行政院國家科學委員會專題研究計畫 期末報告
內控缺失查核與內控缺失揭露可信度之研究
計 畫 類 別 ： 個別型 計 畫 編 號 ： NSC 100-2410-H-004-047- 執 行 期 間 ： 100 年 08 月 01 日至 101 年 08 月 31 日 執 行 單 位 ： 國立政治大學會計學系 計 畫 主 持 人 ： 周玲臺 計畫參與人員： 碩士班研究生-兼任助理人員：李奕萱 博士班研究生-兼任助理人員：邱獻良 報 告 附 件 ： 出席國際會議研究心得報告及發表論文 公 開 資 訊 ： 本計畫涉及專利或其他智慧財產權，2 年後可公開查詢中 華 民 國 101 年 11 月 30 日
中 文 摘 要 ： 沙賓法 404 條要求管理階層耗費心力地製作內部控制評估報 告，及會計師對於內部控制加以查核。雖然過去研究指出遵 行沙賓法 404 條的成本相當昂貴，但對於沙賓法 404 條是否 能提昇內部控制揭露的品質，卻沒有一致的結論。由於施行 沙賓法時，關於內部控制揭露及內部控制查核之各項管制採 行依時間、依公司規模漸近式實施，藉由分析受不同程度管 制的公司，本研究探討不同管制程度是否對於內部控制揭露 的品質有不同的影響。研究結果指出受到完整 404 條管制的 公司，相對於僅受 404(a)條或是 302 條管制的公司，能夠發 佈較佳品質的內部控制揭露。然而本研究並未發現僅受到 404(a)條管制的公司能夠比僅受 302 條管制的公司發佈較佳 品質的內部控制揭露的證據。本研究的實證證據能作為管制 機關在將來制訂內部控制報導之參考。 中文關鍵詞： 沙氏法管制、內部控制揭露品質、內部控制查核，重編 英 文 摘 要 ： Although prior studies indicate that compliance with
Section 404 of SOX, which requires both high-effort management disclosure and internal control audit, is costly, no consistent conclusion that Section 404 improves the disclosure quality of internal control over financial reporting (ICFR) can be drawn from prior research . By analyzing incremental and joint implementation of multiple SOX-based ICFR disclosure and internal control audit mandates, we examine the differential effects of alternative ICFR regulations on the quality of ICFR related public disclosure. Our results reveal that companies subject to full Section 404 issue higher-quality internal control reports than those subject to Section 302 or Section 404(a) only. However, we find no supporting evidence that companies subject to Section 404(a) issue better quality internal control reports than those subject to Section 302 only. This study provides evidence for policymakers to assess the effectiveness of internal control reporting regulations in the future.
英文關鍵詞： SOX regulation, quality of internal control disclosure, internal control audit, restatement
I
ICFR Disclosure Quality under Different
SOX Regimes
Hsien-Lian Chiu Department of Accounting
College of Commerce National Chengchi University
Phone: +886934272115 E-mail: [email protected]
Ling-Tai Lynette Chou Department of Accounting
College of Commerce National Chengchi University Phone: +886-2-29393091 ext. 81245
E-mail: [email protected]
November 2012
1. We thank helpful comments from anonymous reviewers of 2012 AAA Section, Annual, and Northeastern Region Meetings and workshop participants at National Chengchi University. Research funding provided by NSC, Taiwan (100-2410-H-004-047) is greatly appreciated. 2. The data used in this study is extracted from AuditAnalytics and Compustat whose access is
limited to eligible users. To follow the terms of use, we cannot transfer or distribute the data to any other party.
II
Abstract
Although prior studies indicate that compliance with Section 404 of SOX, which requires both high-effort management disclosure and internal control audit, is costly, no consistent conclusion can be drawn from prior research that Section 404 can improve the disclosure quality of internal control over financial reporting (ICFR). By analyzing incremental and joint implementation of multiple SOX-based ICFR disclosure and internal control audit mandates, we examine the differential effects of alternative ICFR regulations on the quality of ICFR related public disclosure. Our results reveal that companies subject to full Section 404 issue higher-quality internal control reports than those subject to Section 302 or Section 404(a) only. However, we find no supporting evidence that companies subject to Section 404(a) issue better quality internal control reports than those subject to Section 302 only. This study provides evidence for policymakers to assess the effectiveness of internal control reporting regulations in the future.
Keyword: SOX regulation, quality of internal control disclosure, internal control audit,
1
ICFR Disclosure Quality under Different
SOX Regimes
1. INTRODUCTION
This study examines the differential effects of the alternative regulations of internal control over financial reporting (ICFR) on the quality of ICFR related public disclosures. In order to restore investors’ confidence after a number of high-profile accounting scandals, the Congress in 2002 passed the Sarbanes-Oxley Act (SOX), one of whose main goals is to improve the reliability of financial information quality by strengthening the assessment and reporting process of ICFR. Section 302 of SOX, effective on August 29, 2002 (SEC 2002), requires management to self-evaluate and conclude on the effectiveness of ICFR in periodic reports. Section 404, effective on November 15, 2004, requires that (1) company management report the effectiveness of internal control structure and procedures based on a formal assessment at the end of each fiscal year (Section 404(a)) and (2) their assessment be attested by external auditors (Section 404(b)).
Many executives contended against high compliance costs from Section 404 regulations1 and a series of hearings were held to address concerns over disproportionately high compliance costs for smaller issuers (Cox 2007). To reduce compliance burden for small companies, Section 404 was implemented incrementally over time and bifurcated by size of public companies. Compliance with Section 404(a) was deferred until December 15, 2007 for non-accelerated filers (companies whose
1
Collins, Pete, “Senior Executives Divided on Cost of Complying with Sarbanes-Oxley Act,” PWC Management Barometer, July 2, 2003,
2
public float is less than 75 million as of six months before the fiscal year-end) and they were granted extensions of exemption from internal control audit six times. In 2009 and 2010, there were also keen debates as to whether a permanent exemption of Section 404(b) for non-accelerated filers should be included in the Dodd-Frank Wall
Street Reform and Consumer Protection Act, part of the Obama financial regulatory
reform plan of 2009. The final bill passed in the mid of 2010 by the Congress rules that non-accelerated filers do not have to comply with Section 404(b) requirement. The incremental and bifurcated implantation of Section 404 is illustrated in Figure 1. Since compliance with the requirements of Section 404 is costly, especially Section 404(b) (Hartman 2007; Eldridge and Kealey 2005; Kinney and Shepardson 2010), we address a crucial research question in this study: whether the burdensome requirements can improve the quality of public ICFR disclosure.
[Insert Figure 1 here]
One purpose of the internal control disclosure requirements is to provide financial statement users with a warning of potential accounting problems that could result from weak internal controls as well as the likelihood that financial statements are of questionable reliability and may be restated later due to internal control weaknesses. Despite the fact that the ICFR disclosure quality cannot be either observed or directly measured, we argue that high quality internal control disclosures should exhibit two features: (1) offering credibility when associated with clean internal control reports and (2) demonstrating relevance for material weaknesses discovered. As to the first feature, high-quality clean internal control reports should imply actually effective internal control systems and therefore the likelihood of
3
associated financial statements to be restated is expected to be lower. As to the second feature, disclosed material weaknesses should be relevant and severe enough to remind market participants of higher likelihood of financial restatements.
