Jurors’ Assessments of Audit Quality on Assurance for Fair Values: The Impact of Audit Quality Indicators and Task Difficulty

• In Event: Behavioral Audit Research

Abstract

An important auditing issue that standard setting bodies, auditing firms, and researchers are currently considering is identifying potential measures of audit quality. The purpose of the current study is to examine the relative importance of audit quality input and process determinants on jurors’ overall assessment of audit quality and how this assessment is affected by a pervasive factor in the audit environment: audit task difficulty. We focus on assurance provided on fair value measurements (FVMs) of securities, an increasingly important financial reporting area and one of concern to regulators as well as challenging for auditors. Drawing on Attribution Theory, we posit that when auditors’ FVM task difficulty is moderate (i.e., level 2) jurors will have higher expectations of auditors in judging overall audit quality and thereby employ a non-compensatory model, demanding both high quality input (resources and knowledge) and process (audit tests performed). In contrast, when FVM task difficultly is high (i.e., level 3), jurors are expected to recognize the challenges of providing assurance on such complex FVMs and employ a compensatory model whereby they focus on evidence of reasonable auditor diligence as demonstrated by a high level of either audit quality input and/or process. We test these expectations employing an experiment involving 318 perspective jurors where we manipulate audit quality input (low vs. high), audit quality process (low vs. high), and auditor task difficulty (moderate vs. high). Given a case setting involving a restatement related to a FV estimate, participants assess audit quality, auditor responsibility, and auditor negligence. The results support our expectations and highlight the importance of both audit input and process information as indicators of audit quality conditional on auditor task difficulty.

Authors

• Arnold Wright, Northeastern University

• Yi-Jing Wu, Texas Tech University