ABSTRACT

The present study tries to test the association between the total numbers of restatements filed in 2004 for selected firms and several audit committee expertise characteristics using correlation analysis and regression analysis. The evidence shows that there is negative association between the total number of restatements filed in 2004 and the combined accounting expertise and industry expertise. The regression results indicate that if at least one audit committee member having the combined accounting financial expertise and industry expertise, or the combined supervisory financial expertise and the industry expertise, the audit committee would effectively reduce the total number of restatements. The results provide evidence that accounting expertise with supplement of industry expertise would better serve the audit committee in its effectiveness in ensuring the integrity of the financial reporting process and the resulting financial statement delivered to the related stakeholders. The other results surprisingly show a positive association between the total number of restatements and only accounting expertise or only industry expertise.

Key words: audit committee, financial expert, industry expert, restatement

1. Introduction

The financial statements for the public company filed with the Securities and Exchange Commission (SEC) aim to keep the various stakeholders such as investors and the creditors informed of the financial information of the firms to make their decisions based on that financial information. Those financial statements are required both quarterly and annually with the SEC to meet the listing requirements. It is great opportunity for the firm management to be on the same page with outside stakeholders (at least on the paper). Thus, the accuracy of the financial statements and whether there are errors in the financial statements are very important for those investors and creditors. The corporate scandals from the companies like Enron, WorldCom, Tyco and others in the early 2000s are also due to the inaccuracy of their financial statements. All those companies must restate their previous statements afterwards.

By restating the financial statements, the firms explicitly admit there was material omissions and/or misstatements in their previous statements. The firms need take measures to correct those deviation from GAAP to make them right (Abbott et al. (2004), Shin et al. (2014)). No restatement for a firm does not necessarily mean the firm having high quality financial reporting. However, the firm definitely had low quality financial reporting if there was restatement announced.

In response to the fraud between 2000 and 2002, the Sarbanes-Oxley Act was enacted in 2002 on the requirement for US public company regarding the board, management, and public accounting firms with the goal to restore the confidence in the public company for the investors and other stakeholders. Particularly, Sarbanes-Oxley Act Section 407 required public company to disclose whether at least one audit committee financial expert (ACFE) serves on its audit committee or to disclose the reason for not having such an expert. There was much debate on the qualification of the criteria for audit committee financial expert. It is well known and established that the audit committee for the board of a firm is charged with the overseeing responsibilities with the accounting and financial reporting process of the firm and the auditing of the firm’s financial statements. The audit committee needs diligently work with the internal auditing and the external auditors to ensure the integrity of the firm’s financial statements. From this perspective, the composition of the audit committee and the audit committee member characteristics would be very important to the integrity and the accuracy of the financial statements of the firm.

In this paper, I would like to study the relation between the audit committee characteristics and the number of restatements announced and filed in the year of 2004. Each quarterly restatement and annual financial restatements will all be counted once toward the total number of the restatements in 2004. Specifically, I would test four types of the expertise on the audit committee following the methodology in Shin et al (2014). The different types of expertise include: accounting financial expert (AFE), non-accounting financial expert (FFE), supervisory financial expert (SFE), industry expert (INDE). Additionally, I also include AFEINDE (possessing accounting financial expertise and industry expertise at the same time) and SFEINDE (possessing supervisory financial expertise and industry expertise at the same time).

There were extensive studies on the impact of audit committee characteristics on the quality of the financial reporting, more specifically on the earning management. Yang and Krishnan (2005) studied the impact of audit committee and quarterly earning management and they found that “the number of outside directorships held by audit committee directors is negatively associated with earnings management behavior. This could reflect possible independence of these directors because of their desire to maintain their reputations or their expertise in dealing with financial reporting issues.” They further found that “stock ownership by independent audit committee directors is positively associated with earnings management. The monitoring benefits of independent directors seem to be eroded in situations where they are given stock ownership.” Yang and Krishnan also concluded that “the average tenure of audit committee directors is negatively associated with quarterly earnings management suggesting a possible positive effect of experience with the firm and its accounting.”

