A note on the effect of PCAOB inspections on the audit quality
of triennial CPA ﬁrms
Alan I. Blankley *, Keejae P. Hong, David Kerr,Casper E. Wiggins
Department of Accounting, Belk College of Business, University of North Carolina at Charlotte, Charlotte, NC 28223, USA
Available online 26 September 2014
This note reports the results of a study conducted regarding PCAOB inspections of trien-
nial CPA ﬁrms. The purpose was to see if there was any evidence that inspections contributed
to improved audit quality. It was found that small ﬁrms did not correct staﬃng deﬁcien-
cies, which were related to previous audit deﬁciencies determined by the PCAOB. However,
deﬁcient ﬁrms did increase their audit fees signiﬁcantly more following their ﬁrst inspec-
tions than non-deﬁcient ﬁrms. This result is consistent with applying greater audit effort
after the inspection.
Interestingly, this response does not persist through second inspections.
© 2014 Elsevier Ltd. All rights reserved.
Early studies examining the results of Public Company
Accounting Oversight Board (PCAOB) inspections of small
public accounting ﬁrms indicated that a major character-
istic of ﬁrms that received an audit deﬁciency judgment was
that they were over-extended.1By and large, these small
ﬁrms had too many issuer clients – too much work – for
the size of their staffs. After studying inspection results
through July 2006, Hermanson, Houston, and Rice (2007)
concluded that ﬁrms that had performed audits where the
PCAOB had identiﬁed deﬁciencies in their engagements
tended to be smaller, had more issuer clients, and were
growing more rapidly than non-deﬁcient ﬁrms.
If it is true that the size of the staff, the number of issuer
clients, and the likelihood of being found deﬁcient are
related, then it seems reasonable to suppose that correc-
tive action to reduce the number of issuer clients, or increase
the number of staff, or both, would help improve the quality
of these ﬁrms’ audits by relieving a constraint on the number
of audit hours available for issuer client engagements. Such
action would be consistent with the notion that the inspec-
tion was instrumental in helping to improve the audit quality
of triennial ﬁrms. Of course, relative staff imbalances could
also be improved through redirecting effort from non-
issuer clients to issuer clients, or by increasing the total hours
worked on issuer engagements as well. This redirected effort
would likely be reﬂected in increased fees for issuer clients;
it would also be consistent with an improvement in audit
quality following the inspection.
To determine whether PCAOB inspections helped to
improve audit quality of small accounting ﬁrms, the in-
spection reports for 947 triennial ﬁrm-years between January
2004 and June 2012 were examined. The study ﬁrst inves-
tigated whether the structural problems identiﬁed by
Hermanson et al. (2007) continued to be an issue after 2006.
The study then examined whether deﬁcient ﬁrms im-
proved their staﬃng relative to the number of issuer clients
after their inspections, and ﬁnally examined audit fee
Note: The authors express their appreciation for the assistance of Dr.
Gregory Jonas, Associate Professor, Case Western Reserve University, and
assistant editor of Research in Accounting Regulation, for his assistance.
* Corresponding author. Tel.: +704 687 7707; fax: 704-687-6938.
E-mail address: firstname.lastname@example.org (A.I. Blankley).
1For convenience, ﬁrms that received an audit engagement deﬁciency
judgment from the PCAOB during an inspection will be referred to as “de-
ﬁcient” ﬁrms, and those for whom the PCAOB did not identify any audit
engagement deﬁciency as “non-deﬁcient” ﬁrms. This use of the term should
not be interpreted to mean the ﬁrm is deﬁcient (or not) in any other sense.
1052-0457/© 2014 Elsevier Ltd. All rights reserved.
Research in Accounting Regulation 26 (2014) 212–216
Contents lists available at ScienceDirect
Research in Accounting Regulation
journal homepage: www.elsevier.com/locate/racreg
changes for both deﬁcient and non-deﬁcient ﬁrms after their
ﬁrst two inspections.
