A Reexamination of Abnormal Audit Fees or Audit Hours and Audit Quality29) －New Data and New Evidence－
Asian Tax Journal Vol. 19 No. 5 (2018), pp. 9-53
Abstract
This paper investigates the effects of abnormal audit fees and abnormal audit hours on audit quality. Specifically, we test the effect of audit quality on both discretionary accruals and real earnings management as firms’ opportunistic earnings management activities. This is, we reexamine the association between abnormal audit fees (or abnormal audit hours) and audit quality using new data recently. (i.e., years 2004－2015). Abnormal audit fees and audit hours are typically proxied by the residuals obtained from audit fee or hour determinant models. Therefore, consistent with prior research (e.g., Park and Choi 2009；Choi et al. 2010；Asthana and Boone 2012；Eshleman and Guo 2014 etc.), we decompose total audit fees (or total audit hours) into normal and abnormal components, and test for an association between audit quality and abnormal audit fees (or abnormal audit hours), and also conditioning our tests on the sign of the abnormal audit fee (or audit hour) metric (i.e., abnormal low audit fees or high audit fees, abnormal low audit hours or high audit hours). To the extent that audit fees are a measure of audit effort in the case of Korea’s audit environment is significantly associated with lower level of audit fees for auditors, higher fees indicate that the auditor worked more hours, could higher audit quality. Thus, we expect that the association between abnormal audit fees or abnormal audit hours and audit quality is positive, and also we expect that the negative association between abnormal low audit fees or abnormal low audit hours and audit quality, and the positive association between abnormal high audit fees or abnormal high audit hours and audit quality. To test this prediction, we utilize audit fees data from 15,912 firms－year observations and audit hours data from 15,652 firms－year observations 2004－2015 period, respectively. The sample firms are all listed firms in Korean Stock Exchange, December year－end firms, and firms do not belong to banking industry. For analysis, we utilize signed discretionary accruals (hereafter DA) calculated using both Dechow et al. (1995) and Kothari et al. (2005) models. We utilize real earnings management (hereafter RM) calculated using the Roychowdhury (2006) models, this study identify three empirical proxies for RM based on prior research, and also similar to Cohen and Zarowin (2010) and Ge and Kim (2014), we utilize a single as well as combinatorial or comprehensive measure of RM. The empirical findings of this paper are following. First, after controlling for several factors that affect earnings management, we find that there is the negative relationship between abnormal audit fees or abnormal audit hours and discretionary accruals as well as real earnings management. The result is not consistent with the prior studies (e.g., Park and Choi 2009) mostly using to data from the pre－IFRS period (i.e., years 1999－2005). But, with data taken from both the pre－ and post － IFRS periods, we are able to probe for a positive association between abnormal audit fees or abnormal audit hours and audit quality, averagely. Specifically, the results show the negative association between abnormal audit fees or abnormal audit hours and earnings management is more pronounced in the case of real earnings management and in some cases, destroy firm value. Therefore, this result suggests that all of both abnormal audit fees and abnormal audit hours are significantly positive associated audit quality. Second, we find that the auditors with abnormal low audit fee or abnormal low audit hour have higher discretionary accruals and real earnings management. This result supports the hypothesis of this study that audit quality is impaired when audit fees or hours are abnormally low. On the other hand, the effect of abnormal high audit fees or abnormal high audit hours on the level of discretionary accruals is insignificant. In contrast, we find that the auditors with abnormal high audit fee or audit hour have lower real earnings management. These results suggest that the association between abnormal audit fees or audit hours and discretionary accruals as audit quality differs systematically between auditors with positive and negative abnormal fees or hours. As a result, the association between abnormal audit fees or audit hours and discretionary accruals is asymmetric non－linearity. Whereas, the positive association between abnormal audit fees or audit hours and real earnings management is asymmetric linearity. Overall, these evidence is consistent with the notion that abnormal audit fees as well as abnormal audit hours are indicative of greater auditor effort and, ultimately, better audit quality in a situation of Korea’s audit environment. Additionally, this result is consistent with concerns raised by practitioners and regulators that lower audit fees could reflect a lower level of effort provided by the auditor and, ultimately, worse audit quality. Thus, this paper provides evidence to help resolve the debate on whether abnormal audit fees and abnormal audit hours are a sign of audit quality. And the findings of this paper also contribute to the related literature on the relationship between earnings management and audit quality by providing empirical results. Furthermore, this information is of interest to accounting researchers as well as investors, practitioners, and regulators who decide which firms to inspect using publicly available information, such as audit fees and audit hours.
Keywords
- Abnormal audit fees
- Abnormal audit hours
- Audit quality
- Discretionary accruals
- Real earnings management
Related Articles
The Effect of Audit Quality on Quarterly Earnings Management
13(3) 9-51
The Effect of Abnormal Audit Fees and Audit Hours on Discretionary Accruals
10(3) 257-293
Abnormal Audit Fees and Earnings Management －Focused on Privately Held Firms－
14(3) 73-121
Effects of Abnormal Audit Fee on Real Earnings Management
13(3) 83-112
The Effect of the Overvalued Stock Price on Financial Reporting Opacity
21(4) 103-150