감사위험과 감사보수 경직성과의 관련성

This paper examines the effect of audit risk on a determination of audit fees. Specifically, this paper investigates audit fee stickiness in terms of audit risks and relationship between audit risks and audit fee per audit hour. Previous studies suggest that audit fee is a function of audit efforts measure by audit hours and audit risks. It is a general procedure that auditors evaluate the audit risks of audited firms and then determine the level of audit fee based on expected audit hours. However, evaluated audit risks at the beginning can be changed because of undetected risks or unexpected factors and these risks affect the level of audit hours and further audit fee. In this case, auditor should decide whether audit fees are increased or not. On the one hand, auditors want to increase audit hours to reduce increased audit risk. This leads to a adjustment of audit fees. On the other, auditors don't want to lose clients due to re-negotiated audit fee. In particular, adjustment of audit fees is practically uncommon in competitive audit market. Many studies find that audit fee are sticky - audit fees do not immediately or fully adjust to changes in their determinants(Zhang et al. 2011). This is a reason that audit firms are tend to increase audit hours without adjusting audit fee. Based on this argument, we predict that this behavior results in a negative relationship between audit risks and audit fee per audit hour. We further examine the effect of non-audit service on this relationship. Many papers related to audit provide the evidence of non-audit service on a determination of audit fee in two ways. First issue is related to auditor independence. Having additional working relationship with the client would result in questions being asked of the independence of the audit firm. If non-audit service is provided with same auditor, suspicions to auditor independence will arise. Second issue is related to audit effectiveness. This perspective suggests that auditors are able to transfer knowledges from a non-audit service to audit and both tasks potentially improves audit effectiveness. Based on both arguments, this paper tests the effect of non-audit service when audit risks increase. This paper empirically tests this prediction using 2,998 firm-year observations from 2002 to 2011. We adopt a difference between book income and taxable income (hereafter, BTD) as proxy for audit risks. We find that audit fee per audit hour (LHF) is negatively associated with increased audit risks measured by BTD. This finding suggests that audit fee stickiness exists although auditors have efforts to reduce increased audit risks. It means that auditors are reluctant to increase audit fee even though audit risks increase. We also find that when non-audit services related to tax is provided, audit risks are positively associated with audit fee per audit hour. This implies that tax service leads to knowledge spillover with audit service, and results in increase of audit fee per audit hour. The findings of this paper have various implications. First, the evidence on negative relationship between audit risks and audit fee per audit hour is a useful indicator to evaluate audit quality of audited firm. This paper provides the empirical evidence on the audit fee stickiness in terms of audit risks. Second, our results suggest the audit fee stickiness in Korea and this may affect audit quality. Since 1999, audit fee regulations in Korea are under competitive market. This environment influences the behavior of audit firms and then the level of audit fee. Therefore, the empirical findings of this paper can provide practical implications to interested parties. Because it is sure that firms experience low level of audit fee in the short term but firms are also suffer from high level of cost of capital due to poor audit quality in the long term. It is an important implication to investors, regulators, related institutions, as well as practitioners. Researchers who are interested in this area can also apply the discussion in this paper for the related studies.