Associations between SOX 404 opinions and auditors, audit committees, and executives

This dissertation examined three issues associated with Section 404 of the Sarbanes-Oxley Act (SOX) that are of current interest to regulators and the accounting profession. For the first essay, I examined auditor changes from 2003 to 2005 for 2,454 non-financial firms that filed their initial SOX 404 opinions prior to July 1, 2005. My results showed that there is a significant association between the receipt of an adverse SOX 404 opinion and auditor resignations - both before and after the issuance of the SOX 404 opinion. The data related to auditor dismissals show that clients are not successful in opinion shopping, since clients dismissing the auditor before the SOX 404 report also are more likely to receive an adverse SOX 404 opinion than clients not changing auditors. My second essay examined the relation between audit committee characteristics and changes in material weaknesses in internal control over financial reporting under the new SOX regime. My results showed that improvements in internal control in the second year of SOX are associated with: (1) the number of audit committee members and (2) financial expertise of audit committee members. My third essay examined the relation between the appointment of the new executives and the subsequent receipt of initial section 404 of SOX opinions. My results showed that adverse SOX 404 reports will be more likely at firms that recently hired a new chief financial officer (CFO).

Internal control reporting by non-accelerated filers

I examine three issues related to internal control reporting by non-accelerated filers. Motivation for the three studies comes from the fact that Section 404 of the Sarbanes-Oxley Act (SOX) continues to be controversial, as evidenced by the permanent exemption from Section 404(b) of SOX granted to non-accelerated filers by the Dodd-Frank Wall Street Reform and Consumer Protection Act of 2010. The Dodd-Frank Act also requires the SEC to study compliance costs associated with smaller accelerated filers. In the first part of my dissertation, I document that the audit fee premium for non-accelerated filers disclosing a material weakness in internal controls (a) is significantly lower than the corresponding premium for accelerated filers, and (b) declines significantly over time. I also find that in the case of accelerated filers remediating clients pay lower fees compared to clients continuing to report internal control problems; however, such differences are not observed in the case of non-accelerated filers. The second essay focuses on audit report lag. The results indicate that presence of material weaknesses are associated with increased audit report lags, for both accelerated and non-accelerated filers. The results also indicate that the decline in report lag following remediation of problems is greater for accelerated filers than for non-accelerated filers. The third essay examines early warnings (pursuant to Section 302 disclosures) for firms that subsequently disclosed internal control problems in their 404 reports. The analyses indicate that non-accelerated firms with shorter CFO tenure, presence of accounting experts on the audit committee, and more frequent audit committee meetings are more likely to provide prior Section 302 warnings. Overall the results suggest that there are differences in internal control reporting between the accelerated and non-accelerated filers. The results provide empirical grounding for the ongoing debate about internal control reporting by non-accelerated filers.