In some cases, only one of the two features can be achieved. For example, if the management and/or auditors are too conservative, they tend to overstate the severity of internal control problems and unduly conclude the internal control systems as ineffective. Their disclosed material weaknesses are less likely to be followed by subsequent restatements. However, suppose that the said management and/or auditors issue clean internal control reports, investors might become more confident about the quality of financial information. On the other hand, management may intentionally withhold the unfavorable information that there exist internal control problems in order not to decrease their equity-based compensation or damage their job security, or they may not be competent enough to perform adequate and sufficient internal control assessment procedures. In this case, their clean internal control disclosures are not credible while their disclosed material weaknesses can be a strong indicator of potential upcoming restatements. Even though the quality of ICFR public disclosures cannot be observed, we measure the quality in our research design by examining the afore-mentioned two features of public ICFR reports.
We compare the relative disclosure quality of ICFR reports issued by companies under different regulation regimes. Due to the incremental and bifurcated implementation of Section 404, we can identify three non-overlapping groups of U.S. companies which are respectively subject to one of the three levels of SOX regulations of ICFR (in descending order):
(1) Full Section 404 – “accelerated filers” (companies whose public float is 75 million or more as of six months before the fiscal year-end) after November
4
15, 2004: Those companies are subject to Section 302, Section 404(a) and Section 404(b). That is, managers of those companies need to make a high effort preparing ICFR reports and ICFR audit is required.
(2) Section 404(a) – non-accelerated filers after December 15, 2007: Those companies are subject to Section 302 and Section 404(a). Only high management effort to prepare ICFR reports is required.
(3) Section 302 – all public companies prior to November 15, 2004 and non-accelerated filers in the period of November 15, 2004–December 15, 2007: Compliance of only Section 302 is required for this group. Neither high effort management ICFR reports nor ICFR audits are mandated.
Our evidence shows that full Section 404 can improve disclosure quality of internal control reports. Given clean ICFR reports, companies subject to full Section 404 are less likely to restate than those of Section 302 and Section 404(a) groups. If the conclusion is ineffective ICFR, companies of the full Section 404 group are more likely to restate than those of Section 302 or Section 404(a) group. Companies under full Section 404 regime are more capable of providing assurance (advance warning) of the reliability of their financial statements if they conclude their ICFR as effective (ineffective). In the additional test, we also find ICFR reports are of better disclosure quality if attested by brand name auditors. On the other hand, we find no supporting evidence that Section 404(a) only can improve the reporting quality of internal control disclosures. Given clean ICFR reports, the likelihood of financial restatements for Section 404 (a) group and that for Section 302 group is not significantly different. Similarly, given disclosed weak controls, the likelihood of restatements between the two groups is not significantly different either. In sum, the results indicate that companies subject to full Section 404 regulation can provide higher-quality internal
5
control reports than those subject to Section 302 or Section 404(a) regulation. The robustness tests indicate our results are not sensitive to size effect and the choices of measurement timing of financial restatements.
Our study makes several contributions. First, while most studies which investigate disclosure quality focus on financial statements, this paper contributes to the contemporaneous literature of the quality of public ICFR disclosures. By developing a measurement method based on the relation between disclosed ICFR effectiveness and financial restatements, we compare the relative quality of ICFR disclosures under differential ICFR regulation regimes. Second, we provide findings regarding whether auditor attestation can improve the quality of internal control disclosure. Although prior research suggests that financial audit can enhance the credibility of financial reports (Healy and Palepu 2001), there is little research investigating whether auditor attestation also has positive effects on the reporting quality of internal control disclosures. This study provides evidence that external auditor attestation can improve the quality of internal control disclosure. Finally, while non-accelerated filers are not required to comply with Section 404(b) according to the Dodd-Frank Wall Street Reform and Consumer Protection Act, the Government Accountability Office is requested to evaluate the effects of the exemption. This study provides useful evidence to policymakers for setting up future internal control auditing and reporting rulings.
The remainder of this paper is organized as follows. The next section presents the related literature and our hypotheses. Section 3 describes our research design and sample selection. Our results are presented in Section 4. Section 5 reports the sensitivity and additional tests. Finally, Section 6 concludes our research.
6
2. RELATED LITERATURE AND HYPOTHESIS
DEVELOPMENT
2.1 Regulations of Internal Control over Financial Reporting
Prior to the passage of SOX, public companies were required to disclose information about internal controls only when a Form 8-K was filed after an auditor change. Sections 302 and 404 of SOX are the first statutory legislation that requires public companies disclose the effectiveness of internal controls. Section 302, effective on August 29, 2002 (SEC 2002), requires managers of public companies to evaluate and conclude on the effectiveness of their internal control systems in periodic reports. Section 404 which became effective on November 15, 2004 contains two subsections. Under Section 404(a) managers are mandated to prepare reports of their assessments of the effectiveness of company internal control systems in annual reporting. Under Section 404(b) public accounting firms that audit the issuers’ financial reports shall also attest to the management internal control assessments. Starting from August 27, 2007 external auditors are required to directly express an opinion on the effectiveness of internal control systems rather than attest to management assessment (SEC 2007b).
Section 404 is the most contentious part of SOX due to its high compliance costs. Some surveys and empirical evidence show that audit fees have increased substantially after Section 404 implementation (Hartman 2007; Eldridge and Kealey 2005). In order to be classified as non-accelerated filers which are exempted from compliance with Section 404, small companies intentionally kept their sizes small while large companies had incentives to become smaller (Gao et al. 2009). Some public companies also chose to go private (Engel et al. 2007) or go dark (Leuz et al. 2008) and foreign companies decided to delist from the U.S. stock markets to
7
circumvent the compliance of Section 404 (Piotroski and Srinivasan 2008).
Due to concerns over overwhelming burden of compliance with Section 404(b) on small companies, U.S. Securities and Exchange Commission repeatedly extended the deadline of compliance with Section 404(b) for non-accelerated filers. In 2009 and 2010, there were also keen debates as to whether a permanent exemption of Section 404(b) for non-accelerated filers should be granted in the Dodd-Frank Wall Street
Reform and Consumer Protection Act. The House Financial Services Committee in
the mid of 2010 finally approved an amendment to exempt small companies from Section 404(b). Although the dispute over the scope of Section 404 has been settled, understanding of the effects of different levels of regulation regime (i.e., from self-evaluation, formal assessment, to formal assessment plus an independent audit) on ICFR disclosure quality remains limited.