Abbott et al. (2004) was very early to investigate the impact of certain audit committee characteristics on the likelihood restatement. They used 88 firms between 1991 and 1999 that restated annual financial statements to study the association between audit committee independence, financial expertise, minimum audit committee size and the likelihood of restatement. They found there was a negative association between the independence and activity level of the audit committee and the occurrence of restatement. There was also a significant negative association between an audit committee with at least one member with financial expertise and restatement.

Cohen et al. (2014) proposed that the industry expertise of the audit committee member could impact the financial reporting quality arguing that accounting guidance, estimates, and oversight of the external auditor are often linked to a company’s operations within a particular industry. Cohen et al. (2014) suggested that the audit committee members who are both accounting and industry experts would perform better than those with only accounting expertise. And they further found that, in certain instances, the supervisory experts possessing the industry experts at the same time would perform better than AC members with supervisory expertise alone. They finally concluded that industry expertise combining with accounting expertise would improve the effectiveness of the audit committee in monitoring the financial reporting process.

Shin et al. (2014) examined how the audit committee expertise level could affect the companies’ capability to detect early accounting errors in their statements. They found there were negative relationships between the restatement period and audit committee members with accounting expertise, non-accounting financial expertise, supervisory expertise, and industry expertise. More particularly, Shin et al. (2014) found that supervisory expertise is incrementally beneficial to an audit committee in that it reduces the length of the restatement period.

In this paper, I fill gap in the literature by studying the association between the combined accounting financial expertise and industry expertise, the combined supervisory financial expertise and industry expertise and the number of restatements announced during 2004 for selected 26 firms listed on NYSE.

I provide evidence from regression analysis that if the audit committee has at least one member having the combined accounting financial expertise and industry expertise, the firm would have less number of financial restatements. I also provide evidence from regression analysis that if the audit committee has at least one member having the combined supervisory financial expertise and industry expertise, the firm would have less number of financial restatements.

The remainder of the paper is organized as follows: The next section will review prior literature and develops hypotheses. The sample selection, methodology, research design would be discussed in section 3. The final section will detail the results discussion and summarize the study.

2. Literature review and hypothesis development

One of the prime objective of the Sarbanes-Oxley Act enacted in 2002 and changes made to stock exchange listing standards is to improve the quality of financial reporting. The audit committee of the board for all the public listed company is charged with the authority and responsibilities to oversee the whole financial reporting and disclosures process. The primary responsibilities of the audit committee are to monitor the financial reporting quality and ensure the integrity of the financial statements. The composition and characteristics of the audit committee such as their expertise and experience would thus be very important in determining quality of the filed financial statements.  DeZoort et al. (2002) establish several critical points for the audit committee to ensure their effectiveness in overseeing high quality financial reporting. This paper will only focus on the audit committee member expertise and study the association between those expertise and the likelihood of financial restatement under their oversight.

The final rules adopted by the SEC in 2003 mandate that at least one member of the audit committee need to be the audit committee financial expert (ACFE).  The final rules define the ACFE as a person who has the following attributes: “an understanding of generally accepted accounting principles and financial statements; the ability to assess the general application of such principles in connection with the accounting for estimates, accruals and reserves; experience preparing, auditing, analyzing or evaluating financial statements that present a breadth and level of complexity of accounting issues that are generally comparable to the breadth and complexity of issues that can reasonably be expected to be raised by the registrant’s financial statements, or experience actively supervising one or more persons engaged in such activities; an understanding of internal controls and procedures for financial reporting; and an understanding of audit committee functions.”(SEC Releases 33-8177 (2003)).  This final rule would include people who have direct and explicit experience as a principal financial officer, principal accounting officer, controller, public accountant or auditor or experience in one or more positions that involve the performance of similar functions which I would classify as accounting financial expert (AFE). The final rule also would qualify people who have “experience actively supervising a principal financial officer, principal accounting officer, controller, public accountant, auditor or person performing similar functions and experience overseeing or assessing the performance of companies or public accountants with respect to the preparation, auditing or evaluation of financial statements.” I would categorize those people as non-accounting financial expert (FFE) and supervisory financial expert (SFE). Companies need comply with their annual reports for fiscal years ending on or after July 15, 2003. Small business issuers must comply with their annual reports for fiscal years ending on or after December 15, 2003. I thus chose 2004 for my study since this is the first year that this requirement would be effective for all issuers.

