ABSTRACT
We examine audit firms’ responses to increased legal liability. Using audit-effort data in China, we find audit firms increase audit effort when they are forced to switch from limited liability corporations (LLCs) to limited liability partnerships (LLPs). Audit fees also increase after the switch from LLCs to LLPs. Furthermore, we show the increase in audit effort is greater for clients that are in financial distress and for clients that report higher discretionary accruals. Overall, our study suggests audit firms respond to increased legal liability by increasing their audit effort, and the increase in audit effort is more pronounced for risky clients.
Acknowledgments
We are especially grateful for the detailed and constructive suggestions from the editor and the anonymous reviewers. We acknowledge financial supports from the National Natural Science Foundation of China (Grant No. 71872078, 71872046, 71572038, 72173108, 71472047).
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1. Please refer to http://www.gov.cn/zwgk/2010-07/23/content_1662348.htm.
2. In 2001, the Chinese Institute of Certified Public Accountants (CICPA) issued a notice requiring a trial filing system for the audit of annual financial statements of all listed companies (please refer to https://lawn.esnai.com/view/54610). Specifically, the CICPA required audit firms to disclose the number of staff and the number of days used to audit each client before May 31 of each year. Information on audit hours is obtained from the CICPA’s database, which contains information on audit hours for all client-auditor pairs in each fiscal year. Because the proprietary audit-hours data (not publicly obtained) are for the period 2004-2011 only, our sample ends in 2011.
3. A recent study by He, Pan, and Tian (Citation2017) examines how shifting from LLCs to LLPs affects audit quality and finds LLPs lead to increased going-concern opinions, less earnings management, and higher audit fees. Our results are consistent with He, Pan, and Tian (Citation2017). However, by using audit-effort data, we more firmly establish the impact of shifting from LLCs to LLPs on audit quality.
4. Negative working capital (i.e. when the current assets are less than the current liabilities) indicates that the firm’s short-term solvency (i.e. liquidity) has suffered serious problems. Sibilkov (Citation2009) finds that low liquidity will increase the discount effect of asset liquidation and increase the expected costs of financial distress. Therefore, we believe negative working capital should predict a higher likelihood of financial distress (DISTRESS). Our results remain robust to the exclusion of negative working capital when we define DISTRESS.
5. We believe signed discretionary accruals are a more appropriate measure of audit risk, because auditors’ litigation risk is greater in the presence of income-increasing earnings management (e.g. Heninger Citation2001) and auditors work harder to detect income-increasing misstatements than income-decreasing misstatements (Barron, Pratt, and Stice Citation2001).
6. We start our sample from 2007 because China adopted the new accounting standards with international convergence in 2007. Our sample ends in 2011 because we do not have the audit-effort data afterwards.
7. Please refer to http://czj.sh.gov.cn/zys_8908/zcfg_8983/zcfb_8985/hj_9035/hjssws_9041/20,110,414/0017-157,518.html.
8. After correcting for the downward bias in the successor’s partial-year audit fee, Barua, Lennox, and Raghunandan (Citation2020) find no evidence of fee discounting in initial-year audit engagements, which indicates audit fees are significantly higher during the auditor-change year.
9. For example, audit firms tend to become LLPs if their client portfolios are less risky on average. In addition, the degree of market development of the region where the audit firm is located could have an impact on the auditor’s behavior. Thus, we use a composite index of regional market development (LOCAL_ECON) for the province where the client firm’s auditor is located. The higher the LOCAL_ECON index, the more developed the province where the audit firm is located.
10. The conclusion of their study is that once Rzu2 < Rxz2Rxu2, the IV estimators are preferable to OLS estimators even if the IV is ‘semi-exogenous.’ Although calculating the Rzu2 and Rxu2 is impossible because the real residual is unknown (we only have the estimated residual), we can do a qualitative analysis to evaluate the validity of IV.
11. The most common design choice is one-to-one matching without replacement and the most commonly used caliper distances are 0.01, 0.03, and 0.10 (Shipman, Swanquist, and Whited Citation2017).
12. In Model (7), we do not control for year fixed effects, because we have included POST to control for time-varying characteristics that affect both LLP and non-LLP audit firms.