![Sample our Economics, Finance,Business & Industry journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5931/TOC-Subject-Sample-2021-Economics-Finance-Business-Industry.jpg"})
Abstract
Small audit firms try to reduce their resource constraints in serving clients by creating inter-organisational relationships (IORs) through membership in accounting networks or in agreements with individual firms. Through these relationships, firms have access to audit manuals/methodology, personnel, and may perform joint audits and joint marketing of their audit services. We identify the existence of IORs from PCAOB Form 2 filings and test whether audit quality and fees vary cross-sectionally as a function of the specific resource obtained from the IOR. We find evidence that firms with access to audit technology through IORs charge higher audit fees and provide higher quality audits, in terms of smaller clients’ performance-matched discretionary accruals, higher auditors’ likelihood of issuing going concern opinions to financially distressed clients, and fewer audit deficiencies identified in PCAOB inspection reports. The findings highlight the importance of access to audit technology through IORs in improving small audit firms’ real and perceived audit quality.
Acknowledgements
Professor Juan Mao thanks the financial support from University of Texas at San Antonio and Samuel Garlett and Si Shen for their excellent research assistance. We greatly acknowledge helpful comments and suggestions from participants at the 2022 PCAOB Conference on Auditing and Capital Markets.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
Data are available from the public sources cited in the text.
Notes
1 Legitimacy theory (e.g. Perrow Citation1961, Suchman Citation1995) has been applied in strategic management to explain the existence of strategic alliances from the institutional perspective (Dacin et al. Citation2007). Dacin et al. (Citation2007) note that ‘Legitimation refers to the social justification of an actor or activity, such that the actor or activity is publicly validated or endorsed’.
2 In sensitivity analyses, we exclude audit firms that are annually inspected by the PCAOB (Crowe Chizek and Company LLC, McGladrey & Pullen LLP, Crowe Horwath LLP, MaloneBailey, LLP, ParenteBeard LLC, McGladrey LLP, Marcum LLP, RSM US LLP, Cohen & Company Ltd (OH), and Crowe LLP) from our sample of non-Big 6 auditors, retaining only audit firms that are triennially inspected. Our main conclusions do not change.
3 This number is calculated based on audit opinions issued by U.S. audit firms for the year 2018 (Audit Analytics). In total, 20,443 audit opinions are issued in 2018 and 5399 are issued by non-Big 6 auditors.
4 Comparable Non-Big 6 global data are unavailable. However, global data on the Non-Big 4 show that such firms enjoy a 26% market share (IAB Citation2021).
5 According to GAO’s survey, 82% of large public companies (i.e. the Fortune 1000) limited their auditor choices to three or fewer firms and about 60% viewed competition in their audit market insufficient (GAO Citation2008).
6 Bills et al. (Citation2016a) first identify the association list from CCH and other sources from 2010 to 2013. They contacted each association on the list to obtain its member list and for those that did not provide the member lists they collect the member lists from the association website.
7 Calculated by the authors based on all Form 2s that PCAOB registered firms filed with the PCAOB during 2010 to 2018. This is an audit-firm level statistic, and the number is smaller than what is calculated based on client-year observations. Over 60 percent of non-financial non-Big 6 clients with assets over one million in our sample are audited by auditors engaged in IORs (untabulated).
8 Bills et al. (Citation2018, Citation2021) seek to identify the IOR-sourced resources that audit partners perceive as contributing to improved audit quality or audit market competitive advantage. In contrast, we attempt to assess the actual (as opposed to perceived) association between IOR-sourced resources and audit quality/fees.
9 ‘Each registered public accounting firm shall submit an annual report to the Board, and may be required to report more frequently, as necessary to update the information contained in its application for registration under this section, and to provide to the Board such additional information as the Board or the Commission may specify, in accordance with subsection (b)(2).’ https://pcaobus.org/About/History/Documents/PDFs/Sarbanes_Oxley_Act_of_2002.pdf
10 To vary the exposition, we occasionally use the generic term ‘other firm’ rather than ‘Identified Entity.’
11 To be clear, not all are sanguine about the effects of investment in functional audit technology. Dowling and Leech (Citation2007) interview 4 audit partners who note audit decision aids can harm audit quality if auditors mechanistically apply the technology in a ‘check the box’ manner.
