The Public Company Accounting Oversight Board (PCAOB) has introduced critical audit matters (CAMs) to expand audit reports to provide external users with more information regarding potential risks of material misstatements. While the intention of CAM disclosures is to highlight the risk of material misstatement and the corresponding audit procedures, early studies suggest that investors may erroneously perceive CAMs as indicators of overall business risk rather than focusing solely on financial misstatement risk.
The findings of a new study underline this misconception. These findings reveal a significant association between the extensiveness of CAM disclosures and increased market uncertainty. Both the standard deviation of equity returns, and the dispersion of analysts‘ earnings forecasts show a positive and significant relationship with the extent of CAM disclosures. Furthermore, firms with more extensive CAM disclosures experience a higher increase in the dispersion of analyst forecasts after the CAMs are disclosed. Additionally, the study finds a significant negative association between short-term market returns around the Form 10-K filing date and the extent of CAM disclosures, even after controlling for other risk-related disclosures.
The study highlights the need for improved communication and understanding of the role of CAM disclosures. Market participants‘ misinterpretation of CAMs as indicators of overall business risk can lead to increased investor uncertainty. Enhancing the explanation of CAMs and the specific audit risks they measure could potentially mitigate such misinterpretations, thereby providing investors with a clearer understanding of the information conveyed through CAM disclosures. Additionally, the study employs an NLP (Natural Language Processing) approach and develops rules for classifying CAMs and audit procedures. This approach can serve as a foundation for future textual analysis research. Therefore, the study also contributes to the emerging field of textual extraction, showcasing the application of specific NLP rules to extract distinguishing information from similar and standardized texts.
You can find the complete study via the following link.