The Importance of the Auditor’s Internal Control Opinions:
Beyond the Financial Statement Audit Opinion
Thomas J. Lopez
Scott D. Vandervelde**
Yi-Jing Wu
University of South Carolina
1705 College Street
Columbia, SC 29208
Ph.: (803) 777-6075
Fax: (803) 777-0712
October 11, 2006
**Corresponding author
This paper has benefited from helpful comments from Wendy Bailey, James Bierstaker, Michael Cipriano, Tim Doupnik, Dutch Fayard, Audrey Gramling, Richard Houston, Scott Jackson, Jennifer Joe, Lisa Koonce, Al Leitch, Terence Pitre, Brad Tuttle, two anonymous reviewers, and the workshop participants at the American Accounting Association 2005 ABO Conference and the University of South Carolina.
The Importance of the Auditor’s Internal Control Opinions:
Beyond the Financial Statement Audit Opinion
Summary: Recent corporate scandals have increased the focus of Congress and the SEC on the protection of investors. In response to these scandals, Congress passed the Sarbanes-Oxley Act of 2002 (SOX) which, among other things, requires two new audit opinions related to a company’s internal controls: (1) the auditor’s opinion on management’s assessment of internal controls, and (2) the auditor’s opinion on the effectiveness of internal controls. The expectation is that the new audit opinions will provide investors with information regarding the reliability of financial statements beyond the traditional financial statement audit opinion, thus helping investors in their company valuations. Our results suggest that the new internal control reporting requirements of SOX provide information to investors beyond what is conveyed in the traditional financial statement audit opinion. However, evidence indicates the value placed on the internal control opinions is primarily contained in the opinion on the effectiveness of internal controls, which is consistent with the direction of some of the discussion at the PCAOB Standing Advisory Group June 2006 meeting. While investigating the potential cause of the relation between the auditor’s opinion on the effectiveness of internal controls and the assessed stock price, we demonstrate that the internal control report has implications for investors’ assessments of earnings quality.
Key words: Internal control reporting, audit opinions, mediation analysis, perceived earnings quality
The data is available upon request.
INTRODUCTION
Recent corporate scandals, such as those at Enron, WorldCom and Tyco, have increased the focus of Congress and the SEC on the protection of investors (PCAOB 2004; Securities Industry Association 2002; SEC 2003). In response to these scandals, Congress passed the Sarbanes-Oxley Act of 2002 (SOX) which, among other things, requires two new audit opinions related to the effectiveness of a company’s internal controls: (1) the auditor’s opinion on management’s assessment of internal controls, and (2) the auditor’s opinion on the effectiveness of internal controls. The assumption implicit in this SOX requirement is the belief by regulators that these new opinions will provide investors with useful information regarding the reliability of the financial statements, beyond that provided by the financial statement audit opinion alone, for making company valuation judgments.
The primary objective of our analysis is to assess whether the implicit assumption of regulators is empirically valid. That is, after controlling for the overall financial statement audit opinion, do investors incorporate internal control audit opinions in their valuation assessment of the firm? And if so, is the relation between the internal control opinion and stock price explained by changes in investors’ perceptions of earnings quality (i.e., reliability of financial statements), as regulators anticipate? In order to investigate the first research question, we experimentally test whether the new internal control opinions affect assessed stock prices. Then, we investigate whether these opinions provide investors with information beyond the financial statement opinion. In order to investigate the second research question, we examine the link between the audit opinion on the effectiveness of internal controls and investors’ perceptions of earnings quality.
A primary goal of SOX is to alert investors to potential financial statement problems that could result from weak internal controls (PCAOB 2004). Inherent in the passage of these new rules is the belief that an adverse internal control audit opinion has the potential to signal investors that the earnings numbers generated by the current internal control system are of lesser quality or of higher risk than previously assumed. This notion and the importance regulators attach to the new internal control reporting requirements is highlighted in a recent speech by Donald Nicolaisen, the Chief Accountant of the SEC.[i] In that speech Nicolaisen states, “I believe that, of all the recent reforms, the internal control requirements have the greatest potential to improve the reliability of financial reporting.”
