MIT Sloan School
of
Management
WORKING PAPER
Working Paper 4330-02
January 2002
THE RELATION BETWEEN AUDITORS' FEES FOR
NON- AUDIT SERVICES AND EARNINGS QU.\LITY
MIT
Sloan School of
Management
50 Memorial Drive
Cambridae, Massachusetts 02142-1 347
MIT
Sloan School of Management
Working Paper 4330-02
January 2002
THE RELATION BETWEEN AUDITORS' FEES FOR
NON-AUDIT SERVICES AND EARNINGS QUALITY
Richard M. Franlcel, Marilyn
5
2002 by Richard M. Frankel, Marilyn
F.
F.
Johnson, Karen K. Nelson
Johnson, Karen K. Nelson.
Short sections of text, not to exceed two paragraphs,
permission provided that
full credit
including
may
.AJl
rights reserved.
be quoted without explicit
© notice is given to the source.
MASSACHUStirSINSlllUlt
This paper also can be downloaded without charge firom the
Social Science Research
OFTECHNGLOGY
Network Electronic Paper Collection:
http://papers.ssm.com/abstract_id=296557
JUL 2 3
2002
LIBRARIES
The Relation Between Auditors' Fees for Non-Audit Services
and Earnings Quality
Richard M. Frankel
MIT
Sloan School ofBusiness
50 Memorial Drive, E52.325g
Cambridge, MA 02459-1261
(617)253-7084
frankel@mit. edu
Marilyn F. Johnson
Michigan State University
Eli Broad Graduate School of Management
N270 Business
College Complex
East Lansing. MI48824-1122
(517) 422-0152
johnl614@msu.edu
Karen K. Nelson
Stanford University
Graduate School of Business
Stanford CA 94305-5015
(650) 723-0106
knelson(a}gsb. Stanford, edu
January 2002
thank Bill Beaver, Walt Blacconiere, Larry Brown, S. P. Kothari, Tom Linsmeier, Roger Martin, Maureen
McNichols, Joe Weber, two anonymous reviewers, and seminar participants at the University of California at
Berkeley, Carnegie Mellon University, Georgia State University, Michigan State University, Northwestern
University, and the 2001 Stanford Summer Accounting Camp for helpful discussions and comments. Karen Nelson
acknowledges the support of the Financial Research Initiative at Stanford University Graduate School of Business.
We thank First Call for providing the analyst forecast data. Waqas Nazir, Nora Richardson, and Xiaoyan Zheng
We
provided valuable data collection assistance.
The Relation Between Auditors' Fees for Non-Audit Services
and Earnings Quality
Abstract:
quality.
We
examine the association between the provision of non-audit services and earnings
Because of concerns regarding the
credibility, the Securities
effect
of non-audit services on financial reporting
and Exchange Commission recently issued revised auditor
independence rules requiring firms to disclose in their annual proxy statement the amount of fees
paid to auditors for audit and non-audit services. Using data collected from proxy statements
filed
between February
more non-audit
and
is
5,
services
2001 and June
from
15,
2001,
their auditor are
we present
more
to report larger absolute discretionary accruals.
likely to just
filing date.
These
meet or beat
We also find that the unexpected
to total fee ratio is negatively associated
results are consistent with
strengthens an auditor's economic
analysts' forecasts
However, the purchase of non-audit services
not associated with meeting other earnings benchmarks.
component of the non-audit
evidence that firms purchasing
with stock returns on the
arguments that the provision of non-audit services
bond with the
client
and that investors price
this effect.
Keywords: Auditor independence, Auditor fees, Earnings management. Discretionary accruals
Data Availability: The data used in
this
study are from the public sources identified in the
text.
I.
INTRODUCTION
This paper provides empirical evidence on the relation between non-audit services and
earnings quality.
We test hypotheses concerning:
of non-audit services from
to the disclosure
its
of non-audit
auditor and
fees.
(1) the association
eammgs management, and
In the past decade there has
between a firm's purchase
(2) the stock price reaction
been a dramatic increase
proportion of fee revenue auditors derive from non-audit services, yet
we know
little
in the
about
how
non-audit services are related to earnings quality.' Concern about the effect of non-audit services
on the financial reporting process was a primary motivation for the Securities and Exchange
Commission (SEC)
to issue revised auditor
rules require firms to disclose the
amount of all
The provision of non-audit
reporting
by providing
independence rules on November
audit
and non-audit fees paid
services to audit clients
may increase
15,
to
2000. The
its
the quality of financial
the auditor with information about the client's business that
performing the audit (Antle
et al.
auditor.
is usefiil in
1997; Pitt and Birenbaum 1997). Non-audit services
increase the auditor's investment in reputational capital
(Aminada
that the provision of non-audit services strengthens the auditor's
1999).
may also
A competing view is
economic bond with the
client,
thereby increasing the auditor's incentive to acquiesce to client pressure (Simunic 1984; Parkash
and Venable 1993; Firth 1997).
