial behavior, agency costs, and

27 27 ownership structure. Journal of Financial Economics, 3: Jones, J Earnings management during import relief investigations. Journal of Accounting Research, 29: Kang, J. K., & Shivdasani, A Firm performance, corporate governance, and top executive turnover in Japan. Journal of Financial Economics, 38: Kang, J. K., & Stultz, R. M Why is there a home bias? An analysis of foreign portfolio equity ownership in Japan. Journal of Financial Economics, 46: Kang, W. S., Kilgore, A., & Wright, S The effectiveness of audit committees for low and midcap firms, Managerial Auditing Journal, 26: Kaplan, S. N., & Minton, B.A Appointments of outsiders to Japanese boards: Determinants and implications for managers. Journal of Financial Economics, 36: Kasznik, R On the association between voluntary disclosure and earnings management. Journal of Accounting Research, 37: Klein, A Audit committee, board of director characteristics, and earnings management. Journal of Accounting and Economics, 33: Leuz, C., Nanda, D., & Wysocki, P. D Earnings management and investor protection: An international comparison. Journal of Financial Economics, 69: Matsumoto, D. A Management's incentives to avoid negative earnings surprises. The Accounting Review, 77:

28 28 Morck, R., & Nakamura, M Banks and corporate control in Japan. Journal of Finance, 54: Morck, R., Nakamura, M. & Shivdasani, A Banks, ownership structure, and firm value in Japan. Journal of Business, 73: Sakawa, H., Moriyama, K., & Watanabel, N Relation between top executive compensation structure and corporate governance: Evidence from Japanese public disclosed data. Corporate Governance: An International Review, 20: Sakawa, H. & Watanabel, N Relationship between managerial compensation and business performance in Japan: New evidence using micro data. Asian Economic Journal, 22: Sakawa, H., Ubukata, M., & Watanabel, N Market Liquidity and Bank-dominated Corporate Governance: Evidence from Japan, International Review of Economics and Finance, 31: Sheard, P The main bank and the governance of financial distress. In Aoki, M. & H. Patrick (Eds.), The Japanese main bank system: Its relevancy for developing and transforming economies: Oxford: Oxford University Press. Shleifer, A., & Vishny, R. W A survey of corporate governance. Journal of Finance, 52: Teshima, N., & Shuto, A Managerial ownership and earnings management: Theory and empirical evidence from Japan. Journal of International Financial Management and Accounting, 19:

29 29 Tokyo Stock Exchange Principles of corporate governance for listed companies. Tokyo Stock Exchange Inc., Tokyo. Toyo Keizai Japan corporate handbook. Tokyo, Japan. Warfield, T. D., Wild, J. J., & Wild, K. L Managerial ownership, accounting choices, and informativeness of earnings. Journal of Accounting and Economics, 20: White, H A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroscedasticity. Econometrica, 48: Wild, J The audit committee and earnings quality. Journal of Accounting, Auditing and Finance, 11: Xie, B., Wallace, N., & Dadalt, P Earnings management and corporate governance: The role of the board and the audit committee. Journal of Corporate Finance, 9: Yoshida, K An analysis of audit quality and earnings management. In Japan Accounting Association Report (Ed.), Researches on Financial Reporting Reliability: (in Japanese). Yoshimori, M Whose company is it? The concept of the corporation in Japan and the west. Long Range Planning, 28:

30 30 FIGURE 1 Japanese Monitoring Systems of Two types (1) Dual audit system (Board Directors and Audit Board members) in Japan (Internal Audit) Board of Directors Audit Board CEO/President (External Audit) Accounting Auditors (2) Committee system (Internal Audit) Board of Directors Audit Committee CEO/President Compensation Committee Nomination Committee (External Audit) Accounting Auditors Each arrow represents the monitoring activities towards the CEO/President.

31 31 Variables are defined in Appendix A. TABLE1 Summary Statistics Number Mean Median S.D. 5% 95% Dependent Variables Jones Modified Jones Modified CFO Jones Board Variables Bank Directors Board Size Outside Directors Committee Bank Audit Board Audit Board Size Other Variables Management Shareholdings Volatility Big N ln (MV) Market to Book ROA Leverage Stock Option Topix

