Auditors who engage in unethical behavior often face subsequent reputational damage, such as loss of market share to competing auditors. KPMG was recently embroiled in a high-profile scandal that involved prominent members of the firm inappropriately receiving confidential information about upcoming PCAOB inspections. This article summarizes a study by Hale and Truelson (2022) that analyzes the effect of the scandal on KPMG’s reputation within the audit marketplace. Specifically, the study examines KPMG’s ability to attract new clients, the types of clients acquired, and audit fees charged to new clients. We then conclude by discussing the implications of the scandal on audit practitioners, regulators, and academics.

Data Availability: Data used in this study are available from public sources identified in the document.

A vital aspect of an external auditor’s assurance process is to uphold a high standard of ethical conduct (Satava, Caldwell, and Richards 2006). When this obligation is not met, the auditor’s reputation may be damaged such that it affects the firm’s market share, audit fees charged to clients, and client acquisition practices (DeAngelo 1981; DeFond and Zhang 2014). This paper summarizes Hale and Truelson's (2022) study on a recent auditor scandal and outlines the implications for audit professionals. In 2017, the public learned that high-ranking members of KPMG had taken the unprecedented action of improperly obtaining confidential information about upcoming inspections from PCAOB staff members (Michaels and Rapoport 2017). The firm then used this proprietary information to “clean up” audit documentation for the particular engagements they anticipated would be inspected in hopes of minimizing negative findings. This provided an undue advantage for KPMG, as firms are generally not aware of which engagements will be selected for inspection until after the audits are completed. The scheme was discovered internally by KPMG when a partner notified the firm that she had been improperly warned of upcoming inspections, which KPMG subsequently self-reported to the PCAOB. KPMG’s attempt to skirt regulatory oversight led to one of the largest fines ever levied upon an accounting firm—$50 million—and jail time for multiple members involved in the scheme, which was dubbed the “steal the exam” scandal.1 This event also provides a unique testing ground for inspecting the effects of a damaged auditor reputation since the firm’s clients were unaware and uninvolved in the unethical behavior, unlike previous high-profile audit scandals. Thus, the actions are a reflection of auditor behavior that unequivocally violates ethical professional guidelines. Also different from certain prior scandals, KPMG quickly admitted guilt in the matter and took swift action by dismissing the implicated partners.

Hale and Truelson (2022) examine the effect of the “steal the exam” scandal on KPMG’s ability to attract new audit clients in the post-scandal period.2 Interestingly, the authors find that reputational damage observed through decreased client acquisitions is dependent upon the comparison group used in the analysis. Specifically, KPMG appeared to struggle to compete for new clients against other Big 4 auditors, but fared well relative to large non-Big 4 auditors following the scandal. Additionally, Hale and Truelson (2022) find that the scandal appeared to impact the types of clients that KPMG acquired in the post-scandal period, which is also associated with higher fees charged to these new clients.

Auditors’ reputations are important; various stakeholders rely on auditors to perform ethically in providing credibility to audited financial statements (Satava et al. 2006). Thus, auditors who engage in questionable behavior often face negative consequences, such as a loss of clients, a reduction in audit or nonaudit fee revenue, and capital market losses for clients. The most prominent example of an auditor’s damaged reputation is likely the national and office-level reputational effects experienced by Arthur Andersen as the Enron audit failure unfolded (Chaney and Philipich 2002; Barton 2005; Krishnamurthy, J. Zhou, and N. Zhou 2006), which contributed to the passage of the Sarbanes-Oxley Act of 2002 (SOX).3 While auditors may engage in ethically questionable behavior that can impact audit engagements and associated audit quality, auditors in the post-SOX environment have also increasingly been required to uphold ethical standards outside of engagement performance. For instance, PCAOB inspections have placed pressure on auditors to pass with minimal findings in order to avoid the perception of substandard work, to maintain or achieve desired positions within the firm (e.g., promotions), and to avoid scrutiny from superiors (Nagy 2014; Aobdia and Shroff 2017; Johnson, Keune, and Winchel 2019; Westermann, Cohen, and Trompeter 2019).

Thus, somewhat ironically, the post-SOX environment has created new opportunities for auditors’ reputational damage. For instance, when Deloitte was sanctioned by the PCAOB for quality control issues in 2007, Deloitte faced a higher risk of client defection despite a lack of direct impact on specific engagements (Boone, Khurana, and Raman 2015). While similar, one may argue that KPMG’s “steal the exam” scandal is more severe due to the admission of guilt by the firm and the collusion between national firm partners and certain PCAOB members. Hence, it is important to consider how the unique nature of this scheme impacted the firm. Hale and Truelson (2022) undertake this examination with a focus on new audit client acquisitions, as clients seeking a new auditor may be especially sensitive to a high-profile auditor scandal. Then, Hale and Truelson (2022) examine the types of clients KPMG acquired post-scandal and how a change in the types of new clients acquired impacted KPMG’s post-scandal audit fees.

Past scandals or audit failures involving audit firms have been associated with a loss of market share following the events (Barton 2005; Krishnamurthy et al. 2006; Weber, Willenborg, and Zhang 2008; Skinner and Srinivasan 2012; Swanquist and Whited 2015). For example, Weber et al. (2008) report that KPMG Germany suffered reputational damage through a loss of clients following a scandal where the auditor failed to detect a client’s fraudulent recognition of revenue over several years. Such an effect may be due to hesitancy from existing or potential clients in associating themselves with an audit firm in the midst of a scandal, which may have negative capital market repercussions.

Hale and Truelson (2022) analyze KPMG’s client acquisition rate relative to other large auditors in the three years preceding and the three years following the “steal the exam” scandal. Since news media first alerted the public to the scandal in April 2017, the pre-scandal period includes fiscal years 2014–2016, while the post-scandal period includes fiscal years 2017–2019.4 In their sample, Hale and Truelson (2022) find that KPMG gained 58 new clients in the pre-scandal period, but only 34 new clients in the post-scandal period, which could be indicative of reputational damage.5 However, in a regression model that controls for other factors related to auditor choice, such as client size, they fail to find evidence that KPMG’s overall acquisition rate decreased post-scandal.6

Yet, Hale and Truelson (2022) also present evidence that the effect of reputational damage through client acquisition may be better detected by examining different classes of audit firms separately. Specifically, they split their sample into a “Big 4” sample, which includes KPMG along with the other Big 4 audit firms, and a “Next Four” sample, which measures KPMG against the next four largest audit firms in the U.S.7 When splitting the sample, a different pattern of client acquisition emerges between the two groups, as shown in Figure 1. In detail, KPMG’s percentage share of new clients fell from 29 percent pre-scandal to 14 percent post-scandal in the Big 4 sample but remained more stable in the Next Four auditor group. In the Next Four sample, the percentage share of new clients went from 27 percent pre-scandal to 22 percent post-scandal. Moreover, when controlling for other factors in the full regression model, Hale and Truelson (2022) find evidence that KPMG was less likely to gain new clients in relation to the other Big 4 auditors following the scandal but was more likely to gain new clients relative to the large non-Big 4 auditors. Interestingly, KPMG’s competitive advantage against the Next Four firms arose despite the fact that KPMG had more deficient audits identified in PCAOB inspection reports post-scandal than it did pre-scandal, while the Next Four firms all had fewer deficient audits identified in the post-scandal period.8 Overall, the analysis suggests that KPMG did suffer reputational damage in the years following the “steal the exam” scandal, observable through the new client acquisition rate compared to Big 4 auditors, but that the damage was at least partially offset by clients gained from firms outside of the Big 4.

FIGURE 1

KPMG’s Client Acquisitions from Other Auditors Pre- and Post-Scandal

This figure shows the number of clients in Hale and Truelson's (2022) sample acquired by KPMG in the pre- and post-scandal periods from other audit firms. The other Big 4 firms are Deloitte, Ernst & Young (EY), and PricewaterhouseCoopers (PwC), while the Next Four is comprised of BDO, Crowe, Grant Thornton, and RSM.

FIGURE 1

KPMG’s Client Acquisitions from Other Auditors Pre- and Post-Scandal

This figure shows the number of clients in Hale and Truelson's (2022) sample acquired by KPMG in the pre- and post-scandal periods from other audit firms. The other Big 4 firms are Deloitte, Ernst & Young (EY), and PricewaterhouseCoopers (PwC), while the Next Four is comprised of BDO, Crowe, Grant Thornton, and RSM.

Close modal

Apart from changes in market share, an auditor’s reputational damage may be observable through other avenues. For instance, given the evidence above suggesting that KPMG had difficulty attracting new clients relative to Big 4 auditors, KPMG may be incentivized to change client acceptance procedures to mitigate market share damage. Specifically, KPMG may be more inclined to accept riskier clients in the wake of the scandal than they would otherwise in order to maintain market share.

To examine this potential shift in acceptance behavior, Hale and Truelson (2022) examine the characteristics of KPMG’s new clients in the post-scandal period. As shown in Figure 2, KPMG was nearly twice as likely to acquire a client with either a recent going concern opinion or a material weakness identified in internal controls in the post-scandal period (38 percent) than in the pre-scandal period (20 percent).9 The authors also find that following the scandal, KPMG was more likely to accept clients with higher levels of earnings management and bankruptcy risk in comparison to large non-Big 4 auditors.10 Overall, this evidence suggests KPMG engaged riskier clients following the “steal the exam” scandal, which is another subtle indicator of the consequences of a damaged reputation. Future research may wish to examine whether and how KPMG deliberately changed its marketing strategy to clients following the scandal.

FIGURE 2

Percentage of Reporting Irregularities for KPMG’s New Clients Pre- and Post-Scandal

This figure shows the percentage of clients acquired by KPMG in the pre- and post-scandal periods with either a going concern opinion (GC) or an identified material weakness in internal controls (MW) in the three years prior to acquisition. The percentage of clients with either a MW or GC in the post-scandal period (38 percent) is significantly higher than the percentage in the pre-scandal period (20 percent) (p-value < 0.05 from paired t-test).

FIGURE 2

Percentage of Reporting Irregularities for KPMG’s New Clients Pre- and Post-Scandal

This figure shows the percentage of clients acquired by KPMG in the pre- and post-scandal periods with either a going concern opinion (GC) or an identified material weakness in internal controls (MW) in the three years prior to acquisition. The percentage of clients with either a MW or GC in the post-scandal period (38 percent) is significantly higher than the percentage in the pre-scandal period (20 percent) (p-value < 0.05 from paired t-test).

Close modal

It is unclear what effect the “steal the exam” scandal would be expected to have on KPMG’s audit fees. On one hand, an audit firm facing negative reputational effects may respond by offering fee discounts to new clients, which may persuade clients’ management to engage with the auditor (Lys and Johnson 1990; Beattie and Fearnley 1995). On the other hand, if KPMG was engaging riskier clients following the scandal, as indicated in the prior section, KPMG may actually increase audit fees to compensate for the additional risk and associated effort for these new clients (Huang, Raghunandan, and Rama 2009; Bronson, Ghosh, and Hogan 2017). Hale and Truelson (2022) provide evidence supporting the latter.

As shown in Figure 3, KPMG’s natural log of audit fees for new clients significantly increased following the scandal, from 13.7 pre-scandal to 14.3 post-scandal.11 This indicates that the raw audit fees for new clients increased from just under $900,000 on average before the scandal to over $1.6 million after. Moreover, while audit fees increased for several large accounting firms in the post-scandal period, KPMG’s fees increased significantly more than the other firms in the Hale and Truelson (2022) sample.12 This evidence suggests KPMG increased audit fees following the scandal, perhaps due to the riskier types of clients acquired in this time period to maintain market share.

FIGURE 3

Audit Fees Charged to KPMG’s New Clients Pre- and Post-Scandal

This figure shows the natural log of audit fees charged to KPMG’s new clients the pre- and post-scandal periods. The value in post-scandal period (14.3) is significantly higher than the value in the pre-scandal period (13.7) (p-value < 0.05 from paired t-test).

FIGURE 3

Audit Fees Charged to KPMG’s New Clients Pre- and Post-Scandal

This figure shows the natural log of audit fees charged to KPMG’s new clients the pre- and post-scandal periods. The value in post-scandal period (14.3) is significantly higher than the value in the pre-scandal period (13.7) (p-value < 0.05 from paired t-test).

Close modal

Hale and Truelson (2022) provide evidence of a varied response from both clients and KPMG following the highly public “steal the exam” scandal. First, they observe an overall loss of clients in comparison to Big 4 peers, but an acquisition advantage in relation to large non-Big 4 auditors. This insight should be of interest to practicing auditors. While it may seem like a firm’s reputational damage stemming from negative PCAOB inspections is a more salient threat to the firm since auditors risk losing clients following PCAOB inspection report findings (Nagy 2014; Aobdia 2018), the results from Hale and Truelson (2022) suggest that an auditor bears risk when attempting to skirt compliance with these regulatory mechanisms. It is difficult to compare reputational damage from a blatant violation of ethical standards to that from negative inspection findings, especially because the SEC never called KPMG’s audit quality into question as a result of the scandal. However, it appears that clients respond to these ethical breaches to a certain extent and that KPMG’s admission of guilt and swift response (e.g., terminating implicated individuals) were not enough to completely soften the blow to the firm’s reputation.

We note that future researchers should consider more nuanced examinations of reputational damage, as the results from Hale and Truelson (2022) are indicative of a subtle shift in KPMG’s place within the audit market, such that the damage is concentrated when comparing KPMG against the other Big 4 firms. One potential conclusion from this analysis is that clients still value a Big 4 audit, which provides KPMG with an advantage against large non-Big 4 auditors, but that within the Big 4 class, other factors become more important to clients seeking a new auditor. Further, we note that KPMG may still hold a competitive advantage in areas in which it has expertise, such as in the financial services sector, where the firm has historically audited a high number of clients.13 While this analysis was not included in Hale and Truelson (2022), we note that new clients in the financial industries are more likely to choose KPMG in the post-scandal period, relative to the broad group of large auditors. Further, we note that the authors’ primary results in the Big 4 sample appear to be attenuated by financial industry firms in that KPMG’s expertise in the financial industry alleviates reputational damage within this group.14 This finding is important for auditors and regulators to consider, as clients may be willing to overlook unethical behavior in order to obtain a high-quality audit.

Perhaps in response to the shift in KPMG’s client acquisition rate, Hale and Truelson (2022) also observe that KPMG appeared to obtain risker new clients following the scandal than it had before. This finding should also be informative to auditors, as it shows a cascading effect of reputational damage resulting from the unethical behavior. While KPMG may be able to retain market share over time, due to the demand for Big 4 auditors in the U.S., a change in the types of clients acquired can have lasting effects on the firm. Specifically, engaging riskier clients should naturally increase KPMG’s inherent risk, which would increase overall audit risk without an alteration to the audit plan. In turn, an increased audit risk could expose KPMG to future reputational damage due to the higher likelihood of restatements and other financial reporting irregularities. KPMG has possibly responded to this increased risk by increasing fees charged to new clients, as Hale and Truelson (2022) show. These increased fees may protect the firm by lowering detection risk to an acceptable level, but they may also serve as a competitive disadvantage in attempting to attract new clients, which underscores the authors’ primary findings. Thus, the findings in totality suggest that auditor reputational effects extend into various aspects of auditor-client relations, and that these effects interplay with one another. We also note that the evidence of KPMG’s increased fees could be indicative of other factors; for instance, the higher fees could represent firm investment in quality control mechanisms aimed at fortifying the firm’s adherence to ethical and regulatory standards.

Overall, the “steal the exam” scandal serves as a case study on an event unique to public accounting. While the reputational damages observed by Hale and Truelson (2022), along with the civil and criminal charges associated with the event, would seemingly serve as a warning to other firms to uphold high ethical standards, another Big 4 firm has recently been sanctioned by the SEC for actions unrelated to specific client engagements. EY was fined $100 million after the SEC determined that the firm’s employees cheated on ethics exams associated with CPA licensing (Ganun 2022). The magnitude of the penalty—which exceeds the penalty levied on KPMG and is, to date, the largest penalty ever imposed on an audit firm by the SEC—indicates that regulators will continue to make a concentrated effort to discipline unethical behavior, even if it does not have an obvious direct effect on audit quality. Future researchers may wish to analyze whether this scandal will have reputational effects on EY similar to those on KPMG observed by Hale and Truelson (2022). Notably, EY is a significantly larger firm than KPMG, even though they are both part of the Big 4 class, which may provide EY a better ability to withstand reputational damages.15 Further, given the frequency of questionable behavior from accounting firms, despite SOX’s propensity to curb more egregious firm actions with a direct effect on financial reporting outcomes, one may wonder whether stakeholders may eventually become “numb” to these negative events.

Accounting Today
.
2021
.
The 2021 top 100 firms
. https://www.accountingtoday.com/the-2021-top-100-firms-data
Altman
,
E. I.
1968
.
Financial ratios discriminant analysis and the prediction of corporate bankruptcy
.
The Journal of Finance
23
(
4
):
589
609
.
Aobdia
,
D.
2018
.
The impact of PCAOB individual engagement inspection process–preliminary evidence
.
The Accounting Review
93
(
4
):
53
80
.
Aobdia
,
D.
, and
N.
Shroff
.
2017
.
Regulatory oversight and auditor market share
.
Journal of Accounting and Economics
63
(
2–3
):
262
287
.
Barton
,
J. A. N.
2005
.
Who cares about auditor reputation?
Contemporary Accounting Research
22
(
3
):
549
586
.
Beattie
,
V.
, and
S.
Fearnley
.
1995
.
The importance of audit firm characteristics and the drivers of auditor change in UK listed companies
.
Accounting and Business Research
25
(
100
):
227
239
.
Boone
,
J. P.
,
I. K.
Khurana
, and
K. K.
Raman
.
2015
.
Did the 2007 PCAOB disciplinary order against Deloitte impose actual costs on the firm or improve its audit quality?
The Accounting Review
90
(
2
):
405
441
.
Bronson
,
S. N.
,
A.
Ghosh
, and
C. E.
Hogan
.
2017
.
Audit fee differential, audit effort, and litigation risk: An examination of ADR firms
.
Contemporary Accounting Research
34
(
1
):
83
117
.
Cahan
,
S.
,
W.
Zhang
, and
D.
Veenman
.
2011
.
Did the waste management audit failures signal lower firm-wide audit quality at arthur andersen?
Contemporary Accounting Research
28
(
3
):
859
891
.
Center for Audit Quality (CAQ)
.
2012
.
Guide to PCAOB inspections
. https://www.thecaq.org/wp-content/uploads/2019/03/guidetopcaobinspections.pdf
Chaney
,
P. K.
, and
K. L.
Philipich
.
2002
.
Shredded reputation: The cost of audit failure
.
Journal of Accounting Research
40
(
4
):
1221
1245
.
DeAngelo
,
L. E.
1981
.
Auditor size and audit quality
.
Journal of Accounting and Economics
3
(
3
):
183
199
.
DeFond
,
M.
, and
J.
Zhang
.
2014
.
A review of archival auditing research
.
Journal of Accounting and Economics
58
(
2-3
):
275
326
.
Francis
,
J.
1984
.
The effect of audit firm size on audit prices: A study of the Australian market
.
Journal of Accounting and Economics
6
(
2
):
133
151
.
Ganun
,
J.
2022
.
Accounting giant Ernst & Young admits its employees cheated on ethics exams
.
NPR
(June 28). https://www.npr.org/2022/06/28/1108044858/accounting-giant-ernst-young-admits-its-employees-cheated-on-ethics-exams
Hale
,
K.
, and
J. M.
Truelson
.
2022
.
Client acquisition following an auditor’s unethical behavior: An examination of reputational consequences following KPMG’s “steal the exam” scandal
.
Accounting Horizons
(forthcoming).
Hogan
,
C.
, and
R.
Martin
.
2009
.
Risk shifts in the market for audits: An examination of changes in risk for “second tier” audit firms
.
Auditing: A Journal of Practice & Theory
28
(
2
):
93
118
.
Huang
,
H.
,
K.
Raghunandan
, and
D.
Rama
.
2009
.
Audit fees for initial audit engagements before and after SOX
.
Auditing: A Journal of Practice & Theory
28
(
1
):
171
190
.
Johnson
,
L. M.
,
M. B.
Keune
, and
J.
Winchel
.
2019
.
U.S. auditors’ perceptions of the PCAOB inspection process: A behavioral examination
.
Contemporary Accounting Research
36
(
3
):
1540
1574
.
Kothari
,
S. P.
,
A. J.
Leone
, and
C. E.
Wasley
.
2005
.
Performance matched discretionary accrual measures
.
Journal of Accounting and Economics
39
(
1
):
163
197
.
Krishnamurthy
,
S.
,
J.
Zhou
, and
N. A. N.
Zhou
.
2006
.
Auditor reputation, auditor independence, and the stock-market impact of Andersen’s indictment on its client firms
.
Contemporary Accounting Research
23
(
2
):
465
490
.
Lys
,
T.
, and
W. B.
Johnson
.
1990
.
The market for audit services: Evidence from voluntary auditor changes
.
Journal of Accounting and Economics
12
(
1
):
281
308
. https://doi.org/10.1016/0165-4101(90)90051-5
Michaels
,
D.
, and
M.
Rapoport
.
2017
.
KPMG fires partners over leak of audit regulator’s confidential plan
.
Wall Street Journal
(April 11). https://www.wsj.com/articles/u-s-audit-regulator-probing-leak-of-confidential-inspection-information-to-kpmg-1491922950
Nagy
,
A. L.
2014
.
PCAOB quality control inspection reports and auditor reputation
.
Auditing: A Journal of Practice & Theory
33
(
3
):
87
104
.
Satava
,
D.
,
C.
Caldwell
, and
L.
Richards
.
2006
.
Ethics and the auditing culture: Rethinking the foundation of accounting and auditing
.
Journal of Business Ethics
64
(
3
):
271
284
.
Skinner
,
D. J.
, and
S.
Srinivasan
.
2012
.
Audit quality and auditor reputation: Evidence from Japan
.
The Accounting Review
87
(
5
):
1737
1765
.
Swanquist
,
Q. T.
, and
R. L.
Whited
.
2015
.
Do clients avoid “contaminated” offices? The economic consequences of low-quality audits
.
The Accounting Review
90
(
6
):
2537
2570
.
Weber
,
J.
,
M.
Willenborg
, and
J.
Zhang
.
2008
.
Does auditor reputation matter? The case ofKPMG Germany and ComROAD AG
.
Journal of Accounting Research
46
(
4
):
941
972
.
Westermann
,
K. D.
,
J.
Cohen
, and
G.
Trompeter
.
2019
.
PCAOB inspections: Public accounting firms on “trial
.”
Contemporary Accounting Research
36
(
2
):
694
731
.
1

As discussed in the implications section, Ernst & Young (EY) was recently involved in a scandal related to ethical violations. The $100 million penalty levied on EY exceeded that imposed upon KPMG for the “steal the exam” scandal (Ganun 2022).

2

Notably, KPMG was not prohibited by the SEC from accepting new clients following the scandal.

3

As noted by Cahan, Zhang, and Veenman (2011), Andersen had multiple audit failures in the years leading up to the Enron scandal.

4

The three-year pre- and post-scandal time windows in Hale and Truelson (2022) align with prior research on auditor reputation (e.g., Boone et al. 2015) and are expected to be long enough to detect reputational damage following the scandal. The post-scandal period begins on June 1, 2017, which would be the earliest a client’s auditor decision would reasonably be affected by news of the scandal. Future researchers may wish to examine the longer-term effects of the scandal on KPMG’s reputational damage.

5

The authors’ sample is derived from observations in the Audit Analytics database with the necessary data regarding client size and audit fees. The sample is limited to clients with a Top 8 auditor and with at least one observation available in each of the pre- and post-scandal periods. The final full sample, after eliminating observations with missing data, is composed of 18,258 observations, with 664 auditor changes.

6
Hale and Truelson (2022) present the following multivariate model to perform the regression analysis:
SWITCHi,t=β0+β1KPMG_CYi,t+β2POSTt+β3KPMG_CYi,t×POSTt+βkCONTROLSk,i,t1+Client Firm Fixed Effectsi+εi,t.

In this model, the dependent variable, SWITCH, measures whether an audit client in the sample switched auditors in the year under examination or not. The variable of interest is the interaction term KPMG_CY × POST (denoted by bold), which measures the incremental likelihood of KPMG obtaining new clients in the post-scandal period in relation to other auditors. The model controls for several client characteristics pulled from prior research on auditor selection (e.g., Boone et al. 2015): size, growth, leverage, profitability, industry, and prior financial reporting or audit irregularities.

7

The Next Four is comprised of BDO, Crowe, Grant Thornton, and RSM, who are likely viable competitors to KPMG in the audit market (Hogan and Martin 2009).

8

PCAOB inspections are conducted annually for auditors with more than 100 public companies, including the eight firms in our sample. Once inspections are finalized, the PCAOB releases public reports for each firm with noted deficiencies in Part I of the report. The deficiencies may be due to inadequate performance of audit procedures and/or failure to identify potential reporting irregularities (Center for Audit Quality (CAQ) 2012).

9

The authors measure these financial reporting irregularities in the three-year window prior to the observation year.

10

Earnings management is measured via abnormal accruals, where higher levels of abnormal accruals are indicative of higher earnings management, consistent with Kothari, Leone, and Wasley (2005). Bankruptcy risk is measured via Altman Z-score developed in Altman (1968). These multivariate analyses control for client size.

11

The natural log transformation of audit fees is common in research on fees (e.g., Francis 1984). This transformation compresses nonlinear distributions, which are common with audit fee data, and decreases variance within the data. This allows for easier detection of association between variables of interest.

12

Specifically, a difference-in-differences estimation using the data in Hale and Truelson (2022) finds that KPMG’s post-scandal increase in audit fees was significantly greater than that of other firms (p-value < 0.10).

13

KPMG reported the largest portion of its revenue in the Financial Services industry in 2019 (KPMG 2019).

14

Specifically, a modified version of the regression model described in footnote 6 is specified to include the three-way interaction FINANCIAL × KPMG_CY × POST. FINANCIAL is an indicator variable equal to 1 if the two-digit SIC code is 60–69 and 0 otherwise. A positive coefficient on this variable in the full model (p-value < 0.05) suggests a competitive advantage for KPMG in this group.

15

EY earns over 65 percent more revenue firmwide than KPMG and employs over 20 percent more total professionals (Accounting Today 2021).