2.2 Related Research
Even though the objective of Section 404 is to provide meaningful disclosures to investors about the effectiveness of internal control systems (SEC 2007a), prior studies do not find consistent evidence about whether the stricter Section 404 regulation can improve the quality of ICFR reports. Some research papers suggest that the reliability of ICFR reports or disclosures may be negatively affected without auditor involvement in testing and reporting the effectiveness of internal controls (Hammersley et al., 2008; Bedard and Graham 2011; Bedard et al., 2009; Hermanson and Ye 2009). On the other hand, a few studies do not find evidence that ICFR audit provides additional benefits to internal control disclosures. Lu et al. (2010) find that public companies in Canada subject to SOX North, which is similar to Section 302, generally make credible internal control disclosures. Kinney and Shepardson (2011)
8
show that small companies which first implement unaudited internal control reports under Section 404(a) regime experience comparably significant increase in material weakness disclosure rates as first-time accelerated filers for the same year do. However, the increased disclosure rate of material weaknesses under Section 404(a) regime does not necessarily mean better quality of ICFR reports because it is uncertain whether or not the public can receive assurance (noteworthy warning) if they are provided clean (negative) ICFR reports. These studies which just take public ICFR reports at face value do not examine the disclosure quality of internal control reports. In this study, we go one step further to examine and compare the quality of ICFR reports under different regulation regimes in the view of the aforementioned two features.
Based on restating companies subject to Section 404 whose original misstatements are resulted from control weaknesses, Rice and Weber (2012) find that despite being subject to outside audit, only a small proportion of these companies disclose control weaknesses during their misstatement period. Although the purpose of both Rice and Weber (2012) and our study is to examine the disclosure quality of ICFR reports under the SOX regulation, there exist two major differences. First, Rice and Weber (2012) only focus on Section 404 and they do not compare the disclosure quality of ICFR under three levels of regulations. Our study investigates whether, relative to Section 302, stricter regulation such as Section 404(a) or full Section 404, can enhance the quality of public internal control disclosures. Second, since the sample of Rice and Weber (2012) only includes restating companies whose misstatements are linked to control weaknesses, the non-random sample may present a potential self-selection problem. We complement their study by drawing on a sample of restating and non-restating companies under three different regimes.
9
2.3 ICFR Reports and Restatements
Internal control over financial reporting is the process designed and maintained by management to provide reasonable assurance regarding the reliability of financial statements (PCAOB 2007). Transactions may not be recorded accurately or in accordance with generally accepted accounting principles if one or more internal control problems exist. Therefore, the effectiveness of a company’s internal control system is a key determinant of the reliability of financial reporting (PCAOB 2004).
Prior research finds that a weak internal control system may negatively affect earnings quality through (1) intentionally biased accruals resulted from earnings management or (2) unintentional errors in accrual estimation (Ashbaugh-Skaife et al. 2008; Doyle et al. 2007a). Moreover, when material weaknesses exist in internal control systems, there is a reasonable possibility that a material misstatement of the entity’s financial statements will not be timely prevented, or detected and corrected (PCAOB 2007; AICPA 2008). Thus, effective internal control systems can decrease the likelihood of material misstatements which lead to subsequent restatements. When management delivers faithful representation of the ICFR, the market participants should be able to infer more (less) credible financial information from effective (ineffective) ICFR reports.
2.4 The Disclosure Quality of ICFR Reports and Section 404(a)
Conceptually, to conclude internal control systems as ineffective, three conditions must be met: there exist deficiencies; managers discover internal control deficiencies and managers decide to disclose those deficiencies (Ashbaugh-Skaife et
10
al. 2007). Thus, management detection capability and disclosure incentives are important factors to high-quality ICFR reports. As to detection capability, material weaknesses may not be discovered if managers are not competent enough to apply adequate and sufficient assessment procedures of internal control systems. Section 404(a) requires management documentation which contains a formal assessment of internal controls. The assessment must be made according to publicly recognized internal controls framework, such as Committee of Sponsoring Organizations of the Treadway Commission (COSO) or Control Objectives for Information and related Technology (COBIT) framework. SEC have issued for managers a clear guidance of complying Section 404(a) (SEC 2007c) with the intent of bringing information regarding effectiveness of internal control systems into public view (SEC 2007d). The requirement of Section 404(a) may enhance detection capability of the management and mitigate the problem of incompetence of managers in examining internal control systems. Thus, we expect that internal control disclosures under Section 404(a) regime are of better quality than those under Section 302 regime. Our Hypothesis 1 is stated as follows:
Hypothesis 1: Compared with companies under Section 302 regime, companies
under Section 404(a) regime are less (more) likely to restate their financial statements even if both their internal control systems are concluded as effective (ineffective).
2.5 The Disclosure Quality of ICFR Reports and Full Section 404
Some prior studies indicate disclosure incentives of control systems are associated with firm size, corporate governance, ownership structure, and firm growth (Bronson et al. 2006; Deumes and Knechel 2008). Thus, even if internal control
11
weaknesses are actually discovered, not all managers have equal level of incentives to issue high-quality ICFR reports. Besides, disclosure of internal control weaknesses is unpleasant news to investors which leads to negative stock market reaction (Hammersley et al. 2008; Beneish et al. 2008). Management compensation and turnover is also influenced by types of internal control reports issued (Wang 2011). As a result, managers may have incentives to withhold unfavorable information related to internal control systems in order not to decrease their equity-based compensation or harm their job security.
External auditors have professional knowledge, training and experience in examining internal control systems and they can help management discover internal control problems. Most internal control weaknesses are detected by auditors rather than by management and material weaknesses are more likely to be disclosed at the fourth quarter when auditors are on-site and when auditors have experience with internal control audits (Bedard and Grahma 2008; Hammersley et al. 2008). In addition, auditors are also regarded as an effective monitoring mechanism for corporate governance (Becker et al. 1998). A survey conducted by the U.S. SEC indicates that investors regard auditor attestation of internal control as an important measure given the auditors’ expertise and independence in the evaluation of internal control systems (SEC 2009). We argue that managers may feel compelled to truthfully disclose weaknesses of their internal control systems if their disclosures need to be attested by external auditors under full Section 404 regime. Moreover, classic theories indicate that low cost signals are not credible (Spence 1973; Ross 1977; Comment and Jarrel 1991) and compliance cost under Section 404 regime is significantly higher than Section 302 and Section 404(a) regime (Kinney and Shepardson 2010). Thus, we posit that the quality of internal control disclosures will be enhanced if auditor
12
attestation is required. We state our Hypothesis 2 as:
Hypothesis 2a: Compared with companies under Section 302 regime, companies
under full Section 404 regime are less (more) likely to restate their financial statements even if both their internal control systems are concluded as effective (ineffective).
Hypothesis 2b: Compared with companies under Section 404(a) regime,
companies under full Section 404 regime are less (more) likely to restate their financial statements even if both their internal control systems are concluded as effective (ineffective).
3. RESEARCH DESIGN
3.1 Sample Selection
Our sample originates from Audit Analytics (Section 302 disclosure, Section 404 reports and restatements) and Compustat (financial information). We begin with 50,608 firm-year observations covered by both Compustat and Audit Analytics for the fiscal year 2002 through 2010. Next, we exclude foreign firms (n=692), firms which are not listed in NYSE, Nasdaq or AMEX (n=6,350), firms operating in financial sector or regulated industries (n=16,233), non-accelerated filers which voluntarily issue Section 404 reports (n=758) and firms with unavailable necessary data (n=1,659). Following the procedure, the final sample is composed of 24,916 firm-year observations.
