Development and Validation of an Archival-Based Measure of Negotiation Likelihood to Corroborate Experimental and Field-based Auditor Client Management Negotiation Research

• In Event: Mixed Methodologies

Abstract

In the last fifteen years, numerous experimental and field studies have examined the factors that influence auditor-client management (ACM) negotiations over financial statements. This paper develops and validates an archival-based measure of the likelihood that ACM negotiations will occur at the end of the fiscal year. To achieve this goal, we adapt a measure of audit efficiency, audit report lag (the length of time between the financial statement year-end date and the auditor’s report date), by incorporating items into the model of its determinants so as to have a residual that we argue is a measure of underlying ACM negotiation likelihood. Based on field research that characterizes two types of negotiation relationships between auditors and client management, we suggest that larger abnormal audit report lag is a viable proxy that measures the likelihood of year-end ACM negotiations. We show that, as would be expected for a measure of year-end negotiation likelihood, a larger abnormal audit report lag is associated with higher audit fees even after controlling for known determinants of audit fees. We show, consistent with the nature of the ACM relationship that would be associated with likelihood of year-end negotiation, that larger abnormal audit report lags are associated with higher levels of discretionary accruals, that is, lower accrual quality. Based on our findings, we suggest that our proxy based on abnormal audit report lag (denoted as NEG) is a valid archival proxy for the differences in year-end ACM negotiation likelihood. We suggest that this proxy will allow researchers to study issues related to published accounting numbers in light of whether negotiations are likely to have occurred in addition to providing regulators and others the means to determine what clients of audit firms are more likely to have different types of ACM relationships.

Authors

• Yan Luo, San Diego State University

• Steven E Salterio, Queen's University - Kingston