The
the
SEC
first set
of tests focuses on one dimension of earnings quality of particular concern
in recent years (Levitt 1998),
i.e.,
management's deliberate intervention
process in the form of earnings management.
We
examine the propensity
to
to
in the earnings
meet or beat
earnings benchmarks and the level of a firm's absolute discretionary accruals to assess whether
' From 1993 to
1999, the average annual g^o^vth rate for revenues for management advisory and similar services was
26 percent, compared to 9 percent for audit services (SEC 2000).
the provision of non-audit services
non-audit services
may
is
The provision of
also affect investors' perceptions of audit quality. If investors require
compensation for perceived changes
(Simunic 1984).
associated with earnings management.
in audit quality,
We assess whether the disclosure
adjustments
in firm
value will occur
of non-audit fees conveys value-relevant
information to the market by examining the share price reaction to disclosure of non-audit fees.
The
the
analysis
SEC between
is
based on auditor fee data collected from 3,074 proxy statements
February
two measures of non-audit
5,
2001, and June
services.
The
first
15,
2001.
measure
(FEERATIO). The second measure disaggregates
audit
client
(RANKAUD) components,
both of which
We
is
filed
with
use the fee disclosures to construct
the ratio of non-audit fees to total fees
total fees into its
we measure
non-audit
(RANKNON)
and
as the percentile ranks for each
of a given auditor.
Results from the
first set
of tests indicate a significant positive association between the
purchase of non-audit fees and earnings management.
significantly higher for firms just
FEERATIO
and
RANKNON are
meeting analyst forecasts. In contrast, both measures are
we
negative and significant for firms just missing analyst forecasts. However,
association
between
increases in earnings
either
FEERATIO
or
find no
RANKNON and two other earnings benchmarks,
and small positive earnings. The
results also indicate that
FEERATIO
RANKNON are significantly higher for firms with high absolute discretionary accruals.
results are robust across several alternative specifications,
more non-audit services manage earnings
and indicate
to a greater extent
small
and
These
that firms purchasing
than other firms.
We interpret this
evidence as consistent with the argument that the provision of non-audit services strengthens an
auditor's
economic bond with the
client.
The earnings management analyses
also indicate that
RANKAUD
is
negatively
associated with meeting analyst forecasts and absolute discretionary accruals. Moreover, the
percentile rank of total fees
discretionary accruals.
to the auditor
not associated with either meeting analysts forecasts or absolute
is
Taken
together, this evidence suggests that the
have different implications
for audit quality.
The
components of fees paid
audit fee evidence
consistent
is
with the argument that audit fees increase an auditor's investment in reputational capital.
Results from the event study indicate a negative association between share prices and the
annoimcement of higher than expected non-audit
the evidence
is
sensitive to the
However, the
fees.
assumed expectations model
for non-audit fees.
marginally significant negative association between share prices and
RANKNON.
However, we find
statistically significant
between share prices and a proxy that
FEERATIO. These
of
We find a
FEERATIO
and
evidence of a negative association
explicitly controls for the expected
some
results provide
statistical significance
component of
consistent with the argument that investors perceive the
provision of non-audit services to strengthen an auditor's economic bond with the client.
The remainder of the paper
rules
organized as follows. Section
II
discusses the
SEC's new
on auditor independence and develops our hypotheses. Section EH describes the research
design. Section
IV presents the empirical
n.
SEC
is
results.
Section
V
summarizes and concludes.
BACKGROUND AND HYPOTHESIS DEVELOPMENT
rule on auditor independence
In
November 2000,
the
SEC
issued Final Rule S7- 13-00, Revision of the Commission
's
Auditor Independence Requirements. The Rule defines independence as "a mental state of
objectivity
and lack of bias" and adopts a two-pronged
independent. The
to as
first
prong
independence "in
fact."
is
test to
determine
if
an auditor
direct evidence of the auditor's mental state,
However, mental
is
commonly referred
states are not directly observable,
and thus the
second prong provides for an assessment of independence "in appearance" based on observation
of external
facts.