3.2 Empirical Models and Variable Definition
13 YEAR BIGN MBR PER LEV ROA SIZE SOX EFFECTIVE a SOX EFFECTIVE SOX a SOX EFFECTIVE RESTAT 11 10 9 8 7 6 5 4 3 2 1 0 404 ) ( 404 404 ) ( 404 ) Prob(
RESTAT is our dependent variable which equals one if the company restates its
financial reports of the same year in which the internal control report is issued and zero otherwise. We exclude technical restatements and only consider financial restatements due to errors or frauds2 because they are de facto reporting failures which are supposed to be prevented, or detected and corrected beforehand by effective internal control systems. EFFECTIVE is coded 1 if a company concludes its internal control system as effective in the 10-K filing and zero otherwise. The effectiveness of internal control systems are supposed to reflect the issuer’s financial reporting quality. Therefore, we expect 1 to be significantly negative.
) ( 404 a
SOX is coded one if the firm-year observation is from Section 404(a) group, that is, non-accelerated filers after December 15, 2007 and zero otherwise.
404
SOX is coded one if the firm-year observation is from the full Section 404 group, that is, accelerated filers after November 15, 2004 and zero otherwise. We expect that clean internal control disclosures under Section 404(a) regime are more credible than those under Section 302 regime according to Hypothesis 1. 4 and 24 are expected to be significantly negative. If Section 404(a) can help provide noteworthy warnings that internal control systems are actually weak when managers claim their control systems as ineffective, we expect 2 to be significantly positive. Under Hypothesis 2a full Section 404, which includes requirement of management ICFR
2 Besides errors and frauds, restatements can be a result of changes in accounting principles. These mandated restatements are not included in our analysis because they are essentially not reporting failures.
14
reports and ICFR audit, is posited to better improve the quality of ICFR disclosures than Section 302. 5 and 35 are expected to be significantly negative and 3 to be significantly positive. Similarly, under Hypothesis 2b if companies under full Section 404 regime issue better-quality ICFR reports than those under Section 404(a) regime, we expect 32 to be significantly positive and 3524 to be significantly negative.
We include SIZE as a control variable, measured by the natural log of total assets whose unit is millions of dollars because prior studies indicate that large companies are more likely to make financial restatements (Baber et al. 2006). Prior studies (Kinney and McDaniel 1989; DeFond and Jiambalvo 1991) find that companies making less profit or having higher leverage are more likely to restate their financial statements. Thus, we include ROA and LEV as control variables. ROA equals net income before extraordinary items divided by total assets and LEV equals total liabilities divided by total assets. We also control firm growth using market to book ratio (MBR) and price to earnings ratio (PER) because Richardson et al. (2002) and Jagadison et al. (2005) show high-growth companies are more likely to restate their financial reports. Finally, Teoh and Wong (1993) and DeFond and Jiambalvo (1993) document that brand name (Big 4) auditors can provide better audit service than non-Big 4 auditors. We include BIGN as a control variable which equals one if the auditor is a brand name auditor and zero otherwise. YEAR is a set of dummy variables representing each of the fiscal years. To mitigate the effect of possible spurious outliners, we winsorize all of the continuous variables at the 0.01 and 0.99 percentiles.
15
4. RESULTS
4.1 Descriptive Statistics
Panel A of Table 1 provides temporal distribution of financial restatements and reported ineffective internal controls. The frequency of restatements dramatically rose in 2003, 2004 and 2005. It might be because the U.S. legal environment and the capital market became more stringent after the enactment of SOX. In the meantime, the frequency of reported ineffective internal controls also dramatically rose in 2004 and 2005. The possible reason is that the mandated management internal control reports and auditor attestation drive managers to discover and disclose internal control problems.
In Table 1, Panel B and Panel C present the descriptive statistics of variables for the full sample and subsamples subject to each SOX regulation regime respectively. Approximately 93% of the observations in our sample claim their controls are effective. The percentages of effective control systems for companies under Section 302, Section 404(a) and full Section 404 regimes are 95%, 82%, and 93%, respectively. Due to the incremental and bifurcated implementation of Section 404, it is not surprising that companies under full Section 404 regime are on average larger than those under Section 302 and Section 404(a) regimes. We notice that firms of Section 404(a) group make less profit than those of full Section 404 group (ROA of -42.77% versus -0.23%; p-value of 0.00). In addition, companies under Section 404(a) regime use more leverage and their P/E and M/B ratios are lower than those under full Section 404 regime. Finally, accelerated filers are more willing to hire brand name auditors than non-accelerated filers.
16
[Insert Table 1 here]
Table 2 presents the correlations among the variables of the full sample. Effective internal controls are negatively correlated with the occurrence of financial restatements. Except the coefficient between BIGN and SIZE, other significant coefficients of correlation are relatively small.
[Insert Table 2 here]
We employ chi-square tests to examine the relation between effective internal control systems and the likelihood of financial restatements. Among 1,815 companies which concluded ineffective internal controls in their 10-K filings, 796 (43.86%) companies restated their financial reports. On the other hand, among 23,101 companies which concluded effective internal controls, only 3,046 (13.19%) companies restated their financial statements. It shows that in general there exists a significant negative relation between the effectiveness of internal control systems and the likelihood of financial restatements (χ2≈1200, p<0.000). Besides, we find that the relation between the effectiveness of internal control systems and restatements under Section 404(a) regime (χ2=152.78, p<0.000) is no more significant than the relation under Section 302 regime (χ2=191.07, p<0.000). It does not support H1 that internal control reports under Section 404(a) regime are of better-quality than those under Section 302 regime.
Finally, we find that the magnitude of the relation between the effectiveness of internal control systems and financial restatements under full Section 404 regime (χ2≈1300, p<0.000) is stronger than that under both Section 302 and Section 404(a)
17
regimes. Among companies which disclosed ineffective controls, the percentages of companies which restated financial statements are 51.59%, 26.73% and 48.70% under Section 404, Section 404(a) and Section 302 regime respectively. On the other hand, among companies which disclosed effective controls, the percentages of companies which did not restate financial statements are 89.68%, 92.22% and 80.59% under Section 404, Section 404(a) and Section 302 respectively. In sum, the requirements of full Section 404 can improve the disclosure quality of internal control reports, supporting H2a and H2b.
4.2 Multivariate Empirical Results
Table 3 presents the logistic regression results for the base model and our main model in column 1 and 2 respectively. The Hosmer-Lemeshow tests show that both of the models have an appropriate fit (p=0.342 and p=0.400, respectively). For the base model, the significantly negative coefficient on EFFECTIVE demonstrates that public management ICFR reports are generally credible. Disclosed ICFR effectiveness suggests lower likelihood of financial restatements. Ceteris paribus, occurrences of effective controls decrease odds of financial restatements by 83.56%. Our result is consistent with Doyle et al. (2007a) who conclude that weak internal control environment may allow for potential unintentional clerical errors or intentional earnings management both of which result in poor earnings quality.