The results indicated that ﬁrms cited for having an audit
engagement deﬁciency during their ﬁrst PCAOB inspec-
tion had signiﬁcantly fewer total staff relative to issuer clients
than non-deﬁcient ﬁrms. Moreover, deﬁcient ﬁrms did not
improve their relative staﬃng issues following the inspec-
tion. What they did appear to do was to redirect effort. It
was found that deﬁcient ﬁrms increased audit fees follow-
ing their ﬁrst inspection visits signiﬁcantly more than non-
deﬁcient ﬁrms. Interestingly, the same increase was not
found in audit fees for deﬁcient ﬁrms following the second
inspection visit. It can be concluded that these results are
consistent with PCAOB inspections helping to improve the
audit quality of small accounting ﬁrms, but that the great-
est marginal improvement in audit quality occurred during
the initial inspection, as the audit review paradigm was
changing. By the time of the second inspection, many –
perhaps most – ﬁrms had already changed their processes
to the point that any additional changes resulting from sub-
sequent inspections were not substantial enough to be
observed in empirical tests of audit fees.
The rest of this note is organized as follows: ﬁrstly, a dis-
cussion of the rationale for the study; secondly, more
information on the empirical tests and their results, and
ﬁnally a conclusion discussing how these results can be of
use to regulators or practitioners.
Rationale for the study
One of the principal objectives of the Sarbanes–Oxley Act
of 2002 was to improve audit quality. Indeed, when the law
established the PCAOB, it charged the new agency with the
task of improving the quality of audits. Since its inspec-
tion process is arguably the Board’s primary means of
achieving that objective, it is important to evaluate whether
PCAOB inspections are, in fact, helping to improve audit
quality. This is especially true for small ﬁrms. Previous re-
search into audit quality has shown, across various measures
of audit quality, that small ﬁrms are associated with lower
quality audits compared to large ﬁrms (see, for example,
DeAngelo, 1981; Francis, 2004; Francis & Krishnan, 1999;
Jackson & Liu, 2010; Palmrose, 1988; Simunic, 1980). Re-
cently, Gunny and Zhang (2013) examined PCAOB inspection
reports and found that for smaller ﬁrms, PCAOB inspec-
tions were effective in identifying lower-quality audits.2Of
course, identifying audit engagement deﬁciencies and pro-
viding feedback during and after the inspection would be
likely to help promote audit quality improvements, as the
PCAOB suggests (PCAOB, 2012). Intuitively, this claim is
appealing. But is there any empirical evidence that the
audit quality of small ﬁrms actually improves after the
Audit quality is a diﬃcult concept to measure. In this
study, two measures were used, both related to audit effort.
Hermanson et al.’s (2007) paper indicates that deﬁcient ﬁrms
were over-extended, so measures of relative staﬃng per
issuer client were evaluated to see whether ﬁrms brought
their staﬃng more in line with the needs of their issuer client
base after an inspection. The logic here is that if the ﬁrm
did not have adequate staﬃng to provide the audit effort
required to perform a quality audit for its issuer client base,
then increasing the available audit effort is a necessary, but
not a suﬃcient, condition to improve audit quality. Thus,
improved staﬃng metrics following an inspection would be
consistent with a tangible effort to improve audit quality.
The second measure of audit quality used is audit fees.
Previous research supports the use of audit fees as a proxy
for audit quality. O’Keefe, Simunic, and Stein (1994) initial-
ly developed a theoretical model linking audit quality (the
level of assurance) with audit effort (the levels of labor
hours). Knechel, Rouse, and Schelleman’s (2009) audit pro-
duction model assumes that the more extensive the
evidence-gathering activities are, the higher the audit quality
will be, suggesting that as audit effort increases, the quality
of the audit increases as well. Effort and quality have also
been linked empirically. Bedard and Johnstone (2004) show
that auditors plan more hours for clients with higher per-
ceived risk of earnings management. Similarly, Lee and Son
(2009) ﬁnd that auditors who lengthened their audit work
permitted less earnings management, and Caramanis and
Lennox (2008) report a negative association between audit
effort and abnormal accruals. Prior research also links audit
fees and audit effort. Whisenant, Sankaragurusuvamy, and
Raghunandan (2003), Blankley, Hurtt, and MacGregor (2012),
Lobo and Zhao (2013), and Asthana and Boone (2012) all
indicate that below-normal audit fees may be associated
with shorter, lower quality audits. Thus, if deﬁcient ﬁrms
increase fees to a greater extent than non-deﬁcient ﬁrms
following an inspection, then that would be evidence sug-
gesting the ﬁrm is increasing audit effort, and would be
consistent with an attempt to increase audit quality.