The Rule
specifies that the
SEC
will not consider an auditor to be independent
with respect to a particular client "if a reasonable investor, with knowledge of all relevant facts
and circumstances, would conclude that the auditor
judgment" (SEC 2000, Section
impartial
The Rule provides four guiding
These principles
state that
is
I).
principles for
independence
not capable of exercising objective and
is
impaired
making independence determinations.
when an
auditor: (1) has a
conflicting interest with the audit client; (2) audits the auditor's
management
The Rule
(3) functions as
or an employee of the audit client; or (4) acts as an advocate for the audit client.
also identifies situations in
certain financial or
under prior
own work;
mutual or
rules.
employment
which the auditor
is
not independent of the client because of
relationships, although these restrictions are
In addition, the
Rule
identifies
more hmited than
and prohibits certain non-audit services
that
it
believes impairs independence regardless of the size of fees they generate.
Finally, the
February
5,
Rule requires companies to disclose
in
proxy statements
filed
on or
after
2001, information regarding fees billed by the auditor for the most recent year.
These fee disclosures are intended
to
provide information useful to investors in evaluating
whether the magnitude of fees for non-audit services has impaired the auditor's independence.
Firms are required to disclose:
(1)
under the caption ''Audit Fees'" the aggregate fees billed for the audit of the annual
financial statements and for reviews of the quarterly financial statements;
(2)
under the caption ''Financial Information Systems Design and Implementation
Fees" the aggregate fees billed for professional services rendered for (i) directly
indirectly operating or supervising the
network, or
company's information system or
or
local area
designing or implementing hardware or software that aggregates
financial statement source data or data significant to the financial statements; and
(3)
(ii)
under the caption "All Other Fees" the aggregate fees billed for
rendered by the auditor other than those described in (1) and
all
(2).
services
Firms are also required
to disclose if the audit
committee (or the board of directors
if there is
no
audit committee) considered whether the provision of non-audit services is compatible with
maintaining the auditor's independence.
Hypothesis development
The provision of non-audit services and earnings management
Auditors
reliability
only
if the
facilitate contracting
between investors and management by
of the financial statements. However, the auditor's monitoring
auditor
is
independent (Watts and
Zimmerman
1986).
compromised when
and reports a misstatement, and independence
auditors
bond between auditor and
audit services.
He
client.
services can impair independence
is
the joint probability
considered
by creating an economic
Simunic (1984) models the joint demand
for audit
defines a decrease in auditor independence as "any situation
incentives such that a self-interested auditor
findings."'^
is
report misstatements they have discovered.
fail to
The provision of non-audit
valuable to investors
DeAngelo (1981) defmes
independence within a broader framework of audit quality. Audit quality
that the auditor discovers
is
attesting to the
He shows
that
when
the
is
more
and non-
which
alters
likely to ignore, conceal, or misrepresent his
same auditor provides both
services and the auditor retains a
portion of the cost savings from "knowledge spillovers," the auditor will be economically bonded
to the cHent.3
Parkash and Venable (1993) and Firth (1997) present empirical evidence using
U.S. and U.K. data, respectively, that firms act as if the purchase of non-audit services
2
Other definitions of independence are also discussed
in the literature.
Magee and Tseng
(1990, 322) state that a
lack of independence implies that "an auditor's decisions are not consistent with his or her behefs about a reporting
and Baiman et al. (1991) developed a conceptual framework for analysing auditor
independence as collusion between the auditor and the manager.
policy." Antle (1982, 1984)
^
Subsequent empirical work, however,
fails to
(Pahnrose 1986; Abdel-khalik 1990; Davis
fmd evidence
et al. 1993).
consistent with the existence of "knowledge spillovers"
jeopardizes independence. Both studies find that firms requiring high quality audits because of
high agency costs will be less likely to purchases non-audit services from their auditor. Gore
al.
et
(2001) present evidence also using U.K. data of a positive association between the provision
of non-audit services and earnings management. They also document that non-Big Five auditors
allow
more earnings management than Big Five
The agency
contrast,
literature
views auditor bias
an evolving body of experimental
auditors.
be a deliberate form of misrepresentation, hi
to
literature suggests that
psychological heuristics
unintentionally and unconsciously lead auditors to bias judgments. Beeler and
present experimental evidence that audit partners are
client is a
budgeted
going concern and
to the audit
when
to revise
downward
there are contingent
more
likely to initially
their initial estimate
Hunton (2001)
assume
that the
of the number of hours
economic rents arising from the provision of
non-audit services. The authors attribute these fmdings to an information processing bias
as predecisional distortion
of information. Bazerman
et al.
known
(1997) argue that the self-serving bias
poses a similar threat to independence and concludes that "the problem
lies
not in our desire to
be unfair, but in our inabihty to interpret information in an unbiased manner."
PoUcy makers
of the American
histitute
Of fundamental
of Certified Public Accountants
importance
in
two
(POB
2000, 119) states
that:
understanding the conflict of interest that arises from the
provision of non-audit services to audit clients
really serving
Board (POB)
offer similar arguments. For example, the Public Oversight
different sets of clients:
is
the fact that in so doing the audit firm
management
in the
is
case of management
consulting services, and the audit committee, the shareholders and
all
those
who
rely on
the audited fmancials and the firm's opinion in deciding whether to invest, in the case of
the audit.