For our main model, the significantly negative coefficient on EFFECTIVE means that internal control disclosures under Section 302 regime are generally credible. Although the signs of the coefficients on SOX404(a) and EFFECTIVE*SOX404(a) are not consistent with our expectation, the coefficients are not significant. The joint
18
test shows that the negative sum of coefficients 2 4 conforms to our expectation but it is insignificant, either. Companies under Section 404(a) regime cannot provide more assurance (noteworthy warnings) if their conclusion as to internal controls is effective (ineffective). Thus, based on the above results we do not find supporting evidence that relative to Section 302, the requirement of Section 404(a) alone helps enhance the disclosure quality of ICFR reports.
The significantly positive coefficient on SOX 404 shows that given disclosure of ineffective internal control systems, the likelihood of restatements is higher for companies subject to full Section 404 than for companies subject to Section 302. The odds of financial restatements increase by 36.98% if the disclosures of ineffective controls are issued by companies subject to full Section 404 rather than Section 302. Users of financial reports can receive more noteworthy advance warnings of potential restatements resulted from weak controls under Section 404 regime. In addition, both
5
and 35are significantly negative, indicating given disclosures of effective controls, the likelihood of restatements is lower for companies subject to full Section 404 than for companies subject to Section 302. The odds of financial restatements decrease by 45.73% if the disclosures of effective controls are issued by companies subject to Section 404 rather than Section 302. In sum, we provide evidence that internal control reports are of better quality if they are issued by companies subject to full Section 404 than by companies subject to Section 302, supporting our H2a.
Finally, 3524 is significantly negative, indicating that compared with Section 404(a), full Section 404 which additionally requires ICFR audit can achieve higher-level credibility for disclosures of effective internal controls. The odds of financial restatements decline by 21.42% if disclosures of effective controls are issued by companies under full Section 404 regime than those under Section 404(a) regime.
19
We also find 32 is significantly positive. Compared with Section 404(a) regime, the materiality of disclosures of ineffective controls under full Section 404 regime is stronger. The odds of financial restatements increase by 53.57% if the disclosures of ineffective controls are issued by companies subject to full Section 404. We provide supporting evidence for H2b that the disclosure quality of internal control reports becomes better if the reports need to be audited.
As to control variables for firm innate characteristics, the coefficient of SIZE is significantly positive, consistent with the finding of Baber et al. (2006). Besides, the significant positive coefficient of LEV indicates that as we expected companies having higher leverage are more likely to restate their financial statements. The coefficient of MBR is significant but the sign of the coefficient is negative. Other control variables are not significantly associated with the likelihood of financial restatements.
[Insert Table 3 here]
4.3 Comparison with the Results of Prior Studies
Based on our sample, the disclosure rates of material weaknesses among Section 302, Section 404(a) and Section 404 groups are 4.64%, 18.27% and 6.6%, respectively. Consistent with Kinney and Shepardson (2011), it is shown that the implementation of Section 404(a) for non-accelerated filers can significantly increase the disclosure rate of material weaknesses. In the further analysis, however, we find that the percentages of restatement companies in those disclosing material weaknesses are 49.74%, 26.83%, 51.60% for Section 302, Section 404(a) and Section 404 groups,
20
respectively. On the other hand, the percentages of restatement companies in those disclosing no material weaknesses are 19.35%, 7.77%, 10.33% for Section 302, Section 404(a) and Section 404 groups, respectively. It is noticed that managers of non-accelerated filers in the initial years of implementation of Section 404(a) seem to overstate internal control problems. Although initial implementation of Section 404(a) can drive management to disclose internal control problems, based on our results, it cannot improve ICFR disclosure quality as much as Section 404.
Rice and Weber (2011) indicate that under Section 404 regime, only a small group of restatement firms disclose material weaknesses during their restatements periods and argue that Section 404 is ineffective. Consistent with Rice and Weber (2011) we also find only 26.11% of restatement firms disclose material weaknesses during the restatement periods for Section 404 group. However, this percentage for Section 302 group is 11.12%, which is even lower than Section 404 group (χ2≈129, p<0.000). The percentage for Section 404(a) group is 43.57%, the highest among the three groups. It may be because management inclines to overstate material weaknesses during the initial implementation years of Section 404(a). Based on aforementioned two disclosure features, our evidence shows that compared with Section 302, Section 404 which requires management ICFR report and ICFR audit, does provide incremental benefits in improving ICFR disclosure quality.
5. SESITIVITY AND ADDITIONAL TESTS
5.1 Periods of Financial Restatements
21
cannot be corrected in a short time and might lead to financial restatements of following fiscal years. Others may argue internal control problems generally have existed a long time prior to their disclosures. Thus, we alter the definition of RESTAT and code RESTAT_PS one if a company restates its current year or prior year financial reports, and zero otherwise. We also code RESTAT_SF one if a company restates its current year or next year financial reports, and zero otherwise. The results with the alternative dependent variables are presented in Table 4 and are quite similar with our main results. Our results are not sensitive to choices of the period of restatements.
[Insert Table 4 here]
5.2 Size Effects
Due to the incremental and bifurcated implementation of Section 404, SOX 404(a) and SOX 404 variables are also proxies for market values of companies. The first, second and third quartile of the market value variable in our sample is around 99.62, 373.90 and 1382.01 million, respectively, showing that our sample contains quite a few companies subject to full Section 404 that are much larger than non-accelerated filers. It is our concern that some innate characteristics of these extreme large companies are very different from those of smaller companies and these characteristics may affect the quality of ICFR disclosure. Although we have controlled for some firm characteristics in our test, it is still possible that our results are attributed to the effects of other firm characteristics that cannot be directly observed or measured.
To address the concern, we use a subsample including companies from the period 2002-2003 in which all companies are subject to Section 302. To examine if
22
the innate characteristics associated with size jointly have an effect on the ICFR disclosure quality, we control the regulation effect and interact SIZE with EFFECTIVE or INEFFECTIVE (equal 1 if a firm concludes its internal controls as ineffective) to estimate the regression models below:
YEAR BIGN MBR PER LEV ROA SIZE SIZE EFFECTIVE EFFECTIVE RESTAT 8 7 6 5 4 3 2 1 0 * ) Prob( YEAR BIGN MBR PER LEV ROA SIZE SIZE E INEFFECTIV E INEFFECTIV RESTAT 8 7 6 5 4 3 2 1 0 * ) Prob(
The result (not tabulated) shows that both coefficients on EFFECTIVE*SIZE and INEFFECTIVE*SIZE are not significant and we do not find any evidence that the quality of ICFR disclosures differs with firm size.
5.3 Impact of 2008 Financial Crisis
Section 404 (a) became effective for non-accelerated filers starting December 15, 2007 during the period of 2007–2008 financial crisis. The crisis reduced consumer wealth, increased unemployment rates and slowed down economic activities. In addition to financial institutions, the majority companies of other industries also performed quite poorly during the crisis. High proportion of our sample period for Section 404(a) non-accelerated filers overlaps the time span of the crisis. Since Doyle et al. (2007b) document that financially weaker companies are more likely to report internal control weaknesses, it may explain why we find companies under Section 404(a) regime inclined to overstate material weaknesses. Besides, it remains unknown whether companies under Section 404 regime also exhibit the same tendency during the crisis.