The study had three empirical tests. First, the audit ﬁrm
staﬃng characteristics were evaluated to determine whether
staﬃng was, in fact, associated with audit engagement de-
ﬁciencies over the test period (2004–2012). Second, it was
determined whether the deﬁcient ﬁrms increased their rel-
ative staﬃng in an attempt to alleviate the constraints that
existed. Finally, it was determined whether deﬁcient ﬁrms
increased audit fees following inspections more than non-
deﬁcient ﬁrms. A discussion on the sources of the study’s
data can be found below.
The data includes information gathered from 1399 ﬁrst
and second PCAOB inspection reports on inspections con-
ducted between January 2004 and June 2012 for triennial
CPA ﬁrms. All inspection reports are publicly available from
the PCAOB website. Inspection reports for ﬁrms that do not
appear in Audit Analytics and those pertaining to third in-
spection visits were excluded. The resulting data set included
947 inspection reports of domestic small ﬁrms, of which 578
2It is interesting to note that Gunny and Zhang (2013) do not ﬁnd the
same effectiveness for annually inspected ﬁrms. For that group, their results
were conﬂicting and they conclude that the PCAOB inspections did not dis-
tinguish audit quality during their test period.
213A.I. Blankley et al./Research in Accounting Regulation 26 (2014) 212–216
represented ﬁrst inspection visits and 369 represented
second inspection visits.
Client data used in the analysis of audit fees were col-
lected from Audit Analytics and matched with the CPA ﬁrm
data from the inspection reports. Client-year observations
used in the audit fee analysis ranged from 3536 for ﬁrst in-
spections to 2517 for second inspections.
The ﬁrst test considered which ﬁrm characteristics were
associated with audit engagement deﬁciencies. Table 1 pro-
vides the mean and median staﬃng and client data taken
from the inspection reports.
From Table 1, observe that deﬁcient ﬁrms have, on
average, 42 professional staff members and 13 issuer clients,
while non-deﬁcient ﬁrms have 67 staff members and only
8 issuer clients. When the relative relationship between total
staff (including partners) and clients is considered, deﬁ-
cient ﬁrms have, on average, only 8 total professionals (staff
plus partners) per issuer client while non-deﬁcient ﬁrms
have closer to 17 total professionals per issuer client. These
differences are statistically signiﬁcant and indicate that over
the 2004–2012 period, deﬁcient ﬁrms were understaffed rel-
ative to non-deﬁcient ﬁrms, and this understaﬃng is
signiﬁcantly associated with the audit engagement deﬁ-
ciency. Additionally, a logistic regression model was run,
which included additional variables to control for year and
other characteristics of the inspection. Results (untabulated)
conﬁrmed that these relative staﬃng differences are sig-
niﬁcantly related to being found deﬁcient in the ﬁrm’s initial
Changes in relative staﬃng
Considering that deﬁcient ﬁrms have fewer staff per client
in the ﬁrm’s initial inspection, then it seems reasonable to
suppose that the relatively poorer staﬃng constrains the
quantity of audit hours or effort available to each issuer
client. One way this constraint could be alleviated would
be to increase the relative staﬃng directly by hiring addi-
tional staff or reducing the number of issuer clients. Changes
the ﬁrm might make to alleviate the constraint on audit
effort would be consistent with a tangible effort to improve
audit quality. However, when changes in relative staﬃng
after the ﬁrst inspection were examined, it was found there
were no statistical differences between the increase in staff-
ing per client for deﬁcient ﬁrms and non-deﬁcient ﬁrms. In
fact, non-deﬁcient ﬁrms appeared to increase staﬃng per
client slightly more than deﬁcient ﬁrms did in the period
between the ﬁrst PCAOB inspection and the second inspec-
tion. An OLS regression model was also developed and tested
where the change in partners plus staff per issuer client was
regressed against a dummy variable for an audit engage-