Earlier studies focused more on the impact of the financial expertise of the audit committee on the financial reporting quality (Abbott et al. (2004), Krishnan et al. (2008) and Dhaliwal et al. (2010)) and the impact of AC on earning management (Klein (2002)). Abbott et al. (2004) provided evidence that there was positive association between the accounting financial expertise of the audit committee on the quality of financial reporting. They found at least one member possessing the financial expertise would greatly reduce the likelihood of financial statement restatement. However, Abbott et al. (2004) did not separate the difference of AFE and SFE as defined in the SEC final rule (2003).

Krishnan and Visvanathan (2008) was able to separate the audit committee members based on their accounting financial expertise and non-accounting financial expertise. They found that there was a positive association between the accounting financial expertise and conservatism. They did not find the same association between nonaccounting financial experts or nonfinancial experts and the conservatism. However, their evidence showed that the positive association would only stand when there was a very strong board governance. In a separate paper by Dhaliwal et al. (2010), it was found that the accounting financial expertise was also positively associated with conservatism and accruals quality. Similar to the results in Krishnan and Visvanathan (2008), Dhaliwal et al. (2010) also did not find evidence for the positive association for audit committee with only SFE.

All the literature above had only focused on the financial expertise of the audit committee. They provided significant evidence of the positive association between the AFE and the financial reporting quality but did not take the industry expertise and industry experience of the audit committee into consideration. There had been extensive studies on the auditor industry expertise on the quality of financial reporting. Romanus et al. (2008) documented evidence that there was a negative association between the auditor industry specialization and the financial restatements. Moreover, Reichelt and Wang (2010) documented evidence for a positive association between auditor industry specialization and earnings quality using discretionary accruals as the measure. This showed that the industry specialization would be very valuable in preventing the financial restatements and detecting potential misstatements at a very early stage during the auditing process. Since the auditor with industry specialization or industry expertise work more effectively for the auditing of financial statements, the audit committee member who possessed the industry expertise in the same industry for the firm they are directors might also be very effective in fulfilling their oversight and monitoring of the financial statements. If the audit committee member had the industry expertise from his current or previous employment, the member would be more effective in monitoring and understanding the industry specific accounting practice. Cohen et al. (2014) provided evidence showing the existing of negative association between the industry expertise of audit committee member and the financial statement restatement. The industry expertise on top of the accounting financial expertise from an audit committee member should greatly help him or her to detect the accounting errors at the early stage by utilization of their industry knowledge of industry specific accounting issues.

Shin et al. (2014) showed evidence through regression analysis that there were negative associations between restatement period and audit committee members possessing accounting expertise, non-accounting financial expertise, supervisory expertise, and industry expertise. More particularly, Shin et al. (2014) documented that supervisory expertise was actually incrementally beneficial to an audit committee in that it would effectively the length of the restatement period for the reasons those members had the big pictures in mind when monitoring the financial reporting.

From all stated above, I developed two hypotheses as follows:

H1: There is a negative relation between the audit committee with both accounting financial expert and industry expert and the numbers of financial reporting restatement.

H2: There is a negative relation between the audit committee with both supervisory financial expert and industry expert and the numbers of financial reporting restatement.

3. Sample selection and hypothesis test

My initial source of the financial restatements was obtained from the Government Accountability Office with GAO publications GAO-06-678. The file can be found at the link in the reference (GAO-06-1079SP). It included 1390 restatement announcements between July 1, 2002 and September 30, 2005. Table 1 shows the sample selection process. For the scope of this paper, I will only focus the restatement announced from January 1, 2004 to December 31, 2004. The reason is that 2004 is the first full year that all issuers regardless of size was mandated to disclose if they had at least one audit committee financial expert.  I start with 370 financial restatement announcements. I then eliminate 236 restatements to focus only on the firms listed at NYSE (New York Stock Exchange). Those listed on AMEX, NASDAQ, and OTC are eliminated.  Another 42 restatements are dropped since no reasons were given in the database on why the firms restated their statements. I eliminate 2 more for the reason of merger and acquisition.  I further drop 3 restatements related to security related matters. I eliminate 4 restatements for other reasons stated in the database from GAO publication. I then eliminate 7 restatements because the related firms made two restatements in the same year and the firms were duplicated. I will address how many restatements a firm announced from January 1, 2004 to December 31, 2004 using the SEC EDGAR filing database. Lastly, I eliminate 50 firms because those firms either turned private or were acquired by parent company outside U.S. or I cannot find the restatement filed on SEC EDGAR during the year 2004. I finally have 26 firms left for this study. I then used the SEC EDGAR database to look through the proxy statements filed in 2004 for each firm for the composition of their audit committee with the financial expertise and the industry expertise.  Table 2 shows the industry distribution of the selected firms based on the two digits SIC code. We can see that the chemical and pharmaceuticals, communications, industrial and commercial machinery and computer equipment, and electric and gas services have the most firms with restatements announcements.