23
We control the effect of time periods by drawing a sample covering observations in 2007, 2008 and 2009 to compare the ICFR disclosure quality between Section 404(a) group and Section 404 group. The model below is estimated and we find 2 is significantly positive, meaning that even in the period of financial crisis, companies under Section 404 regime do not tend to overstate material weaknesses because those accelerated filers reporting ineffective internal controls still show higher likelihood of restatements than their non-accelerated counterparts. So we can find that the difference in regulation regimes plays an important role in ICFR disclosure quality even during economic downturn.
YEAR BIGN MBR PER LEV ROA SIZE SOX EFFECTIVE SOX EFFECTIVE RESTAT 9 8 7 6 5 4 3 2 1 0 404 404 ) Prob(
5.4 Auditor Attribute
Numerous studies following the theoretical foundation of DeAngelo (1981) and Dopuch and Simunic(1980) indicate that brand name auditors enhance financial reporting quality (Palmrose 1988; Becker et al. 1998; Khurana and Raman 2004; Behn et al. 2008). However, no extant research investigates the association between the quality of ICFR reports and brand name auditors. Brand name auditors may provide better-quality ICFR audits because they have more clients subject to full Section 404 and accumulate more experience in ICFR audits. Brand name auditors also have higher incentives to protect and invest in reputation (Dopuch and Simunic 1980). To address this question, we eliminate firms subject to Section 302 and Section 404(a) from our sample and keep only firms subject to full Section 404. We interact
24
BIGN with EFFECTIVE and estimate the regression model below:
YEAR MBR PER LEV ROA SIZE BIGN EFFECTIVE EFFECTIVE BIGN RESTAT 8 7 6 5 4 3 2 1 0 * ) Prob(
The results are presented in Table 5. The significantly positive coefficient of BIGN and the significantly negative coefficient of EFFECTIVE*BIGN show that the quality of ICFR reports can be enhanced through ICFR audit by brand name auditors.
[Insert Table 5 here]
5.5 Remediation Effect
Concluding the ICFR as ineffective may drive managers to remediate discovered internal control problems. Small companies whose organizations are less complex may be able to remediate control weaknesses more easily and quickly. Instead of blaming non-accelerated filers for overstating material weaknesses, it may be that companies under Section 404(a) regime remediate their control weaknesses more quickly and hence do not need to restate financial statement afterwards. Thus, ineffective ICFR systems disclosed by companies under Section 404(a) may not necessarily lead to subsequent restatements. To resolve the concern, we employ chi-square tests to see if there is a relation between regulation regimes and remediation of weaknesses. We find companies under Section 404 regime are more likely than companies under Section 404(a) regime to remediate weaknesses within one year (χ2≈91.61, p<0.000). As a result, we do not find evidence to support the concern that companies under Section 404(a) regime more quickly remediate their control weaknesses and reduce the likelihood of restatements.
25
5.6 Voluntary Regime Shift
Our sample covers firm-year observations between fiscal year 2002 and 2010 and there are two mandatory regime shifts in the sample period, i.e. implementation of Section 404 for accelerated filers in 2004 and implementation of Section 404(a) for non-accelerated filers in 2007. Except the mandatory regime shifts, some companies increase (decrease) its market float and transformed themselves into accelerated files (non-accelerated filers). The regime change may due to capital need for growth, circumvention of burdensome Section 404 and so on. There is a concern that incentives beyond the mandatory regime shift may have implications for our findings. To address the concern, we drop companies which shifted regimes voluntarily from the sample and re-estimate our model. The results (not tabulated) are quite similar to those in Table 3.
6. CONCLUSION
We examine the differential effects of alternative regulations of internal control over financial reporting (ICFR) on the quality of ICFR disclosures. We document that companies under full Section 404 regime provide better-quality internal control reports than those subject to Section 302 regime or Section 404(a) regime. However, we find no supporting evidence that companies under Section 404(a), relative to Section 302 regime, show better quality of ICFR reports. Section 404(a) has been effective for only a few years and therefore it restricts our sample size of companies subject to Section 404(a). Future research to re-examine the effect of Section 404 (a) on the quality of ICFR disclosures may be worthwhile.
26
By developing measurements based on the relation between ICFR disclosure quality and financial restatements, this study concludes that external auditor attestation can improve the quality of internal control disclosure. We also find the quality of ICFR reports further enhanced by brand name auditors. While non-accelerators do not have to comply with SOX 404(b) according to the Dodd-Frank Wall Street Reform and Consumer Protection Act, the Government Accountability Office are requested to evaluate the effects of the exemption. This study has direct implications for legislators and regulators in setting up related rulings in the future.
27
REFERENCES
American Institute of CPAs (AICPA). 2008. Communicating Internal Control Related
Matters Identified in an Audit. Statements on Auditing Standards(SAS) No. 115.
Washington, D. C.: AICPA.
Ashbaugh-Skaife, H., D. Collins, and W. Kinney. 2007. The discovery and reporting of internal control deficiencies prior to SOX-mandated audits. Journal of
Accounting and Economics 44 (1-2): 166–192.
Ashbaugh-Skaife, H., D. Collins, W. Kinney, and R. LaFond. 2008. The effect of SOX internal control deficiencies and their remediation on accrual quality. The
Accounting Review 83 (1): 217–250.
Baber, W. R., S. H. Kang, L. Liang, and Z. Zho. 2006. Shareholder Rights, Corporate Governance, and Accounting Restatements. Working Paper, George Washington University, Syracuse University.
Becker, C. L., M. L. DeFond, J. Jiambalvo and K. R. Subramanyam. 1998. The effect of audit quality on earnings management. Contemporary Accounting Research 15 (1): 1–24.
Bedard, J. C., and L. Graham. 2008. Archival evidence on processes and outcomes of internal control assessment under Section 404 of the Sarbanes-Oxley Act. Working paper, Bentley University.
Bedard, J. C. and L. Graham. 2011. Detection and severity classifications of Sarbanes-Oxley Section 404 internal control deficiencies. The Accounting
Review 86 (3): 825–855.
Bedard, J. C., R. Hoitash and U. Hoitash. 2009. Evidence from the United States on the effect of auditor involvement in assessing internal control over financial reporting. International Journal of Auditing 13 (2): 105–125.
Behn, B., J. H. Choi, and T. Kang. 2008. Audit quality and properties of analyst earnings forecasts. The Accounting Review 83 (2): 327–359.
Beneish, M.D., M.B. Billings, and L.D. Hodder. 2008. Internal control weaknesses and information uncertainty. The Accounting Review 83 (3): 665–703.
Bronson, S., J. Carcello, and K. Raghunandan. 2006. Firm characteristics and voluntary management reports on internal control. Auditing: A Journal of
Practice & Theory 25 (2): 25–39.
Comment, R., and G. Jarrell. 1991. The relative signaling power of Dutch-auction and fixed-price self-tender offers and open-market share repurchases. Journal of
Finance 46(4): 1243–1271.
