The effect of enforcement action on audit fees and the audit reporting lag

Abstract

We study the effect of audit market regulation on auditor contracting by examining a sample of enforcement actions on engagement auditors by the China Securities Regulatory Commission (CSRC). We conjecture that sanctions change auditor behaviour through increased effort and diligence. Specifically, we hypothesise that sanctioned auditors would (a) increase audit fees and (b) increase reporting lag. We find results consistent with these hypotheses. We also provide supporting evidence using a battery of output measures of audit quality including abnormal accruals (various kinds), modified audit opinion, F-score, and restatements. Thus, both input-based and output-based indicators show that sanctioned auditors improve audit quality. Additionally, in nuanced tests, we present evidence on other aspects of auditor behaviour: we find that audit firms combine sanctioned auditors with more seasoned co-partners and assign a lighter workload. By far, this is the most direct and compelling evidence that auditors respond to sanctions. Overall, sanctions appear to encourage sanctioned auditors as well as their firms to take a concerted and strategic approach to improve audit quality.

Keywords:

• governmental sanctions
• audit fee
• audit reporting lag

JEL Classification:

• M49

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 The China auditing market is a predominantly buyer's market (Chen et al. 2010). By way of preliminary information gathering, the authors of this study informally phone-interviewed 11 partners from Big 4 and top 10 Chinese accounting firms in 2015 and queried them whether CSRC sanctions would affect auditors. The purpose of these interviews was to understand whether sanctions were perceived as significant and whether there was merit in pursuing the empirical tests we present in our paper. Based on these interviews, the authors concluded that individual auditors were deeply concerned about the prospect of sanctions. In particular, auditors were concerned about whether sanctions would affect client acquisition and retention. These informal findings are largely in line with those reported by Acito et al. (2018) about PCAOB inspections. We hasten to add that these interviews provide no more than informal or anecdotal evidence on the potential effects of CSRC sanctions.

2 For example, in 1998, the CSRC punished audit firm Cheng Du Shu Du for failing to detect fictitious sales and inflated inventory posted by its client PT Hong Guang. As part of the punishment, the CSRC imposed a lifetime ban on two engagement partners, Yingqin Wang and Xiuhua Zhang.

3 Auditor behaviour can also change prior to potential sanctions as shown in an experimental study by Ugrin and Odom (2010).

4 In tests using the treatment group only, there is no evidence that discretionary accruals decrease post-sanctions. Also, in the difference-in-differences specification where both treatment and control samples are included, the result of a decrease in accruals is significant at the 10 percent level.

5 Our sample is significantly larger than the ones in Firth et al. (2014) and Sun et al. (2016).

6 DeFond (2010) challenges the traditional perspective that litigation and reputation concerns drive audit quality and raises the possibility that oversight mechanisms may be equally influential. He also observes that not much is known about the effectiveness of oversight mechanisms.

7 Chen et al. (2010) provide a detailed overview of institutional improvements in China during 1995–2004. Major financial reporting scandals that led to these changes involve firms such as ZhengBaiWen, MonkeyKing, Yorkpoint, YinGuangXia, and ZhongTianQin.

8 A related question concerns whether audit failure triggers an inspection report denoting deficiency. Gunny and Zhang (2013) report evidence of this link. That is, they show evidence of an association between reports with deficiency and lower audit quality.

9 In a recent, innovative study, Acito et al. (2018) convert information about deficiency, typically an auditor-level variable, into a firm-level variable by relating auditor-level information about deficiency related to specific standards to the importance of such standards to a client firm.

10 Recently, the PCAOB adopted Rule 3211 (Auditor Reporting of Certain Audit Participants, Public Company Accounting Oversight Board (PCAOB) 2015) which requires disclosure of audit engagement partners in Form AP filed on or after 31 January 2017. However, under this rule, U.S. accounting firms provide the firm's name instead of individual name(s) in the audit report. Engagement partners sign and file Form AP directly on the PCAOB website no later than 35 days after the audit report is filed to the SEC. Thus, there are differences in implementation compared to the CSRC situation.

11 We also construct a matching group using a one-to-one method and find that the results are similar to the ones reported. See the detailed explanation in section 5.4.2.

12 According to the Certified Public Accountants Law of People's Republic of China, the CPA qualification is achieved through exams after 1993. Prior to this, the qualification was achieved through work experience.

13 We examine the parallel trends assumption in the difference-in-differences specification by regressing the dependent variable against the treatment dummy (Enforce) as well as the interaction between Enforce and various years prior to and following the year of the sanction. In the vast majority of the cases, the coefficient of the interaction variable is insignificant, supporting the parallel trends assumption.

14 In untabulated tests, we use the residual from the cross-sectional version of the modified Jones (1991) model used in DeFond and Jiambalvo (1994) and the standard deviation of abnormal working capital accruals developed by McNichols (2002) as other measures of audit quality. We find similar results using these alternate measures.

15 We acknowledge the alternate explanation that the reduction in the workload is a natural consequence of sanctions wherein sanctioned auditors lose engagements. This alternate explanation relates to demand rather than supply for services from sanctioned auditors.

16 Goodwin and Wu (2016) find there is no causal relationship between audit partner busyness and audit quality, but they also note that increased workload degrades audit quality in a period of ‘stress’ (the period of accounting scandals). The period following sanctions is plausibly one with elevated stress. In addition, it is ‘difficult to predict the sign of the association between partner workload and audit quality’ (Lennox and Wu 2018, p. 9).

17 We also checked the dates when the sanctioned engagement auditors exit from the auditing market so that we can exclude the alternative explanation that they quit because of retirement.