[Insert Table 1]

[Insert Table 2]

I use a regression model to study the impact of audit committee expertise on the number of restatements filed in 2004. I regress the total number of restatements in 2004 (RESTAT) on audit committee member characteristics and a few control variables. The regression model is listed as follows:

AFE (Accounting financial expert) is a dummy variable that equals 1 if at least one member of the audit committee qualified as accounting expert before 2004 and 0 otherwise. The designated accounting expert need have direct and explicit accounting or auditing experience (Carcello et al. (2006), Dhaliwal (2010), Shin et al. (2014)). Accounting financial expert need to have the experience of CPAs, CFOs, controllers, treasurers, auditors (internal or external). The information including FFE, SFE, INDE, AFEINDE, and SFEINDE will be acquired from the biographs of the audit committee members in the firm’s proxy statement in 2004.

FFE (non-accounting financial expert) is a dummy variable that equals 1 if at least one member of the audit committee has non-accounting financial expertise before 2004 when the firm restated the financial statements and 0 otherwise. A non-accounting financial expert include audit committee members having experience as managerial director in investment banking or venture capital firm, CFAs, VPs with finance responsibilities, or an investment consultant.

SFE (supervisory financial expert) is a dummy variable that equals 1 if at least one member of the audit committee has no explicit and direct experience prior to joining the committee and 0 otherwise. An audit committee member who has been a CEO or president of another firm would qualify for this.

INDE (industry expert) is a dummy variable that equals 1 if at least one member of the audit committee has the experience in similar industry before 2004 and 0 otherwise. Following Cohen et al. (2011) and Shin et al (2014), the industry expert need have the experience in a firm with same two digits SIC code as the firm with financial restatements.

AFEINDE (accounting financial expert and industry expert) is a dummy variable that equals 1 if at least one member of the audit committee is accounting financial expert (AFE) and industry expert (INDE) at the same time before 2004 and 0 otherwise.

SFEINDE (supervisory financial expert and industry expert) is a dummy variable that equals 1 if at least one member of the audit committee is supervisory financial expert (SFE) and industry expert (INDE) at the same time before 2004 and 0 otherwise.

TENU is the total years of longest serving audit committee member.

STKCOM is a dummy variable that equals 1 if there is stock compensation for the board directors in 2004 and 0 otherwise.

ACT is the activity level of the audit committee. It was the total average number of meetings held per quarter by the audit committee in 2003.

NONAUD is the ratio of the non-audit related fee paid to the external auditor over the audit fee and audit-related fees paid to the external auditor in 2003 and 2002.

The dummy variables AFE, FFE, SFE, INDE, AFEINDE, and SFEINDE is to test each of those audit committee expertise characteristics that affect the total number of restatements filed in 2004. Previous and earlier study showed evidence that expertise like AFE, FFE, SFE, and INDE in the audit committee would provide more effective internal control and higher quality of financial reporting as measured discretionary accruals. Those variables were proved effective in reducing earnings management and increasing accruals quality (Abbott et al. (2004); Carcello et al. (2006); Zhang et al. (2007); Dhaliwal et al. (2010)). I would also expect to see a negative association between accounting expertise (AFE, FFE, SFE, INDE) and the total number of restatements filed in 2004. I also expect there would be a negative association between combined accounting expertise and industry expertise and the total numbers of the restatements.

