Do Biased Tax Research Memoranda Influence Supervisors' Initial Judgments in the Reveiw Process?

John A. Barrick
Brigham Young University

C. Bryan Cloyd
University of Illinois at Urbana–Champaign

Brian C. Spilker
Brigham Young University

Abstract: Public accounting firms rely upon the review process, whereby the work of subordinate accounting professionals is reviewed by more experienced professionals, as both a quality control mechanism and a training tool. Prior research documents that accounting professionals are subject to confirmation bias in both information search and information evaluation, where confirmation bias is defined as over (under) weighting information that supports (refutes) the conclusion advocated by the professional. We investigate the effects of confirmation bias underlying staff-level research on supervisors’ initial assessments and recommendations made during the review process. In an experiment involving 56 tax professionals with review experience, we find that the effect of biased memoranda on supervisors’ initial assessments and recommendations depends on whether the memoranda refutes or supports the client’s preferred position. Specifically, we find that supervisors are not unduly influenced by confirmation bias introduced during staff-level research when the bias is against the client’s-preferred position. However, we find that supervisors’ judgments and recommendations are influenced in the direction of an incorrect but client-preferred position when the bias is in favor of that position, and that this effect is more pronounced for more experienced supervisors.

Back to Program