회계이익의 불투명성이 감사보수 및 감사시간에 미치는 영향

본 연구는 3년간의 재량적 발생액의 시계열적 변동성으로 측정되는 회계이익의 불투명성이 감사인의 감사위험으로 반영되는지를 감사보수 및 감사시간 측면에서 실증적으로 분석하였다. 선행연구들에서는 감사보수 또는 비감사보수를 포함한 총보수가 재량적 발생액(DA)에 미치는 영향을 분석한 연구는 많았으나, 보고이익의 질을 결정하는 재량적 발생액이 감사보수에 미치는 효과를 분석한 연구는 소수에 불과했다. 후자의 경우 재량적 발생액이 감사인의 감사위험으로 인지되어 감사보수에 반영되는지를 살펴본 연구들은 검증결과 측면에서 두 변수 간에 양(+)의 관계(Gul, Chen, and Tsui, 2003), 또는 음(-)의 관계(박시연·유관희·유승원, 2012), 그리고 두 변수 간에 관련성을 발견하지 못한 연구(박종일, 2005)도 있어 혼재된 증거를 제시해 왔다. 이와 달리, 본 연구는 Hutton, Marcus, and Tehranian(2009)의 연구에서 제안된 다기간(multi-year)인 과거 3년간 연도별 DA의 절댓값의 합으로 측정되는 회계이익의 불투명성(OPAQUE)과, 또한 본 연구에서 새롭게 제안한 3년간의 DA의 시계열적 변동성을 표준편차로 측정한 OPAQUE_std이 감사보수 및 감사시간 측면에서 감사위험으로 반영되는지를 분석하였다. 이를 위하여 본 연구는 감사보수 및 감사시간 결정모형을 이용해 분석하였고, 2004년부터 2015년까지 12월 결산법인(금융업을 제외) 중 감사보수(감사시간)의 최종표본은 15,708(15,534)개 기업/연 자료를 분석에 이용하였다.

실증결과는 다음과 같다. 첫째, 일정 변수를 통제한 후에도 DA는 감사보수에 대해 유의한 음(-)의 관계로 나타났으나, OPAQUE와 OPAQUE_std 모두는 감사보수에 대해 유의한 양(+)의 관계로 나타났다. 또한 DA는 감사시간에 대해 유의한 음(-)의 관계로 나타났으나, OPAQUE_std는 감사시간에 대해 유의한 양(+)의 관계를, 그러나 OPAQUE는 감사시간과 유의한 결과로 나타나지는 않았다. 둘째, 종속변수를 비정상 측정치로 분석한 경우에도 앞서와 일치된 결과를 보였다. 셋째, 표본을 Big 4와 non-Big 4 감사인으로 나누어 분석하면 OPAQUE 및 OPAQUE_std는 non-Big 4 감사인의 경우 감사보수와 더 뚜렷한 결과를 보인 반면에, Big 4 감사인은 감사시간과 더 뚜렷한 결과를 보였다. 마지막으로, 직급별 감사시간이 공시된 2014년 전후기간으로 표본을 나누어 살펴보면, OPAQUE 및 OPAQUE_std는 2014년 이후기간에서 더 뚜렷한 결과로 나타났다.

이상의 결과를 종합하면, 본 연구는 피감기업의 보고이익의 질을 평가하는데 있어 감사인이 한해 연도의 재량적 발생액 수준보다는 반전효과가 고려된 3년간 다기간의 재량적 발생액의 시계열적 변동성으로 측정되는 회계이익의 불투명성이 높을 때 감사위험의 증가로 인지하여 이를 감사보수 책정과 감사노력(감사시간)에 반영시키고 있음을 보여주었다는데 의의가 있다. 또한 본 연구결과는 새롭게 제안한 OPAQUE_std 측정치의 경우 감사보수뿐만 아니라 감사시간에 대해서도 모두 양(+)의 관계로 나타나 Hutton et al.(2009)의 측정치보다 감사인 측면에서 이익조정 탐지에 있어 더 효과적임을 보여주고 있다. 따라서 본 연구결과는 학계에는 감사품질과 관련된 연구에 추가적인 실증적 증거를 제공해 줄 뿐만 아니라, 실무계 및 규제당국에게는 회계이익의 불투명성이 감사인의 감사보수 책정과 감사노력에 어떤 영향을 주는지와 관련한 시사점을 제공해 줄 것으로 기대된다.

We investigate whether the relation between financial reporting opacity (or opaque earnings) for individual firms based on Hutton et al. (2009) and audit fees as well as audit hours. Specifically, this paper examines the effect of financial reporting opacity related to the audit risk of a auditor on audit fees and audit hours. Auditors who have responsibilities to provide assurance to reliability of the financial statements may concern about audit risks arising from managers’ earnings management incentives. Auditors are expected to assess higher inherent risk for firms with aggressive discretionary accruals and thus higher audit effort and charge higher fees.