ment deﬁciency from the ﬁrst inspection along with size
and year controls. Results from that test conﬁrmed that
having a deﬁciency was not related to changes in relative
staﬃng. In other words, no evidence could be found that
deﬁcient ﬁrms attempted to increase audit effort through
changes in relative staﬃng as a result of the PCAOB
Changes in audit fees
The ﬁnal test examined changes in audit fees following
PCAOB inspections. Firms could redirect audit effort toward
issuer clients by working more hours, resigning from en-
gagements with non-public clients, reducing time spent on
non-public clients, or some combination of the three. In this
case, audit fees are expected to increase as the ﬁrm redi-
rected effort toward its issuer clients. To examine what the
impact of the PCAOB inspection was on subsequent audit
fees, a standard audit fee OLS regression model was used,
where the change in audit fees was measured in two ways,
from the year before the inspection to the year of the in-
spection, and alternatively, from the year before the
inspection to the year after the inspection. This depen-
dent variable was then regressed on two variables of interest
and a number of control variables. The two variables of in-
terest were audit deﬁciency variables for the ﬁrst and second
PCAOB inspections. Each variable was codeda1iftheﬁrm
was found to have at least one audit engagement deﬁcien-
cy and zero otherwise during the inspection. Other variables
in the model controlled for client size, auditor tenure, ﬁscal
year end, the complexity of client operations, various audit
risks, and year ﬁxed-effects.
Results of the test indicated that ﬁrms cited for audit en-
gagement deﬁciencies during their ﬁrst inspection increased
their fees to a signiﬁcantly greater extent than non-deﬁcient
ﬁrms. This was true whether the change in fees was cal-
culated from the year before the inspection to the inspection
year or the year after, although the result was stronger when
the change in fees was measured from the year before to
the year after the inspection. This result is consistent with
deﬁcient ﬁrms expending additional effort in order to
remediate the noted audit deﬁciencies, which presumably
would improve the quality of subsequent audits.
To the extent that the additional effort reﬂects more thor-
ough audits, rather than simply busy work undertaken
Descriptive statistics for initial PCAOB inspections: CPA ﬁrm characteristics by audit deﬁciency.
Firms with audit deﬁciencies
Firms with no audit deﬁciencies
Is there a statistically
CPA ﬁrm characteristics Mean (Median) Mean (Median)
Number of staff 42.23 (11.0) 67.18 (23.5) Yes
Number of issuer clients 13.09 (5.0) 8.14 (3.0) Yes
Staff per issuer client 6.31 (2.0) 13.51 (6.3) Yes
Partners and staff per issuer client 8.19 (2.7) 16.67 (8.7) Yes
214 A.I. Blankley et al./Research in Accounting Regulation 26 (2014) 212–216
merely to satisfy the regulator, then this result is consis-
tent with an increase in audit quality.
Interestingly, this result did not hold when the change
in audit fees following the second inspection visit was
examined.3The coeﬃcient on the variable of interest (the
dummy variable representing audit deﬁciency during the
second inspection) was not statistically signiﬁcant, sug-
gesting that CPA ﬁrms receiving an audit deﬁciency report
at the conclusion of their 2nd PCAOB inspection did not sub-
sequently change their audit fees (i.e., audit effort) to a
greater extent than ﬁrms whose 2nd inspection report did
not reveal an audit deﬁciency. The result was the same re-
gardless of whether the change was calculated from the year
before to the inspection year, or to the year after. This result
may be because subsequent increases in effort were not cap-
tured in fee increases due to pricing constraints in 2008 and
beyond, or it might indicate that audit deﬁciencies in the
second round of inspections were not as easily remedied
by additional audit effort, or that audit effort increased, but
that the increases were not robust enough to reach a thresh-
old of statistical signiﬁcance.
