The Moderating Effect Of Perceived Ethical Leadership On Reduced Audit Quality Behaviors

Abstract

As auditors perform the audit, they are faced with many decisions that ultimately affect the quality of the audit and can lead to the issuance of an improper audit opinion such as accepting weak management explanations without corroborating evidence, superficial review of client documentation, premature sign off of audit procedures, or underreporting of time spent on audit task. In this study, I investigate the impact of the perceived ethical leadership (EL) of the audit supervisor upon the auditor's propensity to engage in reduced audit quality (RAQ) acts. This question is of particular interest considering the renewed focus placed upon audit quality by audit professional and regulatory bodies around the world (CAQ, 2014; PCAOB, 2013; IAASB, 2013; FRC, 2006).In studying the potential impact of perceived supervisor EL, I was particularly interested in two aspects of this potential relationship. First, I was interested in whether perceived supervisor EL would have a direct effect on an auditor's propensity to engage in RAQ acts as this leadership quality has not been previously studied in relation to RAQ acts. Second, I was interested in whether perceived supervisor EL would moderate the relationship between the auditor personal characteristics of locus of control (LOC), professional commitment (PC), and organizational commitment (OC) and an auditor's propensity to engage in RAQ acts. This research addresses these questions using a non-experimental design utilizing a survey instrument and a sample of 114 staff and senior level audit professionals. My results provide support for the hypothesized main effects of perceived EL for the RAQ acts of premature signoff (PMSO) and the composite other RAQ acts variable but generally does not provide support for the RAQ act of underreporting of time (URT). It is not surprising that the results would vary among the three different RAQ acts as Coram et al. (2008) found that auditors perceived the moral intensity of various RAQ acts to be different. With regards to the hypothesized moderating effects of perceived supervisor EL upon the three auditor characteristics, the results are mixed. When considering the likelihood of engaging in PMSO, there was a significant interaction effect between perceived supervisor EL and both auditor PC and OC but not for LOC. Upon further investigating these interaction effects, the results show that the perceived supervisor EL will reduce an auditor's likelihood of engaging in PMSO more for auditor's with lower levels of PC and OC as opposed to higher levels of PC and OC. When considering the likelihood of engaging in URT, there were no significant interaction effects between perceived supervisor EL and any of the three auditor characteristics tested although auditor PC and OC both have a significant negative relationship with the auditor's likelihood to engage in URT. When considering the likelihood of engaging in the composite other RAQ variable, there was a significant interaction effect between perceived supervisor EL and both auditor LOC and PC but not for OC. Upon further investigating these interaction effects, the results show that the perceived supervisor EL will reduce an auditor's likelihood of engaging in OTHER RAQ acts more for auditors with an internal LOC as opposed to an external LOC and more for auditors with lower levels of PC as opposed to higher levels of PC.