Recognition V. Disclosure, Auditor Tolerance for Misstatement, and the Reliability of Stock-Compensation and Lease Information.

Submitted by: Submitted by

Views: 195

Words: 2095

Pages: 9

Category: English Composition

Date Submitted: 12/06/2012 12:43 PM

Report This Essay

This paper investigates whether disclosed numbers, as opposed to recognised numbers, lack reliability because auditors tolerate more misstatements in disclosed amounts. Prior studies that analyzed auditors’ correction decisions indicated that auditors trade-off the risk of litigation and reputation loss against the risk of losing clients. In the light of these findings, the researchers anticipate that auditors permit relatively more misstatement in disclosed numbers. They argue that less attention from users and little use of these numbers in contracting leads to auditors tolerating relatively more misstatement in disclosed amounts. On the contrary, the researchers anticipate that auditors tolerate relatively little misstatement in recognised amounts, because (amongst other reasons) the quality inspections of the PCAOB would strengthen auditor resistance to client pressure for aggressive accounting and audit committees would discourage client pressure. In short, the primary hypothesis that is tested in this study is the following: “auditors allow more misstatement in disclosed amounts than in recognised amounts.” This primary hypothesis (accompanied by four additional hypotheses) is tested in two settings: stock compensation and leasing. These two settings are different: (1) they differ in the financial statement that is affected by misstatement and (2) they differ in subjectivity and mandated recognition. By working with two settings, the researchers are able to test the generalizability of the results and eliminate alternative explanations for the findings.

The first experiment (with a sample consisting of 44 Big Four audit partners evenly spread over the Big Four firms) varies whether a misstatement relates to stock compensation expense that is recognised or disclosed. The results of the first experiment indicate that partners allow significantly more misstatement in disclosed amounts than in recognised amounts. Furthermore, the results indicate that partners...