I also have four audit committee control variables (TENU, STKCOM, NONAUD, and ACT). Those control variables of audit committees could have an impact on the oversight of the audit committee in the financial reporting process. Earlier studies documented evidence that those control variables would increase the financial reporting quality and consequently reduce the chance of restatements. Abbott et al. (2004) showed that a negative association between the audit committee independence and activity level and the likelihood of filing restatements. Xie et al. (2003) documented a negative relation between audit committee meeting frequency and earnings management. I also expect a negative association between ACT and RESTAT.

[Insert Table 3]

4. Results and discussion

Table 3 is the summary of the descriptive statistics of the dummy (Panel A) and non-dummy (Panel B) variables used in this study.  The sample mean (median) of the total number of restatement files in year 2004 is 2.73 (2.50) with a standard deviation of 1.56. The number of the restatement filed in 2004 ranges from 1 to 5 times. For one firm, it filed the amendments to each quarterly statements 10Q and the annual financial statements 10K. The sample mean (median) for the longest sitting years of AC members is 10.385 (10) with a standard deviation of 6.72 years. It is also varied widely. It is also worth looking at the statistics for the ratio of non-auditing fee to auditing related fee for the firms in 2003 and 2002. The mean (median) of this ratio is 0.43 (0.26) with the standard deviation of 0.471. The minimum ratio is 0.078 with the maximum ratio of 2.23. The ratio of 2.23 was very surprising and it told us the firm paid non-auditing fee twice more than the audit and audit-related fee. The average of the AC meeting averaged per quarter ranged from 0.25 to 11.75 which also warrant the investigation into that.

With regard to the statistics for the dummy variables, the mean for the firms with at least one AC member possessing the accounting financial expertise (AFE), non-accounting financial expertise (FFE), supervisory financial expertise (SFE), industry expertise (INDE) are 61.5%, 85%, 96%, and 50% respectively. However, the mean was only 8% for the firms with at least one AC member possessing combined AFE and INDE. The mean for the firms with at least one AC member possessing combined SFE and INDE is 42%. With all this data, it is critical to look at the association of the above mentioned various expertise and the total number of restatements filed in 2004 for the selected firms.

[Insert Table 4]

Table 4 shows the correlation analysis between all the variables used in the study. All the variables were defined earlier in the paper and also noted below the table. The total number of restatement filed in 2004 including the quarterly statement restatements and annual restatements has positive correlations with non-audit fee to audit fee ratio, the stock compensation for the AC members, and the tenure of the longest sitting AC member. It was consistent with earlier literature. It also has negative correlations with the supervisory financial expertise and the combined accounting financial expertise and industry expertise. This suggest that supervisory financial expertise and the combined the accounting and industry expertise could prevent the misstatement or the number of potential restatements of the financial statements. However, it is noted that there is positive correlation between the number of restatement and other accounting expertise characteristics for the AC characteristics. It was surprising to me and I need look deeper to see what is the reason behind it. It might be that sample size is too small and I also need include more control variables for the firm characteristics, the external auditor variables.

[Insert Table 5]

Table 5 presents the regression test results on the impact of various audit committee expertise on the total number of the restatements filed in 2004. It is shown that there is negative relation between the number of restatements and the combined accounting financial expertise and industry expertise (AFEINDE), and also the combined supervisory financial expertise and the industry expertise (SFEINDE). This regression results are consistent with my expectation from both hypothesis 1 and hypothesis2. However, I need to point it out that the significance level is not high in both cases. Surprisingly, the coefficients of all other individual accounting expertise variable are positive. Even though the significance level is not high, all the individual accounting expertise characteristics (AFE, FFE, SFE) and individual industry expertise (INDE) individually will not be able to significantly reduce the total number of restatements using the data from the selected sample I have. The small sample size and the lack of other control variables featuring the firm characteristics might contribute to the results that I obtain.

Aside from the sample size and sample selection, the regression results suggest that if there is at least one audit committee member having the accounting financial expertise in the same industry using the two-digits SIC code, the audit committee then could effectively reduce the number of restatements that the firm they sit on might have. It is not surprising since those combined expertise will enable that audit committee member to have better knowledge of accounting standards, the financial reporting process, and the specific industry practice for fulfilling their oversight responsibilities.  Similarly, the combined supervisory financial expertise and industry expertise will also effectively enable the audit committee member in monitoring the financial reporting process.