The results of this study are best described as “mixed.”
There is clear evidence that staﬃng/client imbalances are
associated with lower quality audits, as determined by the
PCAOB. It was also found that most CPA ﬁrms appear to
respond to deﬁciencies identiﬁed by PCAOB inspections by
increasing the extent of work performed by the ﬁrm’s
current professional employees but not by hiring addition-
al employees. The cost of this increase in workload is passed
on to clients in the form of higher audit fees. These actions
appear to be suﬃcient in remediating audit deﬁciencies, as
the number of CPA ﬁrms with 2nd-round deﬁciencies is
much less than 1st-round deﬁciencies. Thus, the PCAOB’s
stated objective of improving the quality of audits and the
reliability of ﬁnancial information appears to be succeeding.4
Firms with deﬁciencies identiﬁed in their 2nd round of in-
spections appear to be able to resolve those deﬁciencies
more eﬃciently than deﬁciencies identiﬁed in the 1st round,
as no association was found between 2nd-round deﬁcien-
cies and a subsequent increase in audit fees.
One limitation of the study is that staﬃng data for the
ﬁrms is only available in total whereas fees and number of
clients are only available for listed clients. The results could
be biased to the extent that these ﬁrms have a large portion
of their business with non-listed clients or to the extent
growth rates are substantially different between the non-
listed client business versus listed client business. However,
the study’s creators believe that any such bias would not
affect the ﬁndings. Furthermore, the ﬁndings regarding the
association between audit fees and audit engagement de-
ﬁciencies would not be affected.
Practical applications of the research ﬁndings
Both practitioners and inspection teams need to be sen-
sitive to the staﬃng suﬃciency issue reported by this study.
For practitioners, partners need to recognize that the clear-
est indicator of a substandard audit from the PCAOB’s
previous inspection is the staff-to-client suﬃciency level.
It may be that the ﬁrm simply needs to hire additional staff
to improve the quality of the ﬁrm’s audits or to prepare for
an inspection visit. The ﬁrm also might redirect staff from
non-issuer clients to issuer clients. If neither alternative is
feasible, then the ﬁrm may need to think about whether they
have the resources necessary to conduct public ﬁrm audits
or to take on additional clients, and should consider whether
or not to reduce the number of its issuer clients.
Given the PCAOB’s ”risk-based” approach to conduct-
ing inspections, inspection teams should take into
consideration staff-to-client ratios as well when determin-
ing the appropriate scope of work to be performed in
conducting inspections. In addition, teams should consid-
er how those ratios have changed or improved since the
previous inspection. If the partner(s) have shown no incli-
nation to address the issue, and the quality of the audits is
consistently deﬁcient, then the team may need to discuss
whether or not the ﬁrm has the manpower to conduct issuer
client audits. However, as the study indicates, a substan-
tial increase in fees subsequent to the previous inspection
may indicate that the ﬁrm has redirected its audit effort to
improving the quality of those audits, so it is recommend
that inspection teams review audit fee behavior after the
ﬁrm’s previous inspection and compare it with fee behav-
ior from similar ﬁrms following their inspections.
Finally, the results suggest some caution may be in order
for both regulators and practitioners, but particularly for
regulators. The fact that far fewer ﬁrms were considered de-
ﬁcient in their second inspections than in their ﬁrst
inspections suggests that audit quality improved after the
initial inspections. However, the fact that an increase in fees
following the second inspections is not observed suggests
that marginal improvements in audit quality may be more
diﬃcult (and costly) to achieve as audit ﬁrms have already
made improvements large enough to be observable in audit
fees. It may be that the low-hanging fruit for improving audit
quality has already been realized, and subsequent regula-
tions may increase costs without corresponding beneﬁts.