32 32 TABLE 2 Panel A Correlation Matrix (All firms) Jones 2. Modified Jones 0.96 * 3. Modified CFO Jones 0.75 * 0.77 * 4. Board Size * * * 5. Committee Outside Directors * 0.43 * 7. Bank Directors * * * * 0.28 * 8. Management Shareholdings 0.12 * 0.12 * 0.11 * * * * * 9. Volatility 0.20 * 0.20 * 0.21 * * * 0.04 * 10. Big N * * 0.07 * 0.11 * 0.03 * * 11. ln (MV) * * * 0.44 * 0.14 * 0.19 * 0.11 * * * 0.17 * 12. Market to Book 0.10 * 0.10 * 0.09 * * 0.10 * * 0.11 * 0.05 * 0.21 * 13. ROA * * 0.02 * 0.06 * * 0.26 * * 0.09 * 0.24 * 0.39 * 14. Leverage 0.05 * 0.05 * 0.06 * 0.12 * 0.04 * 0.02 * 0.02 * * 0.24 * * 0.07 * * 15. Stock Option 0.04 * 0.04 * 0.05 * * 0.10 * 0.17 * 0.10 * 0.15 * * 0.15 * 0.14 * 0.18 * * 16. Topix * * * 0.35 * 0.09 * 0.17 * 0.10 * * * 0.13 * 0.77 * 0.14 * 0.13 * * * indicates significance at the 0.05 level or better. N =11,689 Panel B Correlation Matrix (Dual audit system only) Jones 2. Modified Jones 0.96 * 3. Modified CFO Jones 0.75 * 0.77 * 4. Board Size * * * 5. Outside Directors * 6. Bank Directors * * * * 7. Audit Board Size * * * 0.39 * 0.07 * 0.06 * 8. Bank Audit Board * * * 0.10 * 0.03 * 0.05 * 0.14 * 9. Management Shareholdings 0.12 * 0.12 * 0.11 * * * * * * 10. Volatility 0.20 * 0.20 * 0.21 * * * * * 11. Big N * * 0.09 * 0.03 * 0.06 * 0.04 * * 12. ln (MV) * * * 0.44 * 0.15 * 0.10 * 0.46 * 0.13 * * * 0.16 * 13. Market to Book 0.10 * 0.10 * 0.08 * * * 0.11 * 0.04 * 0.21 * 14. ROA * * 0.05 * * * 0.25 * * 0.09 * 0.24 * 0.37 * 15. Leverage 0.05 * 0.05 * 0.05 * 0.12 * * 0.13 * * 0.24 * * 0.07 * * 16. Stock Option 0.04 * 0.04 * 0.05 * * 0.14 * 0.09 * * 0.15 * * 0.14 * 0.14 * 0.17 * * 17. Topix * * * 0.35 * 0.15 * 0.09 * 0.40 * 0.12 * * * 0.13 * 0.77 * 0.14 * 0.13 * * * indicates significance at the 0.05 level or better. N =11,437.

33 33 TABLE 3 Estimated Results (OLS) DA β β β Σβ Control ε 6 (1) (2) (3) (4) Modified Jones Board Size *** *** *** *** (-4.19) (-4.20) (-4.46) (-4.21) Outside Directors (1.38) (1.45) (1.57) Committee (-0.60) (0.20) (-0.63) Bank Directors (-0.39) (-0.84) Management Shareholdings *** *** *** *** (5.83) (5.83) (5.78) (5.81) Volatility *** *** *** *** (9.14) (9.15) (9.11) (9.13) Big N (0.19) (0.19) (0.27) (0.19) ln (MV) (-1.38) (-1.35) (-1.25) (-1.32) Market to Book * * * * (2.45) (2.45) (2.46) (2.45) ROA (-1.26) (-1.26) (-1.27) (-1.28) Leverage (0.51) (0.51) (0.55) (0.52) Stock Option *** *** *** *** (3.66) (3.67) (3.86) (3.71) Topix *** *** *** *** (-3.98) (-4.01) (-3.94) (-4.01) Constant ** ** ** ** (3.07) (3.02) (3.03) (3.01) Num. obs Adjusted R-squared F Test *** *** *** *** ***, **, *, and indicate significance at the 0.1, 1, 5, and 10%, respectively. Dependent Variables are Modified Jones discretionary accruals (DA) in Model (1)-(4). Standard errors are robust to heteroscedasticity (White, 1980) with t-values shown in parentheses. Robust t-statistics are shown in parentheses. Variables are defined in Appendix A.

34 34 TABLE 4 Estimated Results related to Audit Board Member (OLS) DA β β β Σβ Control ε 6 (1) (2) (3) (4) (5) Modified Jones Jones Modified CFO Jones Board Size *** *** *** *** *** (-3.97) (-4.16) (-3.95) (-3.67) (-4.90) Outside Directors (1.39) (1.43) (1.45) (1.77) (1.14) Audit Board Size (-1.24) (-1.07) (-1.00) (-0.79) Bank Audit Board *** *** ** *** (-3.86) (-3.79) (-3.11) (-3.86) Management Shareholdings *** *** *** *** *** (5.72) (5.74) (5.66) (5.70) (6.92) Volatility *** *** *** *** *** (8.88) (8.93) (8.89) (9.22) (8.83) Big N (0.20) (0.28) (0.28) (0.40) (0.72) ln (MV) (-1.13) (-1.20) (-0.98) (-1.40) (-0.26) Market to Book * * * * * (2.36) (2.36) (2.36) (2.37) (2.41) ROA *** (-1.27) (-1.29) (-1.33) (-0.49) (-3.83) Leverage (0.44) (0.55) (0.62) (1.43) (-0.23) Stock Option *** ** ** *** *** (3.39) (3.23) (3.24) (3.49) (4.95) Topix *** *** *** *** ** (-3.89) (-3.87) (-3.84) (-3.75) (-2.92) Constant ** ** ** ** * (3.22) (2.94) (3.04) (3.16) (2.46) Num. obs Adjusted R-squared F Test *** *** *** *** *** ***, **, *, and indicate significance at the 0.1, 1, 5, and 10%, respectively. Dependent Variables are Modified Jones discretionary accruals (DA) in Model (1)-(3), Jones discretionary accruals (DA) in Model (4), and Modified CFO Jones discretionary accruals (DA) in Model (5). Standard errors are robust to heteroscedasticity (White, 1980) with t-values shown in parentheses. Robust t-statistics are shown in parentheses. Variables are defined in Appendix A.