Date of Award


Document Type


Degree Name

PhD in Accountancy


Department of Accountancy

First Advisor

Jean C. Bedard

Second Advisor

Steven M. Glover

Third Advisor

Lyn Graham


This dissertation consists of two empirical studies that investigate fair value measurement issues currently facing the accounting profession--one from the perspective of the auditor, and the other from the perspective of the financial statement user. The results of each study are described below.

This first study examines experienced auditors' descriptions of specific client experiences in which auditing fair value measurements (FVMs) was particularly challenging. Based on a field survey of high-level engagement team personnel from several large firms, we identify a number of key issues currently facing the profession in auditing FVMs. First, when asked about challenges faced in auditing FVMs, respondents cited most frequently the difficulties inherent in auditing management assumptions. Second, although 70 percent of respondents identified a range of estimation uncertainty equal to or exceeding materiality, inherent risk assessments for some FVMs are still within the low/moderate range. Lastly, relatively few audit adjustments were proposed. Auditors noted that when management assumptions are subjective and estimation uncertainty is large, it is difficult to prove that their assumptions and estimates are more accurate than the client's. Supplemental regression results also show that auditors are more likely to propose a decreasing audit adjustment when they have developed their own independent estimate, and are less likely to propose decreasing audit adjustments for financial instruments, as compared to other types of FVMs.

In my second study, I examine the extent to which the high degree of inherent risk noted in my first study is effectively communicated to financial statement users under conditions of high and low management aggressiveness. Specifically, I conduct an experiment that investigates which of three disclosure formats most effectively communicates the risk of high-uncertainty FVMs to users--a narrative sensitivity disclosure currently required in the U.S., a standard quantitative disclosure currently required under IFRS, or an "enhanced" disclosure I propose that also displays the impact of these changes on net income. Results of my study suggest that the standard quantitative sensitivity currently required under IFRS may have the unintended consequence of decreasing users' risk assessments when management aggressiveness is high. Increased perceptions of trust, competence, and reliability upon receiving this disclosure partially explain this relationship. As predicted, however, the enhanced disclosure condition is more effective than the standard condition at communicating risk to users under conditions of high management aggressiveness. Thus, the additional information in the enhanced disclosure condition appears to counteract the tendency to decrease risk assessments, thus providing increased benefit to users at little incremental cost to preparers.