Hence, the relation between audit fees and financial reporting quality, including earnings management, has received considerable attention in audit research. However, prior literatures on the association between discretionary accruals and audit fees primarily investigate whether audit fees or total fees affect the level of discretionary accruals, which look at audit quality from the side. But, there is a few studies that investigate whether discretionary accruals affect on the determinants of audit fees (or hours). Furthermore, studies examining the relation between the levels of discretionary accruals and audit fees document report mixed results. For example, Gul et al. (2007) show a positive relation between discretionary accruals and audit fees. In contrast, Park et al. (2012) show a negative relation between discretionary accruals and audit fees. In addition, Park (2005) find no relation between discretionary accruals and audit fees on average. Despite a growing literature on the determinants of audit fess and hours of auditor and also financial reporting opacity (Hutton et al., 2009; Kim and Zhang, 2014; Kang and Choi, 2016 etc.) respectively, there is little evidence of whether and why the financial reporting risk associated with opaque earnings influences audit production. To extend this line of research, we do attempt to higher opaque earnings relate to higher inherent risk or audit risk which translates into higher audit effort and higher audit fees in both cases. This is, we predict that firms with opaque financial reporting have higher both audit fees and audit hours because auditors perceive these firms to be more audit risk.

For analysis, the first proxy (OPAQUE) of this study is the earnings management measure of information opaqueness of Hutton et al. (2009). Following Hutton et al. (2009), this paper measures financial reporting opacity as the prior three years moving sum of absolute value of discretionary accruals, where discretionary accruals (DA) are estimated by the modified Jones model (Dechow et al., 1995). The second proxy (OPAQUE_std) of this study is the earnings management measure of information opaqueness, which is prior three years standard deviation of discretionary accruals as documented more econometrically by our suggestion. These measure intends to capture both the abnormally high accruals in the year of overstatement and the subsequent reversal of prior accrual overstatements. This study sample covers listed firms in non-financial industries with fiscal year-end in December from 2004 to 2015.

The empirical findings of this paper are following. First, after controlling for several factors that affect audit fees and audit hours, we find that audit fees significantly increase for opaque earnings clients over the past three years. These result is consistent with this study using both OPAQUE and OPAQUE_std measures. Second, we find that auditors increase audit hours for opaque earnings clients, suggesting that opaque earnings positively influence audit efforts. These result suggests that auditors perceive opportunistic earnings management activities as increase in audit risk. Our results are robust to a variety of sensitivity checks such as an alternative specification. However, we do not find the positive association between our first proxy for opaque earnings (OPAQUE) as documented by Hutton et al. (2009) and audit hours. Whereas, we find the positive association between our second proxy for opaque earnings (OPAQUE_std) and audit hours. Third, consistent with prior research (Park et al., 2012), we find the negative association between discretionary accruals with a single-year measure (DA) and audit fees as well as audit hours. Fourth, when we partition our samples into Big 4 and non-Big 4 auditor subsamples, we find the positive association between opaque earnings and audit fees is more significant for non-Big 4 auditees than Big 4 auditees. However, we find the positive association between opaque earnings and audit hours is more significant for Big 4 auditees than non-Big 4 auditees. These provides evidence consistent with the extant literature on audit quality that Big 4 auditors have more reputation concerns to protect than non-Big 4 auditors and are therefore more concerned about risk of financial reporting opacity as a result of managerial opportunism. Thus, Big 4 auditors would exert more effort than non-Big 4 auditors. Finally, when we partition the samples into pre- and post-2014 period (position-specific audit hours disclosure since 2014), we find that the positive relation between opaque earnings and audit hours is stronger during post-2014 period. Overall, using two proxies for opaque earnings (or financial reporting opacity), we provide novel evidence that the audit fees and audit hours increases significantly with financial reporting opacity, whereas audit fees and audit hours decreases significantly with the level of discretionary accruals in the short-run. In particular, we confirm that our firm-specific measure of opacity (OPAQUE_std), which prior three-year standard deviation of discretionary accruals is a reliable predicator of audit fees as well as audit hours, compared with opacity measure by Hutton et al. (2009).

Our study adds to the literature in the following ways. First, our study suggests that improving financial reporting opacity can increase auditors’ perception of audit risk. Moreover, while Hutton et al. (2009) employ a self-constructed measure of opacity, we also use a new objective measure opacity, such as prior three years standard deviation of discretionary accruals level, which considered to be refine proxy for low-quality financial information. Second, our study is related to the audit pricing literature. To the best of our knowledge this is the first study to examine the relation between opaque earnings and audit fees as well as audit hours. Lastly, this paper contributes to a deeper understanding of these determinants by examining how financial reporting opacity affects audit fees and audit hours. Therefore, the results of this paper may provide useful information to academics as well as investors and regulatory bodies that evaluate the reliability of external audit by using audit fees and audit hours information.

Keywords:

키워드: