Comments of the Auditing Standards Committee of the Auditing Section of the American Accounting Association on the PCAOB’s A Firm’s System of Quality Control and Other Proposed Amendments to PCAOB Standards, Rules, and Forms Free

We are pleased to provide feedback on the Public Company Accounting Oversight Board’s (PCAOB’s) proposed quality control (QC) standard, A Firm’s System of Quality Control and Other Proposed Amendments to PCAOB Standards, Rules, and Forms (Public Company Accounting Oversight Board (PCAOB) 2022). Given the magnitude of the proposal, the timeline for comments, and the Auditing Standard Committee’s previous response (Brown et al. 2020) to the related Concept Release (Public Company Accounting Oversight Board (PCAOB) 2019), we focus on two key areas in this response: (1) the overall standard and selected questions and (2) recent research that we encourage the PCAOB to consider.¹ In the sections below, please note our focus on balancing processes and outcomes as well as the possible implications of the PCAOB’s accountability focus for attracting and retaining audit talent.

We commend the PCAOB’s principles-based approach to QC in that it affords firms the opportunity to appropriately respond to the unique quality risks identified with reference to firm and engagement circumstances. In this regard, we note the wealth of literature informing firms of the ways in which they may be able to effectively establish quality objectives, identify and assess quality risks, design and implement quality responses, and monitor the QC system. We are unable to cover all of this literature in our response, and the relevance of the research will be contingent on firm and engagement circumstances (see, for example, Knechel, Krishnan, Pevzner, Shefchik, and Velury (2013), Simnett, Carson, and Vanstraelen (2016), and Simnett and Trotman (2018) for broad overviews of this research). We do, however, consider some of the more broadly applicable research in our response to Questions 81, 82, and 84 in the next section.

In considering the overall proposed standard, we concentrate on the higher-level, principles-based requirements and the incremental requirements beyond those that are required by International Standard on Quality Management (ISQM) 1 (International Auditing and Assurance Standards Board (IAASB) 2020) and Statement on Quality Management Standards (SQMS) 1 (American Institute of Certified Public Accountants (AICPA) 2022a). In summary, we speak to our concerns around the requirements for firms that have not and do not plan to perform engagements pursuant to PCAOB standards (Question 5), the accountability and certification requirements (Questions 12, 21, 25, and 62), the explicit recognition and required responses to risks of intentional misconduct (Questions 16 and 17), assessing quality risks independent of quality responses (Question 19), and the conditional requirements applicable to firms that issue audit reports with respect to more than 100 issuers (Questions 22, 23, 28, and 47).

On the basis of our review of the research, we do not believe that it is appropriate to require firms that have not and do not plan to perform engagements pursuant to PCAOB standards to design a QC system in accordance with QC1000. We come to this conclusion on account of the difficulties firms will likely have in designing a QC system based on largely hypothetical circumstances and the likely inappropriateness of the previously designed QC system should the need for it to be implemented and operated arise. While we acknowledge the PCAOB’s statutory mandate as it relates to QC standards applicable to all registered firms, we believe that the design requirements applicable to firms that have not and do not plan to perform engagements pursuant to PCAOB standards be limited to client acceptance components.

To understand what would be expected (and the challenges therein) of firms required to design a QC system to address the unique quality risks in a hypothetical future engagement, we turned to research on scenario planning (see Amer, Daim, and Jetter (2013) for a review of this literature). Audit firm personnel designing a QC system to address hypothetical future circumstances will face many challenges and will need to overcome numerous cognitive biases (see Schoemaker (2004) for a discussion of some of these challenges). The challenges cited by the PCAOB that limit the practical implementation and operation of a QC system when there is no engagement are also evident in the design of a QC system when there is no engagement.

The designed QC system would inevitably need to be modified if it was ever to be implemented and operated. Such modification would be necessary in light of the realized circumstances present when the firm needs to perform an engagement pursuant to PCAOB standards. These modifications are unlikely to be sufficient. Auditors have been shown to be susceptible to an anchoring bias in different judgment and decision-making settings (Joyce and Biddle 1981; Harding and Trotman 2009; Pike, Curtis, and Chui 2013). When judgments and decisions are influenced by anchoring bias, a salient starting point is overrepresented in the final judgment/decision that is made (Tversky and Kahneman 1974). That is, there is insufficient adjustment from the anchor.

Consistent with the intrusion of an anchoring bias in risk assessment, auditors are often not sufficiently responsive to new circumstances that come to their attention during the audit and normatively require an amendment to the original audit plan (Glover, Jiambalvo, and Kennedy 2000; Hoffman and Zimbelman 2009; Bauer, Hillison, Peecher, and Pomeroy 2020). Our concern, therefore, is that should a previously designed QC system subsequently need to be implemented and operated, there likely will be insufficient amendments to the original system to take into account the realized and contemporary circumstances of the environment, firm, and engagement that have given rise to the need to implement and operate the plan. We are concerned that a QC system designed for a hypothetical future situation, and amended for actual realized circumstances, will not be effective in achieving the reasonable assurance objective.

Requiring firms that have not and do not plan to perform engagements pursuant to PCAOB standards to design a QC system in accordance with QC1000, therefore, is likely to be costly and realize little, if any, benefit. Indeed, it may even have the unintended consequence of inappropriate QC systems being implemented and operationalized. In such a situation, the confidence believed to vest with prospective clients by requiring firms to have a PCAOB-compliant QC system designed and ready for implementation and operation may be misplaced.

We believe that a more effective QC system would be designed with knowledge of the actual circumstances of the engagement being undertaken pursuant to PCAOB standards. For firms that have not and do not plan to perform engagements pursuant to PCAOB standards, the PCAOB may wish to limit the design of the QC system to components associated with client acceptance quality objectives.

The proposals in QC1000 emphasize and envisage greater accountability (especially for the firm’s principal executive officer and firm personnel assigned operational responsibility for the QC system as a whole; see paragraphs .11, .14, .15, and .79). Particularly regarding explicit certification, but also with reference to roles and responsibilities, this extends responsibilities beyond those envisaged in ISQM 1 and SQMS 1.

While the imposition of accountability is largely positive in that it, for example, mitigates agency costs and controls effort and behavior (Quinn and Schlenker 2002; Stewart, Snyder, and Kou 2023), improves decision-making (Hall, Bowen, Ferris, Royle, and Fitzgibbons 2007), reduces unethical behavior (Pitesa and Thau 2013), and minimizes self-superiority beliefs (Sedikides, Herbst, Hardin, and Dardis 2002), it is not a panacea. We are concerned that the negative consequences of increased accountability have not been fully considered in the decision to increase accountability requirements.

Research indicates that increased accountability can have negative effects, including a reduced sense of responsibility (Dose and Klimoski 1995), amplified cognitive biases (Lerner and Tetlock 1999), increased stereotyping (Gordon, Rozelle, and Baxter 1989), cognitive laziness (Tetlock 1992; Hall, Frink, and Buckley 2017), squandered resources (Adelberg and Batson 1978), and increased job tension and emotional exhaustion (Hall et al. 2006). Furthermore, accountability can be manipulated by those who are politically skilled, for instance, through the use of impression management to increase trust and lower the extent to which they are held accountable (Hall et al. 2017). As concluded by Lerner and Tetlock (1999), only certain subtypes of accountability lead individuals to exert greater cognitive effort. Accountability interacts with decision-maker characteristics and the environment to produce a variety of effects, not all of which are positive.

Peecher, Solomon, and Trotman (2013, 597) speak to many of these accountability issues in an audit regulatory setting and conclude that “auditors’ current accountabilities are unlikely to motivate them to target high quality financial statement audits or to invest in research and development activities that would improve longer term audit quality.” They propose that an accountability framework for auditors should more effectively balance rewards versus penalties and process versus outcomes. In this regard, we note that while QC1000 proposes accountability for the operation of the QC system (which would capture elements of process and outcome), the reporting/communication associated with these accountabilities is focused on negative outcomes (e.g., violations), and the certification proposed in paragraph .79 is outcomes based and negatively oriented.

We therefore encourage the PCAOB to balance accountabilities in proposed QC1000 such that processes and outcomes, as well as rewards and penalties, are more appropriately weighted. For example, in addition to specifying responsibilities and accountabilities for QC deficiencies (see paragraphs .17 and .58–.76), requirements could also include responsibilities and accountabilities for QC breakthroughs (i.e., identified ways of doing things better). By doing so, we believe that this will not only help realize the benefits of accountability but also encourage auditors to excel and not simply aim to meet minimum quality thresholds to avoid sanction (see also our response to Questions 21 and 62).

On the basis of research that we have reviewed, we do not believe that the proposed definition of quality risks should explicitly address risks of intentional misconduct by firm personnel and other participants. Intentional misconduct, by its nature, would have a reasonable possibility of adversely affecting the firm’s achievement of quality objectives and, since these risks are not assessed with reference to the possibility of occurrence, would represent a quality risk requiring a response. Our concern is that the necessary response, especially as it relates to monitoring of colleagues’ behavior and incidences of intentional misconduct, may negatively impact intraorganizational trust and have a detrimental effect on the firm’s ability to achieve quality objectives.

Research in psychology and management highlights that monitoring for incidences of intentional misconduct can signal and engender distrust for and of employees (Frey 1993). Monitoring of a particular behavior signals that the organization expects such behavior to occur (Cialdini 1996). This erosion of trust can have significant deleterious effects on employee behavior and attitudes and team performance (see Kramer (1999) for a review and De Jong, Dirks, and Gillespie (2016) for a meta-analysis). To illustrate, trust is important in facilitating communication, cooperation, learning, and performance quality and quantity (Burke, Sims, Lazzara, and Salas 2007). Trust also increases “organizational citizenship behavior,” motivating employees to perform tasks above and beyond those specifically assigned to them (Organ 1990; Rousseau and McLean Parks 1993). To the extent that trust is eroded by monitoring for intentional acts of misconduct, the QC system, which will be relying on the behaviors that trust engenders, will be less effective than would otherwise be the case.

In an audit setting, trust within audit teams is associated with higher-quality audit outcomes (Pratt and Jiambalvo 1981; Nelson, Proell, and Randel 2016; Bauer, Estep, and Malsch 2019). The psychological safety that accompanies trust facilitates knowledge exchange and learning. To illustrate the potentially deleterious consequences of an erosion of trust in audit teams and firms, trust is positively associated with employee voice and elevates a willingness to whistle blow (Seifert, Stammerjohan, and Martin 2014; Wilson, McNellis, and Latham 2018). To the extent that explicitly addressing risks of intentional misconduct by firm personnel erodes trust, we fear that, rather than contributing to, it will constrain the achievement of quality objectives. We are not suggesting that the firm should ignore risks of intentional misconduct but rather suggest that this may be more effectively considered and responded to as part of the broader understanding of quality risks and not explicitly addressed.

In our response to Question 16, we note our concerns with explicitly addressing risks of intentional misconduct. We believe that doing so will undermine trust within the firm and audit teams and make it less likely that quality objectives will be achieved (Frey 1993; Cialdini 1996; De Jong et al. 2016).

If the PCAOB decides to continue to explicitly address risks of intentional misconduct, we believe that the threshold of “reasonable possibility of occurring” should apply. Intentional misconduct speaks to a questionable ethical and moral disposition, which would have a pervasive effect on all dimensions of QC. So, all acts of intentional misconduct within the risk population, irrespective of their possibility of occurrence, would represent a quality risk. In designing, implementing, and operating responses to all intentional acts of misconduct, the firm would be embedding a culture of distrust and institutionalizing constraints on the achievement of audit quality. If intentional misconduct is to be explicitly addressed, we believe it should be subject to the same threshold as other risks, thereby minimizing the negative effects on trust.

We commend the inclusion of a note to paragraph .20(b) emphasizing that the assessment of quality risks is based on inherent risk (i.e., regarding the effect of any related quality response), in that it reduces the possibility of quality risks being de-emphasized on the potentially misplaced belief that there are effective quality responses. However, we note that auditors may have trouble achieving this.

As noted in the commentary accompanying proposed paragraph .20(b), the identification and assessment of quality risks is similar to the approach to assessing inherent risk in Auditing Standard (AS) 2201 (Public Company Accounting Oversight Board (PCAOB) 2007). Regarding assessing inherent risk, Miller, Cipriano, and Ramsay (2012) report results consistent with the understanding that auditors assess inherent risk assuming an average or expected level of controls. This finding is consistent with earlier studies examining the audit risk model, which speak to the interdependencies between inherent and control risk and the challenges in distinguishing inherent and control risk factors (Cushing and Loebbecke 1983; Waller 1993; Messier and Austen 2000). To the extent that this may also be the case when identifying and assessing quality risks and that quality risks may be overlooked (or underemphasized) on the belief that they will be addressed by expected (but not necessarily realized) quality responses, a note accompanying paragraph .20(b) may not be sufficient to prompt or remind auditors of the independence of quality risks from quality responses. Supplemental direction may be needed.

As we note above in our response to Question 12, we believe that the quality objective of holding leadership accountable (paragraph .25(b)) is appropriate, but that such accountability should not be limited to deficiencies and sanctions, and focused on outcomes, but extended to process and an acknowledgement of positive behaviors and rewards (see Peecher, Solomon, and Trotman 2013). We are concerned that the quality objectives as presently drafted stifle initiatives to exceed minimum standards and give rise to the possibility that the QC system will not be sufficiently responsive, over time, to a rapidly changing environment (including technological advancements). We encourage the PCAOB to consider extending governance and leadership objectives to explicitly encourage superior performance and highlight the need to consider risks to the achievement of such performance and responses to those risks (and not just discourage inferior/inappropriate performance and actions).

We have noted the conditionality of several specified quality responses in that they are applicable only to firms issuing audit reports with respect to more than 100 issuers. Based on our review of the research, we do not believe that conditional requirements are appropriate, especially as they relate to conditions associated with firm size. To the extent that a specified quality response is considered important, then we believe it should apply to all firms.

We are concerned that the conditional requirements associated with firm size may give rise to audit quality differences between larger audit firms that issue audit reports for more than 100 issuers and smaller audit firms that issue audit reports for less than 100 issuers. The evidence on audit quality effects associated with firm size is inconclusive (see, for example, Lawrence, Minutti-Meza, and Zhang (2011) and DeFond, Erkens, and Zhang (2017) for competing views). If larger firms (i.e., firms that issue audit reports for more than 100 filers) are required to implement a beneficial quality response and smaller firms (i.e., firms that issue audit reports for less than 100 filers) do not need to implement such a response, there is a risk that real audit quality differences may arise.

Moreover, to the extent that participants in the financial reporting ecosystem perceive that audit quality deteriorates with reductions in firm size, actual variation in audit quality is a moot point. It is argued that participants in the financial reporting ecosystem use auditor size as a way to evaluate audit quality (which is largely unobservable; e.g., DeAngelo 1981), and research findings are consistent with the understanding that larger audit firms are perceived to provide higher levels of audit quality (Knechel, Naiker, and Pacheco 2007; Boone, Khurana, and Raman 2010; Kilgore, Harrison, and Radich 2014).

If larger firms are subject to the implementation of specified quality responses and smaller firms are not and financial reporting ecosystem participants become aware of the differing expectations, there is a risk that perceptions of poorer audit quality in smaller firms (whether well founded or not) will be exacerbated. To the extent that this is the case, rather than enhancing capital market confidence in the audit work of those in smaller firms, there is a risk that the conditional requirements based on firm size will undermine that confidence. We, therefore, encourage the PCAOB to carefully consider the use of conditional requirements in QC1000. Where the quality response is seen to be appropriate in some circumstances and not others, it need not be a specified quality response, and firms can consider its appropriateness in light of the circumstances in each situation.

As we note above in our response to Question 22, we believe that conditional requirements associated with firm size risk exacerbating perceptions that audit quality is lower in smaller firms (Knechel et al. 2007; Boone et al. 2010; Kilgore et al. 2014). The specified quality responses should apply to all firms or, alternatively, be left to the individual firms to reflect on their applicability in achieving quality objectives.

In addition to our concerns regarding the conditionality of the requirement to incorporate an oversight function, we have further concerns around the requirement to have an independent member as part of that oversight function. The independent member(s) would be in the minority (often quite significantly), and oversight/monitoring functions are more effective with a greater proportion of independent members (see DeZoort, Hermanson, Archambeault, and Reed (2002), Bédard and Gendron (2010), and Carcello, Hermanson, and Ye (2011) for reviews of the literature on audit committees). We see little merit in requiring an independent member to be part of the oversight function without also considering the balance of the oversight function and the contribution of independent member(s) to that balance.

Given the potential negative consequences of increased accountability (see our response to Question 12), the PCAOB may wish to consider including additional specified quality responses (or an elaboration of paragraph .27) focused on responses to address the risks to audit quality of increased accountability. In our response to Question 12, and with reference to Peecher et al. (2013), we highlight the potential merit in more effectively balancing the emphasis on process and not just outcome as well as rewards and not just sanctions.

As we note above in our response to Question 22, we believe that conditional requirements associated with firm size risk exacerbating perceptions that audit quality is lower in smaller firms (Knechel et al. 2007; Boone et al. 2010; Kilgore et al. 2014). We, therefore, do not believe that the proposed threshold of more than 100 issuers is appropriate. The specific quality response should apply to all firms or, alternatively, not be listed as a specified quality response and be left to individual firms to reflect on its applicability to achieving quality objectives. We do not make any comment on the appropriateness of having an automated process for identifying direct or material indirect financial interests.

As we note above in our response to Question 22, we believe that conditional requirements associated with firm size risk exacerbating perceptions that audit quality is lower in smaller firms (Knechel et al. 2007; Boone et al. 2010; Kilgore et al. 2014). We, therefore, do not believe that the proposed threshold of more than 100 issuers is appropriate. The specific quality response should apply to all firms or, alternatively, not be listed as a specified quality response and be left to individual firms to reflect on its applicability to achieving quality objectives. We do not make any comment on the appropriateness of monitoring in-process engagements.

In our earlier response to the PCAOB concept release (Brown et al. 2020), the Auditing Standards Committee members who participated at the time expressed reservations about requiring certification of the evaluation of the QC system. Research has revealed that there is little, if any, effect of CEO/CFO certifications required under the Sarbanes-Oxley Act (Griffin and Lont 2005; Lobo and Zhou 2006; Bhattacharya, Groznik, and Haslem 2007), and the Committee was of the view that requiring QC certification would also add little value. We continue to hold that view.

In addition, the accountability envisaged in QC certification and the annual submission of Form QC is outcomes and negatively oriented and is similar to other PCAOB accountability frameworks. We also note above in our response to Question 12 that such accountability frameworks focusing on outcomes and sanctions are unlikely to motivate high-quality QC systems, instead encouraging firms to meet minimum standards and avoid sanctions (see Peecher et al. 2013). We encourage the PCAOB to reconsider whether the individual certifications of the evaluation of the QC system is appropriate.

Some questions near the end of the QC proposal seek input on other research that the PCAOB should consider. In the sections below, we discuss research that we believe is relevant to the PCAOB’s QC efforts.

Recent academic research examines several factors that can affect audit quality and suggest the need for standard setting. In the sections below, we highlight selected studies examining audit firm culture and tone at the top, engagement-level issues, use of other participants, partner uniqueness and compensation, scope of services, and audit firm inspections.

The PCAOB and IAASB have both identified the importance of creating a culture where leadership of the firm sets a tone at the top emphasizing audit quality and professionalism (IAASB 2020; PCAOB 2022). Within the current standard-setting process, we encourage the PCAOB to consider recent academic findings related to the influence of audit firm culture and tone at the top on audit quality.

Alberti, Bedard, Bik, and Vanstraelen (2020) perform a literature review of academic research from the past decade on various key factors within audit firm culture that influence audit quality. One important finding is that academic research continues to identify commercialism as the more dominant focus within firms’ cultures as opposed to professionalism. This finding is due to various firm factors, such as a marketing and profit-based focus. For example, Carter and Spence (2014) investigate multiple drivers that allow Big 4 auditors to be successful in the profession. The authors consider someone as being successful when they have reached the pinnacle of the profession at the partner level. One of their key findings is that auditors who are seeking a promotion to partner shift their focus more toward commercialism and, in turn, away from professionalism (e.g., ethics and technical auditing expertise). In addition to this study, other academic studies identify a significant movement within public accounting firms toward commercialism and away from professionalism (Guo 2016; Picard, Durocher, and Gendron 2018). A shift toward commercialism, to the detriment of professionalism, suggests the need for standard setting or possibly staff guidance.

Academic research has also studied the audit quality effects of a more commercial focus. Specifically, Pyzoha, Taylor, and Wu (2020) provide a direct evaluation of the influence of firm-level commercial versus audit quality tone at the top messaging on the quality of auditor judgments while completing a complex valuation task. The study examines the influence of firm-level tone at the top messaging when the tone emphasizes commercial goals (e.g., earnings, profits, and revenues), audit quality goals (e.g., QC, training, and independence), or a balance of both types of goals. Results show that a firm’s messaging approach can be an effective psychological mechanism to influence the quality of auditor judgments whether management engages a valuation specialist or not; that is, auditors become more professionally skeptical and, therefore, less likely to be biased by management or management’s specialist when firm-level tone at the top messaging incorporates more of an audit quality focus. These findings are especially important to consider in light of the PCAOB’s findings that auditors have a tendency to exhibit lower levels of professional skepticism when management uses the work of specialists (Public Company Accounting Oversight Board (PCAOB) 2016a, 2016b, 2017).

Academic research has studied the influence of various cultural factors at the engagement level on audit quality. Alberti et al. (2020) summarize various studies relating to the importance of audit team leadership on engagements that emphasize the need for professional skepticism and ethical behaviors. For example, Harding and Trotman (2017) find that auditors become more professionally skeptical when the partner creates a skeptical environment by encouraging the audit team members to be more skeptical of both management and their own audit judgments. Professional skepticism, ethical awareness, and quality outcomes also can be enhanced when an engagement partner is open to new ideas and reinforces training and skepticism during fraud brainstorming sessions (Dennis and Johnstone 2016; Gissel and Johnstone 2017).

Conversely, academic studies also have identified quality-threatening behaviors based on engagement leadership behaviors. Brazel, Jackson, Schaefer, and Stewart (2016) find that skeptical auditors can be penalized if they ultimately do not find a material misstatement. Also, Agoglia, Hatfield, and Lambert (2015) find that managers will reward seniors for performing the unethical act of underreporting time when the client is a more desirable one. Additionally, multiple studies find that audit quality can be threatened when auditors are concerned with pressures at the engagement level, such as meeting budgets or satisfying preferences established by management or their supervisor (Peecher, Piercey, Rich, and Tubbs 2010; Koch and Salterio 2017; Messier and Schmidt 2018). It is clear that engagement-level issues can directly affect audit quality.

Academic research provides evidence of the need for QC standard setting when audit teams use the work of other participants, most notably specialists. Hux (2017) performs a synthesis of academic research on the use of various specialists during the audit, including valuation, forensic, information technology, and tax specialists. The author discusses research that shows that the use of specialists can lead to higher-quality fraud brainstorming (Brazel, Carpenter, and Jenkins 2010) and identification of control deficiencies and misstatements (Jenkins, Negangard, and Oler 2018). Research also finds that valuation specialists are often used when auditors do not have sufficient valuation expertise to audit complex estimates and specialists play a critical role in proposing audit adjustments (Griffith, Hammersley, and Kadous 2015; Cannon and Bedard 2017; Glover, Taylor, and Wu 2017). Nonetheless, academic research shows that the use of specialists is not always associated with higher-quality audits (Boritz, Kochetova-Kozloski, and Robinson 2015; Zimmerman, Barr-Pulliam, Lee, and Minutti-Meza 2023). In sum, the PCAOB should consider the academic literature on the use of specialists during standard setting on QCs related to the use of other participants.

Academic research finds evidence that partner-related characteristics influence audit quality. The academic studies discussed in this section rely on data from other countries because they either use (1) partner identity, which was not available in the United States until recently, or (2) partner compensation data, which is only available in certain countries. The PCAOB should keep in mind how these countries differ from the United States (e.g., different capital market structures) when considering how these studies generalize to the U.S. audit market and PCAOB regulations.

Despite audit firms having unique audit methodologies and using QC systems, academic research suggests that audit partners have unique effects on clients’ financial reporting. Using data from Sweden, Knechel, Vanstraelen, and Zerni (2015) find evidence suggesting that audit partners exhibit persistent aggressive or conservative reporting decisions and that the market recognizes that certain partners are aggressive and penalizes their clients. Using data from China, Wang, Yu, and Zhao (2015) find that audit partners associated with materially misstated annual financial statements are more likely to be associated with future financial statements that are materially misstated and that audit firm QCs do not mitigate this association. Thus, these two studies suggest that certain audit partner characteristics persistently affect audit quality.

Academic research also suggests that partner compensation structure and wealth affect audit quality. Using Swedish data, Knechel, Niemi, and Zerni (2013) find evidence that Big 4 audit firms have unique compensation structures, and these structures result in variation in partner compensation within each audit firm. Dekeyser, Gaeremynck, Knechel, and Willekens (2021) use the setting of Belgian Big 4 audit firms and their clients, which are primarily private. They find that incentive compensation varies across partners within the same audit firm and that incentive compensation (based on audit fees) is negatively related to audit quality measures. However, partner net wealth is positively associated with proxies for audit quality.

Brenk, Majoor, and Wright (2021) conduct an experiment in the Netherlands with audit partners and managers, manipulating profit-sharing plan type, client importance, and reinforcement sensitivity (i.e., the sensitivity that a partner has to achieve rewards and avoid punishment). The authors find the highest (lowest) audit quality when the profit-sharing plan is based on firm (partner) performance and client importance is low (high). However, these results only exist for participants whose reinforcement sensitivity is high. Thus, individual differences between partners can affect how they respond to compensation schemes.

A long literature has examined whether providing nonaudit services to audit clients affects audit quality (e.g., because of independence impairment) and generally finds little evidence of negative audit quality effects. However, some recent papers find evidence that nonaudit services can negatively affect audit quality through mechanisms other than independence impairment. For example, Beardsley, Imdieke, and Omer (2021) find evidence that audit offices that provide greater levels of nonaudit services to audit clients are associated with lower audit quality, even after controlling for client-specific nonaudit services. The authors posit that nonaudit services can distract auditors even on clients that purchase minimal amounts of nonaudit services. Also, Donelson, Ege, Imdieke, and Maksymov (2020) find a negative (positive) correlation between nonaudit-related (audit-related) consulting firm acquisitions by audit firms and proxies for audit quality. Interview evidence suggests that these associations are due to shifts toward a culture of commercialism from nonaudit-related acquisitions and expertise transfers from audit-related acquisitions. Thus, the provision of nonaudit services could negatively affect audit quality through mechanisms other than independence impairment.

Research has examined both internal audit firm inspections and PCAOB inspections of audit firms. Regarding internal firm inspections, Aobdia and Petacchi (2022) find evidence that such inspections, while often predictable, lead to increased auditor effort and improved financial reporting in the inspection year. However, positive inspections lead to auditors decreasing effort in the future, while negative inspections lead to more sustained increases in auditor effort. Overall, internal inspection programs can improve auditor effort and financial reporting quality (see further discussion of internal inspections in Question 82 below).

The PCAOB’s audit firm inspection program also has been the focus of academic research. Aobdia (2018) finds that PCAOB inspections citing audit deficiencies lead to increases in auditor effort. These increases in effort go beyond the inspected engagements (i.e., auditors appear to increase effort on some noninspected engagements). Khurana, Lundstrom, and Raman (2021) provide evidence that initial PCAOB inspections improved audit quality more for Big 4 audit firms than for other firms. Thus, both studies point to some positive outcomes related to PCAOB inspections.

Another line of research explores auditors’ perceptions of PCAOB inspections. While auditors acknowledge that audit quality increases due to PCAOB inspections, they also point to some downsides. Johnson, Keune, and Winchel (2019) find evidence that auditors view the PCAOB as powerful and that inspections have affected the audit and firms’ QC. They also find that auditors view the PCAOB inspection environment as “antagonistic” and do not fully agree with the PCAOB’s focus. Westermann, Cohen, and Trompeter (2019) find that auditors acknowledge an increase in audit quality derived from PCAOB inspections, but they also highlight negative effects, including overdocumentation, too much focus on auditing standards and too little on accounting, and concerns about attracting and retaining people in the auditing profession. The authors state “the inspection process has created excessive stress and tension, beyond budget and fee pressures, which some auditors perceive as affecting the pool of talented auditors that firms may be able to attract and retain in the future” (Westermann et al. 2019, 694).

Finally, given large audit firms’ challenges in reducing PCAOB inspection audit deficiencies, Hendricks, Landsman, and Pena-Romera (2022) examine the relation between firms’ hiring of former PCAOB employees and future inspection results. They find that firms’ hiring of former PCAOB employees is positively related to the number of prior audit deficiencies and that future deficiencies decrease after the hiring of former PCAOB employees. However, the authors find no evidence that other measures of audit quality improve with the number of former PCAOB employees hired. Thus, the hiring of former PCAOB employees appears to assist firms with meeting the requirements of PCAOB inspectors (e.g., documentation and evidence gathering) but does not appear to affect actual audit quality (as reflected by the quality of clients’ financial statements).

Most research on nonaffiliated or foreign affiliates of globally networked firms’ QC systems’ design, implementation, and operation focuses on internal quality reviews (IQR). For example, Downey, Bedard, and Boland (2021) examine IQRs from GNF-affiliated firms. They investigate factors associated with the selection of IQR engagements and find support for the idea that GNFs use a risk-based approach to target internal review engagements. They report that, on average, 9.9 percent of non-U.S. component auditors have an IQR over two years. To compare, the PCAOB inspection rate for these same firms is 3.5 percent annually.

Research provides evidence that IQRs are important contributions to audit firm quality. Houston and Stefaniak (2013) use survey data to document audit partners’ perceptions of IQRs specifically and relative to PCAOB inspections. Partners find IQR reviewers to be qualified and focused on improving audit quality (rather than finding deficiencies), and they provided timely and meaningful feedback to improve audit quality. Downey and Westermann (2021) survey U.S.-based global group audit leads who report that negative consequences from deficient audits are not limited to individual partners. GNF member firms may be subject to training plans, greater involvement by the global firm, and more regular reviews for compliance with the global methodology.

Bik and Hooghiemstra (2018) examine factors that may undermine compliance with the global methodology for one GNF. The authors use IQR data from the network’s member firms. They find that cross-national cultural differences may undermine the uniform application of global audit methodologies aimed at consistent control measures. Flasher and Schenck (2019) assess the association of audit firm arrangements with audit quality. Using PCAOB inspection reports for audit firms headquartered outside of the United States, they test the relationship between auditing firm arrangements (specifically, “Big 4 network affiliates; BDO and Grant Thornton network affiliates; firms with other network affiliations; firms with associations and alliances; and unknown or no stated affiliation”) and audit quality. Their analysis suggests that when firms have “formal connections (i.e., networks and larger associations/alliances arrangements),” audits within these arrangements will have similar levels of quality The implication is that formal connections are a vehicle to share and enforce QC practices. However, they caution that networked firm affiliations may not be equally beneficial.

Hong, Kerr, and Wiggins (2022) corroborate this caution. Using PCAOB inspection reports, they find that unaffiliated firms performing referred work experience better inspection outcomes than those performing as principal auditor. They do not see similar differences between referred-work and principal-auditor engagements for affiliated firms, supporting the idea that GNFs perform a QC function across their membership.

While GNFs may share and enforce QC practices, Ege, Kim, and Wang (2020) suggest that it may be possible to detect QC issues within a global audit firm network by considering the financial reporting properties of the clients of the affiliates of the global audit firm network. The authors find evidence that global audit network affiliates that have poor QCs (as evidenced by PCAOB sanctions) affect their clients’ financial statements in a way that is different from the affiliates that do not have poor QCs.

A key ingredient of audit quality is the auditing profession’s ability to attract and retain talent. The talent issue is a very serious concern today for accounting firms and university accounting programs. Accounting enrollments are dropping, as is the number of candidates taking the CPA Exam each year (American Institute of Certified Public Accountants (AICPA) 2022b; Ellis 2022; Foley 2022). Many have pointed to the effects of low starting pay in public accounting, long working hours, the 150-hour rule, and even PCAOB regulation of the auditing profession in reducing the supply of future accountants (Hermanson, Houston, Stefaniak, and Wilkins 2016; D. Hermanson, H. Hermanson, and S. Hermanson 2020). Further, research finds evidence that audit quality may suffer when there is high outside demand for auditors during busy season (Ege, Kim, and Wang 2022). Auditors with more outside employment options may be distracted or less motivated to work hard, or they may actually leave the firm and disrupt the audit. Ege et al. (2022) also find that increasing auditor pay and drawing staff from other offices are ways to mitigate the negative impact on audit quality. In a related vein, Chen, Dong, Han, and Zhou (2020) use Chinese data and find evidence that greater audit partner workload compression is related to lower audit quality. Overall, the profession needs good people, and audit quality suffers when people leave, staffing is too low, or workloads are too high.

We raise this issue to highlight the balancing act that we believe the PCAOB should carefully consider in setting standards, conducting inspections, and regulating the auditing profession. On one hand, the PCAOB must be a vigilant and strong protector of audit quality, which fosters sound financial reporting. On the other hand, the PCAOB’s regulatory efforts can affect the attractiveness of public company auditing as a profession (Hermanson et al. 2016; Westermann et al. 2019). If too many people view the PCAOB’s oversight as a reason to leave public company auditing, then this will decrease audit quality, opposite of the PCAOB’s intent.

In this spirit, we encourage the PCAOB to closely monitor trends in the firms’ abilities to attract and retain talent and also to carefully monitor how the PCAOB’s regulatory efforts and even public messaging may be affecting the attractiveness of the profession, especially in periods when auditors have many other options for employment. Very tough regulation can create unintended consequences, one of which is the potential for lower audit quality if people leave.