Cox, C. 2007. Sarbanes-Oxley Section 404: new evidence on the cost of small companies. Presented at U.S. House of Representatives Committee on Small Business. December 12.
DeAngelo, L. 1981. Auditor size and audit quality. Journal of Accounting and
Economics 3 (3): 183–199.
DeFond, M. L., and J. Jiambalvo. 1991. Incidence and circumstances of accounting errors. The Accounting Review 66 (3): 643–655.
DeFond, M. and J. Jiambalvo. 1993. Factors Related to Auditor-Client Disagreements OverIncome-Increasing Accounting Methods. Contemporary Accounting
Research 9 (2): 415–431.
28
profession: a descriptive and normative view. In Regulation and the Accounting
Profession, 34, (2): edited by J. Buckley and F. Weston, 283–289. Belmont, CA:
Lifetime Learning Publications.
Doyle, J., W. Ge, and S. McVay. 2007a. Accruals quality and internal control over financial reporting. The Accounting Review 82 (5): 1141–1170.
Doyle, J., W. Ge, and S. McVay. 2007b. Determinants of weaknesses in internal control over financial reporting. Journal of Accounting and Economics 44 (1–2): 193–223.
Deumes, R., and W. R. Knechel. 2008. Economic incentives for voluntary reporting on internal risk management and control systems. Auditing: A Journal of
Practice & Theory 27 (1): 35–66.
Eldridge, S., and B. Kealey. 2005. SOX costs: auditor attestation under Section 404. Working Paper.
Engel, E., R. M. Hayes, and X. Wang. 2007. The Sarbanes-Oxley Act and firms’ going private decisions. Journal of Accounting and Economics 44 (1–2): 116–145. Gao F., J. S. Wu, and J. Zimmerman. 2009. Unintended consequences of granting
small firms exemptions from securities regulation: evidence from the Sarbanes Oxley Act. Journal of Accounting Research 47 (2): 459–506.
Hammersly, J.S., L.A. Myers., and C. Shakespeare. 2008. Market reactions to the disclosure of internal control weakness and to the characteristics of those weaknesses under section 302 of the Sarbanes Oxley Act of 2002. Review of
Accounting Studies 13 (1): 141–165.
Hartman T. E.. 2007. The cost of being public in the era of Sarbanes-Oxley. Foley & Lardner LLP.
Healy P. M. and K. G. Palepu. 2001. Information asymmetry, corporate disclosure, and the capital markets: a review of the empirical disclosure literature. Journal
of Accounting and Economics 31 (1–3): 405–440.
Hermanson, D. R. and Z. Ye. 2009. Why do some accelerated filers with SOX Section 404 material weaknesses provide early warning under Section 302? Auditing: A
Journal of Practice & Theory 28 (2): 247–271.
Jagadison, K., J. C. Aier, M. T. Gunlock and D. Lee. 2005. The financial expertise of CFOs and accounting restatements. Accounting Horizons 19 (3): 123–135. Khurana, I., and K. Raman. 2004. Litigation risk and the financial reporting
credibility of Big 4 versus non-Big 4 audits: Evidence from Anglo-American countries. The Accounting Review 79 (2): 473–495.
Kinney, W., and L. McDaniel. 1989. Characteristics of firms correcting previously reported earnings. Journal of Accounting and Economics 11 (1): 71–93.
Kinney, W. R., and M. L. Shepardson. 2010. Do control effectiveness disclosures require SOX 404(b) internal control audits? A natural experiment with small U.S. public companies. Journal of Accounting Research 49 (2): 413–448. Leuz, C., A. Triantis, and T. Wang. 2008. Why do firms go dark: Causes and
economic consequences of voluntary SEC deregistrations. Journal of
Accounting Research 45 (2–3): 181–208.
Lu, H., G. Richardson, and S. Salterio. 2010. Direct and indirect effects of internal control weaknesses on accrual quality: evidence from a unique Canadian regulatory setting. Contemporary Accounting Research. Forthcoming.
Palmrose, Z. V. 1999. An analysis of auditor litigation and audit service quality. The
Accounting Review 63 (1): 55–73
Piotroski, J. D. and S. Srinivasan. 2008. Regulation and bonding: the Sarbanes-Oxley Act and the flow of international listings. Journal of Accounting Research 46
29 (2): 383–425.
Public Company Accounting Oversight Board (PCAOB). 2004. An Audit of Internal
Control over Financial Reporting Performed in Conjunction with an Audit of Financial Statements. Auditing Standard (AS) No. 2. Washington, D. C.:
PCAOB.
Public Company Accounting Oversight Board (PCAOB). 2007. An Audit of Internal
Control Over Financial Reporting That Is Integrated with An Audit of Financial Statements. Auditing Standard (AS) No. 5. Washington, D. C.: PCAOB.
Rice, S. C. and D. P. Weber. 2012. How effective is internal control reporting under SOX 404? Determinants of the (Non-) Disclosure of Existing Material Weaknesses. Journal of Accounting Research 50 (3): 811–843.
Richardson, S. A., A. I. Tuna, and M. Wu. 2002. Predicting Earnings Management: The case of earnings restatements. Working Paper, University of Pennsylvania, Hong Kong University of Science and Technology.
Ross, S. 1977. The determination of financial structure: the incentive signaling approach. The Bell Journal of Economics 8(1):23–40.
Securities and Exchange Commission (SEC). 2002. Certification of Disclosure in
Companies’ Quarterly and Annual Reports. Release Nos. 33-8124, 34-46427.
Washington, D. C.: SEC.
Securities and Exchange Commission (SEC). 2007a. Approves new guidance for
compliance with Section 404 of Sarbanes-Oxley. Press release 2007-101. (Cox,
C.) May 23.Washington, D. C.: SEC.
Securities and Exchange Commission (SEC). 2007b. Amendments to rules regarding
managements report on internal control over financial reporting. Release Nos.
33-8809, 34-55928. Washington, D. C.: SEC.
Securities and Exchange Commission (SEC). 2007c. Commission guidance regarding
management;s report on internal control over financial reporting under Section 13(a) or 15(d) of the Securities Exchange Act of 1934. Release Nos. 33-8810;
34-55929; FR-77; File No. S7-24-06. Washington, D. C.: SEC.
Securities and Exchange Commission (SEC). 2007d. Statement by SEC Staff: SEC’s
proposed interpretive guidance to management for Section 404 of Sarbanes-Oxley Act. (Hewitt, C.) May 23.Washington, D. C.: SEC.
Securities and Exchange Commission (SEC). 2009. Study of the Sarbanes-Oxley Act
of 2002 section 404 internal control over financial reporting. Washington, D. C.:
SEC.
Spence, M. 1973. Job market signaling. Quarterly Journal of Economics 87(3): 355– 374.
Teoh, S. H. and T. J. Wong. 1993. Perceived auditor quality and the earnings response coefficient. The Accounting Review 68(2): 346–67.
Wang, X. 2010. Increased disclosure requirements and corporate governance decisions: Evidence from chief financial officers in the pre- and post-Sarbanes-Oxley periods. Journal of Accounting Research 48 (4): 885–920.
