Discussion of Relevant but Delayed Information in Negotiated Audit Fees. (Hackenbrack, Jenkins, and Pevzner) Discussant Ryan Wilson

Transcription

1 Discussion of Relevant but Delayed Information in Negotiated Audit Fees (Hackenbrack, Jenkins, and Pevzner) Discussant Ryan Wilson

2 Motivation Auditors possess private information about their clients that is impounded into the negotiated audit fee. Auditors will change audit fees from year-to-year in response to changes in client-specific risk. As a result, changes in audit fees are a signal of changes in client-specific risk that would be useful to investors if the fee information was disclosed in a timely manner. Disclosing this information would decrease opacity and incidence of stock price crashes.

3 Benefits of Audit Fee Data Incorporates auditors private assessment of client specific risk. Potential measure of the quality of firms accounting information (opacity). More global measure of changes in client risk and/or information quality (opacity) than measures based on accruals. Unlike other measures of earnings quality, unexplained changes in audit fees should be uncorrelated with fundamental firm characteristics like operating volatility.

4 Challenges of using audit fees If auditors observe increases in client risk and respond by increasing audit procedures this should reduce client risk. Hence, we would expect no relation between changes in audit fees and changes in client risk in the cross-section of firms. Do I expect more crime at a mall with lots of police or a mall with no police? Prior research conjectures that audit and non-audit fees impair auditor independence (e.g. Frankel et al. 2002). The increased bond between auditors and clients then leads to reduced earnings quality. It is not clear what to control for when examining the association between idiosyncratic risk and changes in audit fees (litigation risk, corporate governance, earnings quality, SEOs, IPOs, etc.?)

5 Key Assumption: How are audit fees related to client risk? Auditor s Identify increase in client risk Higher Engagement Risk Increased testing to reduce overall engagement risk to acceptable levels Higher Audit Fees Fee premium in exchange for accepting higher engagement risk Higher Audit Fees Auditor s must be constrained in their ability to remediate client risk through increased audit effort.

6 Key Assumptions (Continued) The audit fee negotiated at the beginning of the year already impounds information about changes in client specific risk. Auditors are constrained in their ability to pass on unexpected costs during the audit process. I agree auditors are constrained in their ability to pass on unexpected costs. However, I would expect any changes in fees that occur during the audit process (however small) to be most informative about changes in client risk.

7 Contemporaneous Studies Hribar Kravet and Wilson (2010): We also assume audit fees contain information about auditors private client specific knowledge. Auditors have insight into the quality of the firm s accounting information. We find unexplained audit fees are associated with future incidence of fraud and restatement.

8 Contemporaneous Studies (Continued) Picconi and Reynolds (2010): Find high abnormal audit fees are associated with poor future performance for smaller firms. Specifically, higher audit fees are associated with poor future operating performance, stock returns, and the likelihood of delisting. Alternative explanations: Is it possible auditors increase fees when they realize the business model is flawed? In that case the client no longer represents and annuity of future fees to the firm?

9 Research Design Primary analysis focuses on the estimation of the following equation: % Audit Fees_Ind = γ 0 + γ 1 * Idiosyncratic Risk + γ t CONTROLS + e Controls are intended to capture the increase in fees due to higher audit effort that is unrelated to the increase in idiosyncratic risk. Include controls for events auditors might be aware of that would require additional effort: Accounting Irregularities Material Control Weakness Are these related to the bad news events driving the stock price crash? These measures could have been alternative test variables similar to securities lawsuits and credit rating downgrades.

10 Research Design (Continued) Supplemental tests control for changes in discretionary accruals/conservatism as possible intervening variables: Increase in Book-Tax Differences Increase in Discretionary Accruals Decrease in conservatism Is the decline in earnings quality associated with the cause of the crash?

11 Litigation Risk Venkataraman Weber and Willenborg (2008): Find audit fees are almost twice as high in IPO years compared to pre-ipo years. They conclude litigation risk is being priced in the audit fee. If you control for litigation risk how does this affect the association between changes in audit fees and crashes? Are investors able to identify changes in litigation risk at the same time auditors do?

12 Research Design (Continued) Do credit rating downgrades and securities class action lawsuits increase the required audit work? If securities lawsuits stem from either a material misstatement or omission from the financial statements will this require a restatement? A restatement would likely increase audit work. Possible control for changes in fees that occur during the engagement: Audit Delay: Number of days between the fiscal year-end and the date of the auditor s report.

13 What are the implications of disclosing audit fees at the beginning of the year? How would investors respond to this information? Would they panic when observing large changes in audit fees? Would they demand an explanation for the change in fees? If so, why not just disclose the information that led to the change in fees rather than the change in fees? Would this disclosure lead to a decline in tipping points? Or would it simply accelerate the tipping point?

14 Future Directions Assumption: The bad news contained in audit fees will be revealed before the proxy statement containing the prior year s audit fee is filed. Would it be interesting to examine the market reaction to significant changes in audit fees when they are disclosed? Do investors realize audit fees may contain useful information about idiosyncratic risk?