For this reason, regulators need to be careful to weigh the
beneﬁt of new regulations carefully against the cost of
Asthana, S., & Boone, J. (2012). Abnormal audit fee and audit quality.
Auditing: A Journal of Practice and Theory,31(3), 1–22.
3It should be noted here that there were far fewer deﬁciency judg-
ments in the second (and third) round of inspections than in the ﬁrst round.
4However, this conclusion is subject to the caveat that audit fees are a
good proxy for audit effort, which leads to higher quality audits. There is
support in the auditing literature for both assumptions, but the possibil-
ity remains that fees increase merely because the PCAOB demands
additional work related to satisfying process requirements, which may, or
may not, be related to audit quality. For example, the demand to supply
additional documentation over an audit issue could require more effort
from the auditor to supply, and hence, lead to higher fees, but may not
yield an improvement in audit quality because the documentation re-
quirement is unrelated to evidence gathering, the quality of the evidence,
or fostering improvements in audit judgment or decision-making.
215A.I. Blankley et al./Research in Accounting Regulation 26 (2014) 212–216
Bedard, J., & Johnstone, K. (2004). Earnings manipulation risk, corporate
governance risk, and auditors’ planning and pricing decisions. The
Accounting Review,79(2), 277–304.
Blankley, A. I., Hurtt, D., & MacGregor, J. E. (2012). Abnormal audit fees
and restatements. Auditing: A Journal of Practice and Theory,31(1),
Caramanis, C., & Lennox, C. (2008). Audit effort and earnings management.
Journal of Accounting and Economics,45(1), 116–138.
DeAngelo, L. (1981). Auditor size and audit quality. Journal of Accounting
and Economics,3, 183–199.
Francis, J. (2004). What do we know about audit quality? British Accounting
Francis, J., & Krishnan, J. (1999). Accounting accruals and auditor reporting
conservatism. Contemporary Accounting Research,16 , 135–165.
Gunny, K. G., & Zhang, T. C. (2013). PCAOB inspection reports and audit
quality. Journal of Accounting and Public Policy,32(2), 136–160.
Hermanson, D. R., Houston, R. W., & Rice, J. C. (2007). PCAOB inspections
of smaller CPA ﬁrms: Initial evidence from inspection reports.
Accounting Horizons,21(2), 137–152.
Jackson, S., & Liu, K. (2010). The allowance for uncollectible accounts,
conservatism, and earnings management. Journal of Accounting Research,
Knechel, W. R., Rouse, P., & Schelleman, C. (2009). A modiﬁed audit
production framework: Evaluating the relative eﬃciency of audit
engagements. The Accounting Review,84(5), 1607–1638.
Lee, H. Y., & Son, M. (2009). Earnings announcement timing and earnings
management. Applied Financial Economics,19(4), 319–326.
Lobo, G., & Zhao, Y. (2013). Relation between audit effort and ﬁnancial
report misstatements: Evidence from quarterly and annual
restatements. The Accounting Review, forthcoming.
O’Keefe, T., Simunic, D., & Stein, M. (1994). The production of audit services:
Evidence from a major public accounting ﬁrm. Journal of Accounting
Palmrose, Z. (1988). An analysis of auditor litigation and audit service
quality. The Accounting Review,63, 55–73.
PCAOB. (2012). Information for audit committees about the PCAOB inspection
process. August 1: 1–26. Available at: <http://pcaobus.org/Inspections/
Simunic, D. (1980). The pricing of audit services: Theory and evidence.
Journal of Accounting Research,18, 161–190.
Whisenant, S., Sankaragurusuvamy, S., & Raghunandan, K. (2003). Evidence
on the joint determination of audit and nonaudit fees. Journal of
Accounting Research,41(4), 721–744.
216 A.I. Blankley et al./Research in Accounting Regulation 26 (2014) 212–216