5. Conclusion and future work

I test the relations and association between the total numbers of restatements filed in 2004 for selected firms and several audit committee expertise characteristics using correlation analysis and regression analysis. As expected, there is negative association between the total number of restatements filed in 2004 and the combined accounting expertise and industry expertise. The regression results indicate that if at least one audit committee member having the combined accounting financial expertise and industry expertise, or the combined supervisory financial expertise and the industry expertise, the audit committee would effectively reduce the total number of restatements. Surprisingly, the results show a positive association between the total number of restatements and only accounting expertise or only industry expertise. The results provide evidence that accounting expertise with supplement of industry expertise would better serve the audit committee in its effectiveness in ensuring the integrity of the financial reporting process and the resulting financial statement delivered to the related stakeholders. For future work, I would propose to test the hypotheses on much larger sample size and also the controls of the firm characteristics, the auditor characteristics, and other audit committee characteristics which might make a difference on the results.

Table 1. Number of restatement announcements in the sample

Table 2. Restatement Sample observations by Two-Digit SIC Code

Table 3. Panel A: Descriptive Statistics of dummy variables

Table 3. Panel B: Descriptive Statistics of control variables

RESTAT = total number of financial statements amended or restated in 2004 via SEC Edgar

AFE = 1 if at least one member of the audit committee qualified as accounting expert before 2004 and 0 otherwise.

FFE = 1 if at least one member of the audit committee has non-accounting financial expertise before 2004 when the firm restated the financial statements and 0 otherwise.

SFE = 1 if at least one member of the audit committee has no explicit and direct experience prior to joining the committee and 0 otherwise.

INDE = 1 if at least one member of the audit committee has the experience in similar industry before 2004 and 0 otherwise.

AFEINDE = 1 if at least one member of the audit committee is accounting financial expert (AFE) and industry expert (INDE) at the same time before 2004 and 0 otherwise.

SFEINDE = 1 if at least one member of the audit committee is supervisory financial expert (SFE) and industry expert (INDE) at the same time before 2004 and 0 otherwise.

TENU is the total years of longest serving audit committee member.

STKCOM is a dummy variable that equals 1 if there is stock compensation for the board directors in 2004 and 0 otherwise.

ACT is the activity level of the audit committee. It was the total average number of meetings held per quarter by the audit committee in 2003.

NONAUD is the ratio of the non-audit related fee paid to the external auditor over the audit fee and audit-related fees paid to the external auditor in 2003 and 2002.

Table 4. Correlation Analysis of restatement times and audit committee characteristics

RESTAT = total number of financial statements amended or restated in 2004 via SEC Edgar

AFE = 1 if at least one member of the audit committee qualified as accounting expert before 2004 and 0 otherwise.

FFE = 1 if at least one member of the audit committee has non-accounting financial expertise before 2004 when the firm restated the financial statements and 0 otherwise.

SFE = 1 if at least one member of the audit committee has no explicit and direct experience prior to joining the committee and 0 otherwise.

INDE = 1 if at least one member of the audit committee has the experience in similar industry before 2004 and 0 otherwise.

AFEINDE = 1 if at least one member of the audit committee is accounting financial expert (AFE) and industry expert (INDE) at the same time before 2004 and 0 otherwise.

SFEINDE = 1 if at least one member of the audit committee is supervisory financial expert (SFE) and industry expert (INDE) at the same time before 2004 and 0 otherwise.

TENU is the total years of longest serving audit committee member.

STKCOM is a dummy variable that equals 1 if there is stock compensation for the board directors in 2004 and 0 otherwise.

ACT is the activity level of the audit committee. It was the total average number of meetings held per quarter by the audit committee in 2003.

NONAUD is the ratio of the non-audit related fee paid to the external auditor over the audit fee and audit-related fees paid to the external auditor in 2003 and 2002.

Table 5. Regression analysis results of association between the total number of restatements and audit committee expertise characteristics

AFE = 1 if at least one member of the audit committee qualified as accounting expert before 2004 and 0 otherwise.

FFE = 1 if at least one member of the audit committee has non-accounting financial expertise before 2004 when the firm restated the financial statements and 0 otherwise.