Results from recent studies (Beneish et al. 2005; De Franco, Guan, and Lu 2005; Hammersley et al. 2005) suggest that management’s disclosure of an internal control weakness/deficiency is associated with a negative market reaction. However, each of these studies relies on relatively small samples in their empirical analyses and focus solely on management’s disclosure of a potential weakness in the system of internal controls.[ii] We believe there are two important distinctions between our study and existing research. First, we focus on the effect of the auditor’s internal control reporting requirements of SOX rather than on management’s disclosure of a potential internal control weakness. We believe this is an important distinction because a report from an independent party should be more salient to investors.
Second, rather than employing an archival event study methodology, we utilize an experiment to investigate the value placed on the SOX-mandated internal control opinions. From a methodological perspective, the experimental design allows for control of confounding variables, thus allowing the isolation of the incremental effect of each internal control opinion on investors’ firm valuation judgment. By doing so, the experimental design enables causal inferences to be made regarding the relation between auditors’ internal control opinions and investors’ firm valuation judgment (Bailey 1982; Mutchler 1985; Maines 1994; McDaniel and Hand 1996; Kennedy 1999). Archival studies are limited in this respect in that both auditors’ internal control reports are disclosed simultaneous with the annual report, which includes the financial statement audit opinion. Thus, archival studies cannot isolate the incremental effects of each audit opinion on stock price or examine whether the internal control audit opinion provides additional information to investors beyond the financial statement audit opinion alone.[iii]
To address our empirical questions, we perform two separate experiments. In our first experiment, we investigate the value investors placed on each of the internal control opinions beyond the financial statement opinion. Our experimental results suggest that investors do not attach the same valuation implications to both internal control audit opinions. We find no significant firm valuation difference between an unqualified and adverse audit opinion on management’s assessment of internal controls. This result provides evidence consistent with the tone of some of the discussion at the PCAOB Standing Advisory Group (SAG) June 2006 meeting, pointing out that the opinion on management’s assessment likely does not provide information to financial statement users (PCAOB 2006). This result provides evidence of interest to the SAG related to the debate of the value of the auditor’s opinion on management’s assessment. On the other hand, our results suggest that the auditor’s opinion on the effectiveness of internal controls influences investors’ firm valuation judgments. Consistent with this conclusion, our experimental results show that the assessed stock price for firms with an adverse auditor opinion on the effectiveness of internal controls is significantly less than for firms with an unqualified opinion. This result is robust to whether the company has an unqualified opinion on the financial statements or an unqualified opinion with an explanatory paragraph.[iv] Thus, our results indicate that an adverse internal control opinion on the effectiveness of internal controls provides information to investors beyond that contained in the financial statement audit opinion alone.
In our second experiment, we examine a potential cause of the relationship between the opinion on the effectiveness of internal controls and the assessed stock price. One possible explanation for the observed stock price effect is that the opinion affects investor perceptions of earnings quality where earnings quality is defined as the extent to which there is uncertainty with respect to past, current, and future earnings. Thus, in experiment two, we examine whether an unqualified versus an adverse audit opinion on the effectiveness of internal controls differentially affects investor perceptions of factors commonly associated with earnings quality. The factors we test are: (1) the perceived risk of financial statement misstatement, (2) the perceived risk that a material error will not be detected, and (3) the perceived risk of financial statement restatement. We find that an adverse audit opinion on the effectiveness of internal controls relative to an unqualified opinion is significantly associated with a higher assessed risk of financial statement misstatement and a higher assessed risk of a future financial statement restatement. On the other hand, we find that the assessed risk that a material error will fail to be detected is not significantly associated with an adverse opinion on the effectiveness of the system of internal controls.[v]
Finally, we test the factors of earnings quality in a mediation model (Baron and Kenny 1986; Jackson and Hatfield 2005) in order to provide evidence regarding the causes of the relationship between the auditor’s opinion on the effectiveness of internal controls and the assessed stock price. We find that the assessed risk of material misstatement, which includes the risk of earnings management, partially mediates the relationship between the opinion on the effectiveness of internal controls and the assessed stock price in the predicted direction. As discussed further in the paper, Baron and Kenny (1986) state that in most cases partial mediation, rather than full mediation, is the most one can hope to find. This arises because it is difficult to precisely measure the mediating variable which often is a psychological variable (e.g., investor perception); and most dependent variables have more than one cause (e.g., change in stock price). Thus, consistent with our expectation, our results suggest that investors perceive a lower quality of earnings for firms with an adverse internal control report compared to firms with an unqualified report and this fact explains, at least in part, the lower stock price for firms with an adverse opinion.