12 We acknowledge that the wording of the question is unclear because the question specifies joint audits whereas group audits (not joint audits) typically are performed in the U.S. This ambiguity was noted in a letter from PricewaterhouseCoopers (PwC) to the PCAOB in which PwC offered feedback on the PCAOB-proposed Form 2 annual report. PwC wrote ‘Some of the relationships that firms must disclose with respect to audit-related memberships, affiliations and similar arrangements are defined in terms that are overbroad or ambiguous, thereby creating uncertainty about what must be reported to the Board. For example, Item 5.2.a.2 refers to ‘joint audits’ . . .’ However, the wording of Question 5.2.a(2) was not changed in response to the PwC letter.
In email correspondence, we sought clarification from the PCAOB as to the meaning of ‘joint audit’ in Question 5.2(a)(2). The email reply received from PCAOB was: ‘’Joint audits,’ in Item 5.2 of Form 2, is undefined . . . The PCAOB has not otherwise made any public pronouncements on the meaning of ‘joint audits’ in Item 5.2 of Form 2, and the staff of the PCAOB cannot provide legal advice to private parties.’ Since survey evidence in Bills et al. (Citation2018) makes clear that firms participating in an IOR often market and provide group audits with their strategic partners, we interpret this question as seeking to identify not only instances in which a joint audit is to be performed with an Identified Entity, but also instances in which a group audit is performed with the Identified Entity.
13 Burke et al. (Citation2020) document that mere use of component auditors does not lead to lower audit quality but the extent of participation of component auditors is negatively associated with audit quality.
14 Carson et al. (Citation2022) examine the audit quality and audit fee effects of group audits by Australian auditors. The study finds that involvement of a component auditor is associated with higher audit quality, but audit quality decreases with the amount of work performed by the component auditor.
15 We attempted to implement more stringent fixed effects (i.e., client fixed effects, auditor fixed effects) but abandoned this effort because (1) non-linear models failed to converge with these fixed effects and (2) p-values of test variables in linear models declined and often were statistically insignificant. The nonconvergence in non-linear models can be attributed to the inherent challenges in maximizing a likelihood function when dealing with numerous fixed effects that require estimation. Loss of significance in linear models occurs because a fixed effects specification has the effect of demeaning each variable by the mean value of that variable within each fixed effect unit (e.g. each client firm for client fixed effects), such that the analysis becomes an examination of the association between across-time deviations from the mean value of the dependent variable and across time deviations from the mean value of each independent variable. When explanatory variables tend to be ‘sticky’ (i.e., the value in year t is similar to the value in year t-1, which is characteristic of TECHNOLOGY, MARKETING and PERSONNEL), the deviations from means tend to be minor, causing significance levels to decline.
16 The starting year of 2010 is the first year for which we can obtain PCAOB Form 2 filings. The ending year of 2018 is the last year of data availability at the time of data collection.
17 We match the Form 2 data into the inspection data if the year of Form 2 is equal to the year of inspection.
PCAOB Form 2 is filed by registered audit firms before June 30 each year to cover their activities from April 1 of last year to March 31 of the current year. For example, PCAOB Form 2 for 2018 will cover the audit firm’s activities from April 1, 2017 to March 31, 2018. PCAOB inspection reports are typically issued for audits most recently completed by the firm. If the PCAOB issued an inspection report in 2018, it typically refers to the audits done by the firm in 2017. https://pcaobus.org/oversight/inspections/inspection-procedures ‘The inspection team generally selects the audits most recently completed by the firm but may also select audits completed in prior years if, for example, there are no recently completed audits.’
18 All continuous explanatory variables are winsorised at the 1 and 99 percent levels.
19 In sensitivity analyses reported in , we also use a variety of other discretionary accruals used in prior studies, and our results are robust to those measures.
20 Our additional analyses reported in Panel A suggest that these higher fees associated with PERSONNEL are mainly driven by clients audited by audit firms in smaller IORs.
21 This is consistent with the population of all 3499 inspection reports as of March 2022 (0.104 and 0.524 for GAAP_DEF and GAAS_DEF, respectively).
22 We use a truncated regression model in the analysis of signed discretionary accruals due to the truncation in the dependent variable created by the partition.
23 We find some evidence of lower audit quality when audit firms have access to joint marketing and provision of audits when the dependent variable is the audit firm’s likelihood of having GAAS deficiencies in its inspection report.