30 FIGURE 1
31 TABLE 1 Descriptive Statistics Panel A: Temporal distribution of restatements and ineffective controls
Restating Firms Ineffective Control Firms Year Number Freq. (%) Number Freq. (%)
2002 384 9.99 12 0.66 2003 643 16.74 45 2.48 2004 696 18.12 270 14.88 2005 595 15.49 342 18.84 2006 429 11.17 268 14.77 2007 329 8.56 240 13.22 2008 304 7.91 210 11.57 2009 269 7.00 218 12.01 2010 193 5.02 210 11.57 Total 3842 100.00 1815 100.00
Panel B: Descriptive Statistics of Regression Variables
Variable a Mean Median Std. Dev. Min Max EFFECTI VE 0.93 1.00 0.26 0.00 1.00 SIZE 5.78 5.78 2.04 0.42 10.57 ROA(%) -7.80 3.37 41.88 -297.95 30.02 LEV 0.50 0.45 0.35 0.05 2.56 PER 11.80 13.44 49.86 -221.20 254.25 MBR 2.88 2.05 4.73 -15.55 30.03 BIGN 0.75 1.00 0.43 0.00 1.00
Panel C: Descriptive Statistics for Firms Subject to Different ICFR Regulations
Section 302 only (n=8320) Section 404(a) only (n=2835) Full Section 404 (n=13761) F-test of Mean Differenceb Variable Mean Median Std. Dev. Mean Median Std. Dev. Mean Median Std. Dev. F Statistic P-value
32 EFFECT IVE 0.95 1.00 0.21 0.82 1.00 0.39 0.93 1.00 0.25 312.62*** 0.00 SIZE 5.21 5.02 1.90 3.22 3.36 1.43 6.65 6.55 1.63 5496.81*** 0.00 ROA(%) -8.42 2.42 38.45 -42.77 -6.76 85.03 -0.23 4.53 21.97 1345.05*** 0.00 LEV 0.48 0.44 0.34 0.62 0.43 0.63 0.48 0.46 0.27 208.92*** 0.00 PER 10.72 12.47 51.84 2.12 -1.04 35.37 14.44 15.76 50.92 75.15*** 0.00 MBR 2.86 1.97 4.71 2.11 1.17 5.99 3.06 2.23 4.43 48.37*** 0.00 BIGN 0.77 1.00 0.42 0.20 0.00 0.40 0.86 1.00 0.35 3527.90*** 0.00 a Variable Definitions:
RESTATE = financial restatement; 1 if a company restated its financial statements of the same year in which the internal control report is issued, and 0 otherwise;
EFFECTIVE = effective controls; 1 if a company concludes its internal control system as effective in the 10-K filings and 0 otherwise; SIZE = firm size; the natural log of a company’s total assets (unit of total assets: millions of dollars)
ROA = return on assets (net income before extraordinary items divided by total assets); LEV = leverage; total liabilities divided by total assets;
PER = price earnings ratios; year-end closing price divided by earnings per share; MBR = market to book ratios; market value divided by book value;
BIGN = audit firm type; 1 if a company is audited by a Big-N auditor, and 0 otherwise b ***, **, * significant at a p-value no larger than 0.01, 0.05, or 0.1, respectively
33
a Pearson correlation are presented below the diagonal and Spearman correlation are presented above. b Correlations significant at the two-tailed 0.05 level are in bold fonts.
c Variable Definitions:
RESTATE = financial restatement; 1 if a company restated its financial statements of the same year in which the internal control report is issued, and 0 otherwise;
EFFECTIVE = effective controls; 1 if a company concludes its internal control system as effective in the 10-K filings and 0 otherwise;
SIZE = firm size; the natural log of the company’s total assets (unit of total assets: millions of dollars) ROA = return on assets (net income before extraordinary items divided by total assets);
LEV = leverage; total liabilities divided by total assets;
PER = price earnings ratios; year-end closing price divided by earnings per share; MBR = market to book ratios; market value divided by book value;
BIGN = audit firm type; 1 if a company is audited by a Big-N auditor, and 0 otherwise TABLE 2
Pearson / Spearman Correlation Matrixabc
RESTATE EFFECTIVE SIZE ROA LEV PER MBR BIGN SOX 404(a) SOX 404 RESTATE -0.221 0.003 -0.049 0.036 0.005 -0.026 0.014 -0.041 -0.073 EFFECTIVE -0.221 0.125 0.113 -0.044 0.067 0.048 0.146 -0.153 0.028 SIZE -0.001 0.138 0.333 0.280 0.269 0.048 0.518 -0.433 0.485 ROA -0.007 0.146 0.453 -0.132 0.515 0.252 0.142 -0.193 0.198 LEV 0.033 -0.108 -0.011 -0.444 -0.082 -0.044 0.081 0.003 0.028 PER 0.006 0.027 0.104 0.128 -0.059 0.173 0.149 -0.180 0.144 MBR -0.020 0.027 -0.031 0.015 -0.139 0.015 0.110 -0.170 0.139 BIGN 0.014 0.146 0.519 0.229 -0.055 0.052 0.022 -0.462 0.263 SOX 404(a) -0.041 -0.153 -0.449 -0.299 0.128 -0.070 -0.059 -0.462 -0.398 SOX 404 -0.073 0.028 0.475 0.201 -0.053 0.059 0.042 0.263 -0.398
34
TABLE 3
Logistic Regression of Restatements on Disclosed Effectiveness and Internal Control Regulations
YEAR BIGN MBR PER LEV ROA SIZE SOX EFFECTIVE a SOX EFFECTIVE SOX a SOX EFFECTIVE RESTAT 11 10 9 8 7 6 5 4 3 2 1 0 404 ) ( 404 404 ) ( 404 ) Prob(
Base Model Main Model Variablea Sign Predictions Coeff. Estimate p-value Coeff. Estimate p-value EFFECTIVE － -1.805 0.000 -1.479 0.000 SOX404(a) ＋ -0.114 0.464 SOX404 ＋ 0.315 0.013 EFFECTIVE *SOX404(a) － 0.059 0.728 EFFECTIVE *SOX404 － -0.611 0.000 SIZE ＋ 0.040 0.001 0.054 0.000 ROA(%) － 0.001 0.365 0.000 0.696 LEV ＋ 0.204 0.001 0.215 0.000 PER ＋ 0.000 0.361 0.000 0.314 MBR ＋ -0.010 0.013 -0.009 0.020 BIGN － -0.046 0.387 -0.022 0.684 Intercept ? -1.403 0.000 -1.581 0.000
YEAR ? (omitted) (omitted)
Joint Test 4 2 － -0.055 0.607 5 3 － -0.296 0.000 2 3 ＋ 0.429 0.002 4 2 5 3 － -0.241 0.014 Pseudo R2 0.083 0.085 Hosmer-Lemeshow p-valueb 0.342 0.400 # of observations 24916 24916 a Variable Definitions:
RESTATE = financial restatement; 1 if a company restated its financial statements of the same year in which the internal control report is issued , and 0 otherwise;