SFE = 1 if at least one member of the audit committee has no explicit and direct experience prior to joining the committee and 0 otherwise.

INDE = 1 if at least one member of the audit committee has the experience in similar industry before 2004 and 0 otherwise.

AFEINDE = 1 if at least one member of the audit committee is accounting financial expert (AFE) and industry expert (INDE) at the same time before 2004 and 0 otherwise.

SFEINDE = 1 if at least one member of the audit committee is supervisory financial expert (SFE) and industry expert (INDE) at the same time before 2004 and 0 otherwise.

TENU is the total years of longest serving audit committee member.

STKCOM is a dummy variable that equals 1 if there is stock compensation for the board directors in 2004 and 0 otherwise.

ACT is the activity level of the audit committee. It was the total average number of meetings held per quarter by the audit committee in 2003.

NONAUD is the ratio of the non-audit related fee paid to the external auditor over the audit fee and audit-related fees paid to the external auditor in 2003 and 2002.

References

Abbott, L., S. Parker, and G. Peters. “Audit committee characteristics and restatements.” Auditing: A Journal of Practice & Theory 23 (2004): 69-87.

Carcello, Joseph V., Carl W. Hollingsworth, April Klein, and Terry L. Neal. “Audit Committee Financial Expertise, Competing Corporate Governance Mechanisms, and Earnings Management.” Unpublished paper, University of Tennessee, 2006. Available at https://ssrn.com/abstract=887512 or http://dx.doi.org/10.2139/ssrn.887512

Cohen, Jeffrey R., Udi Hoitash, Ganesh Krishnamoorthy, and Arnold M. Wright. “The Effect of Audit Committee Industry Expertise on Monitoring the Financial Reporting Process.” The Accounting Review 89 (2014): 243-273.

DeZoort, F.T., D. R. Hermanson, D. S. Archambeault, and S. A. Reed. “Audit committee effectiveness: a synthesis of the empirical audit committee literature.” Journal of Accounting Literature 21 (2002): 38-75.

Dhaliwal, D. S., V. Naiker, and F. Navissi. “The association between accruals quality and the characteristics of accounting experts and mix of expertise on audit committees.” Contemporary Accounting Research 27 (2010): 787-827

Klein, A. ‘Audit committee, board of director characteristics, and earnings management.’ Journal of Accounting and Economics 33 (2002): 375-400.

Krishnan, G. V., and G. Visvanathan. “Does the SOX definition of an accounting expert matter? The association between audit committee directors accounting expertise and accounting conservatism.” Contemporary Accounting Research 25 (2008): 827-858.

Romanus, R. N., J. J. Maher, and D. M. Fleming. “Auditor industry specialization, auditor changes and accounting restatements.” Accounting Horizons 22 (2008): 389-413.

Reichelt, K., and D. Wang. “National and office-specific measures of auditor industry expertise and effects on audit quality.” Journal of Accounting Research 48 (2010): 647-686.

Sarbanes-Oxley Act. Section 407. 2002, Available at: http://www.sarbanes-oxley-act.biz/SarbanesOxleySection407.htm

U.S. Securities and Exchange Commission, Releases 33-8177, 2003, Available at: https://www.sec.gov/rules/final/33-8177.htm

Shin, H., R. Z. Xu, M. Lacina, and J. Zhang. “Audit Committee Expertise and Early Accounting Error Detection: Evidence from Financial Restatements.” Journal of Forensic & Investigative Accounting 6 (2014): 181-221.

U.S. Government Accountability Office. GAO-06-678. 2006, Available at http://www.gao.gov/special.pubs/gao-06-1079sp/file1.html

U.S. Securities and Exchange Commission Edgar Filers, Available at https://www.sec.gov/edgar.shtml

Xie, B., W. N. Davidson, and P. J. Dadalt. “Earnings management and corporate governance: The role of the board and the audit committee.” Journal of Corporate Finance 9 (2003): 295-316.

Yang, J., and J. Krishnan, “Audit committees and quarterly earnings management.” International Journal of Auditing 9 (2005): 201-219.

Zhang, Y., J. Zhou, and N. Zhou, “Audit committee quality, auditor independence, and internal control weaknesses.” Journal of Accounting and Public Policy 26 (2007): 300-327