Research on Professional Skepticism
Strategic Firms, Market Consequence, and Auditor Professional Skepticism – after the Adoption of ASU 2014-15, Ole-Kristian Hope & Jingjing Wang, Rotman School of Management, University of Toronto
ASU 2014-15 requires management of U.S. firms to disclose going concern (GC) uncertainties in the notes to the financial statement, for each annual and interim reporting period. Both anecdotal evidence (e.g., Sears) and empirical research (Wang 2019) show that management acknowledging that the firm has substantial doubt in GC triggers substantially negative market reactions. Managers tend to withhold bad news due to career concerns and they tend to believe that the firm’s status would improve in the future and thus may never have to release the bad news (Kothari, Shu, and Wysocki 2009; Graham, Harvey, and Rajgopal 2005). Given the negative market consequences following GC disclosures, managers could trade-off the cost and benefit of releasing the news to investors, and may even try to avoid disclosures by engaging in strategic activities, such as cutting on investment and other discretionary expenditures. Auditors may be reluctant to issue a GC opinion for troubled firms (Venuti 2004) in order to avoid the “self-fulfilling prophecy” of a corporate death spiral. The researchers intend to explore how to identify companies that are behaving strategically (and how they are doing so) to avoid GC disclosures and thereby lower market efficiency. Findings would contribute to an enhanced platform of understanding for professional skepticism.
Aspects of Professional Skepticism – Understanding, Application, and Dissemination, Grant Thornton (Eric Au, Caroline Hillyard, Jennifer Fiddian-Green) & Minlei Ye, Rotman School of Management, University of Toronto
This study proposes to explore the utility of using artificial intelligence and machine learning techniques to enhance professional skepticism in regard to the provision of assurance guidance and audit quality in Grant Thornton. The main theme of the research was to apply artificial intelligence technology within the audit process. This was to be considered in a landscape where the profession is considering how to continue to apply professional skepticism with these emerging technologies. The key elements of our research objectives are currently:
- The development of a machine learning algorithm in evaluating an aspect of the audit process that is currently heavily dependent on the auditor's professional judgement
- A comparison of the outcome of the machine learning algorithm to the outcome of the application of professional judgement.
- While there is a long-term goal in applying the research to multiple audit risk areas, the current focus being discussed is the audit risk related to missing going concern note disclosures as required by the application financial reporting framework. The considerations for this include the following:
- The narrowing of scope to a specific audit risk (going-concern note disclosure) will improve the practicality of successfully finishing the research.
- Going-concern note disclosures are a major risk consideration in the current audit landscape
- The key inputs to develop the model are available in the public domain, which include analyst reports/calls, MD&A, the financial statements, and the benefit of hindsight in whether the company actually had going-concern issues (if using samples with enough history).
- If this research is successful, it would enable the conduct of future research that could determine how and the most effective way of applying professional skepticism when auditing in conjunction with machine learning technology.