Our study makes at least two important contributions to the existing literature. First, our results confirm the assumption of regulators that the new auditor’s internal control reporting requirements of SOX provide information to investors regarding the reliability of financial statements beyond what is conveyed in the traditional financial statement audit opinion. However, our results also suggest that the auditor’s opinion on managements’ assessment of internal controls provides no price-relevant information to investors. This evidence is important to the PCAOB as evidenced by the direction of the discussion at the SAG June 2006 meeting that suggested that the opinion on management’s assessment could be removed as a requirement in the audit process (PCAOB 2006). Second, we demonstrate that the internal control report has implications for investors’ assessments of earnings quality. Our experimental setting allows us to measure several dimensions of investors’ perceptions of earnings quality and to directly test the mediating effect of investors’ perceptions on their assessed stock price valuations. This is important because it provides empirical support for conjectures made in the prior literature on the relationship between audit opinions and stock prices (Choi and Jeter 1992). More importantly, our examination of the relationship between the opinion on the effectiveness of internal controls and the assessed stock price directly addresses regulators’ claims that additional disclosure about the system of internal controls will provide investors with useful information about the quality of a firms’ earnings (i.e., reliability of financial statements) (SEC 2003).
The remainder of this paper is organized as follows. Section two presents background literature and the hypotheses development. Section three presents the experimental design and results for our tests of the auditor’s internal control opinions. Section four presents the experimental design and results for our tests of the mediating variables. Section five presents a summary and conclusion.
Background and Hypotheses Development
Background and Prior Literature
In the aftermath of recent corporate scandals, regulators suggested that effective internal control systems will lead to more reliable financial information, which will help investors make more informed investment decisions (SEC 2003). The implementation of SOX and PCAOB Auditing Standard No. 2 (AS2) modified the system of audit disclosure from merely attesting to the financial statements, to also opining on the financial reporting process (i.e., the system of internal controls).[vi] Under this new reporting system, the auditor must express two opinions related to internal controls over financial reporting: (1) the auditor’s opinion on management’s assessment of internal controls, and (2) the auditor’s opinion on the effectiveness of internal controls (PCAOB 2004). Issuing audit opinions related to the process by which transactions are recorded provides information to financial statement users regarding the reliability of the financial statements.
Under the new reporting guidelines, a deviation from an auditor’s standard unqualified opinion on the effectiveness of internal controls will result if even one material weakness exists in the financial reporting system. A material weakness in internal controls is defined as “a significant deficiency or combination of significant deficiencies, that results in more than a remote likelihood that a material misstatement of the annual or interim financial statements will not be prevented or detected” (PCAOB 2004, A-10). According to AS2, in the event that the auditor identifies one or more material weaknesses in internal control over financial reporting, “the auditor must express an adverse opinion on the company’s internal control over financial reporting” (PCAOB 2004, A-75). Previously these material weaknesses remained private information, unless they warranted a modified financial statement audit opinion or were part of a required 8-K filing in conjunction with a change in auditors. Thus, AS2 substantially changes the information environment of the audited company with respect to its system of internal controls.
The new internal control reporting requirements are part of the SEC’s aggressive efforts to protect investors from corporate fraud and to reassure them that corporate failures, like Enron, will not recur (SEC 2002a; SEC 2002b; SEC 2003). The assumption underlying these mandatory internal control audit opinions is that they will affect investors’ judgments, especially when a material weakness exists. Given that a company can receive an unqualified financial statement opinion even in the presence of a material weakness in internal controls, the new reporting requirements are expected to affect investor’s judgments beyond the information in the financial statement audit opinion.