Financial restatements and auditor quality

Abstract

This thesis examines the association between auditor quality and the likelihood of financial restatements for Australian listed companies. I focus on two types of restatements: error-based restatements, which are financial restatements to correct a prior-period error-based misstatement, and stealth restatements, which are restatements made without qualitative disclosures required by the regulation. I document that 32%, 23% and 1.9% of Australian listed companies made restatements, stealth restatements and error-based restatements respectively in my sample of 1,465 Australian listed companies in the 2012/2013 reporting period. I construct an Error Restatement Model and a Stealth Restatement Model to examine the effects of auditor quality and auditor quality changes on the likelihood of different types of restatements. I find some evidence that prior-period auditor quality, measured by whether an auditor is a Big 4 audit firm, is associated with a lower likelihood of a current-period error-based restatement. However, a change in auditor to a Big 4 audit firm (or to an industry specialist audit firm) is associated with a higher likelihood of an error-based restatement arising from the first year audit engagement. I find evidence that higher auditor quality, measured by whether an auditor is an industry specialist audit firm, is associated with a lower likelihood of a stealth restatement and that a change in auditor increases the likelihood of a stealth restatement, but no evidence that higher quality auditors are more effective at detecting or deterring client restatements without qualitative disclosures in first year audit engagements.