.
.
It is
obvious that in serving these different chents the
And
it is
problem, regardless of whether,
The above arguments imply
quality
owed
fhm
is
subject to
one client or the other.
equally obvious that the existence of dual loyalties creates a serious appearance
conflicts of interest that tear at the firagile fabric of loyalty
to
in particular cases, the fabric actually tears apart or not.
that the provision of non-audit services results in
and an increased likelihood the auditor
will
lower audit
waive earnings management attempts.
An
auditor concerned about the possible loss of non-audit fee revenue
may be
less likely to disagree
with management's accounting choices, giving firms greater leeway to manage earnings.
A competing viewpoint suggests that the provision of non-audit services strengthens audit
quality
and can thus reduce earnings management. Academic research and policy arguments
provide support for this view. For example, a
1
997 White Paper submitted
Standards Board by Pitt and Birenbaum (1997, 4) states
.
.
.the
overwhelming economic
interest
to the
also
Independence
that:
of accounting firms
in their reputational capital
provides a powerful incentive to safeguard independence. Non-audit services increase the
firm's investment in reputational capital, contribute importantly to the quality of audit
services and provide other benefits to clients and the public.
Arrunada (1999) formally models the argument
that non-audit services increase audit
quaUty by providing information about the client firm's business that complements the audit
function and increasing an audit firm's investment in reputational capital. Using similar
arguments
to
DeAngelo
(1981), he
audit firms because the gains
the reputational losses that
shows
that reputational penalties constrain the behavior of
from acquiescing
to
any one
would be imposed by other
reputation for independence. Relatedly,
Dopuch
et al.
demands
client's
clients
who
are outweighed
value the audit firm for
by
its
(2001) analytically demonstrate that the
increase in reputational capital resulting from the provision of non-audit services will increase
auditor independence and audit quality provided that the probability of misstatement risk
is low."*
In addition to the reputational cost arguments, institutional factors support the contention
that non-audit fees
legal sanctions
'^
do not adversely effect auditor independence.
from poor quality
audits.
First, auditors
face significant
Aggregate legal claims against the Big Six auditing
Although reputational incentives may constrain the behavior of an individual employee seeking
to
maximize
whose
value of the audit firm, they are less effective in constraining the behavior of an individual employee
compensation depends on retaining a particular
client (Trompeter, 1994).
the
firms exceeded $30 billion at the end of 1992.^ Second, no evidence exists that the provision of
non-audit services poses a threat to auditor independence (Antle
et al.
1997;
Walhnan
1996).
Third, investors do not appear to be concerned about the provision of non-audit services (Antle et
.
al. 1
997). Finally, insurers perceive
non-audit services (Antle
no additional legal
liability arising fi-om the
provision of
et al. 1997).
Because of the competing views on the provision of non-audit services,
we
test the
following hypothesis, stated in null form:
HI: The provision of non-audit services
earnings management.
to audit cUents
is
not associated with
The provision of non-audit services and shareholder wealth
The provision of non-audit
services
may affect
For example, the provision of non-audit services
investors' perceptions of audit quality.
may reduce
the probability of truthful audit
reporting if the non-audit services generate economic rents. If investors require compensation for
a perceived increase in the probability
of nontruthfiil reporting, then a loss in firm value
occur (Simunic 1984).* In addition, a perceived reduction in audit quality
share price adjustments due to a
earnings stream (Teoh and
of non-audit services
downward revision
Wong
may improve
may trigger negative
in the perceived persistence
1993).^ Alternatively, the
will
of the firm's
knowledge gained by
the provision
the auditor's ability to detect accounting irregularities.
As
a
However, see Coffee (2001) for a discussion of how three recent legal developments - The Private Securities
Reform Act of 1995, the Supreme Court's decision in Central Bank of Denver N.A. v. First Interstate
Bank of Denver, N.A., and the Private Securities Litigation Uniform Standards Act - have resulted in a significant
^
Litigation
decrease in auditors'
^
For research
liability for securities fraud.
that relates uncertainty to valuation, see
Barry and BrowB (1985).
(1991) and Dechow et al. (1996) present evidence that companies
receiving accounting enforcement actions by the SEC for manipulating earnings experience a decline in stock prices
around the announcement of the enforcement action. The stock price response for our sample is not expected to be
^
Consistent with these arguments, Feroz et
al.
as extreme, however, because we do not focus on a sample of firms identified by the
Accepted Accounting Principles.
SEC
as violating Generally
result, investors
from
services
may
infer higher audit quality
an increase
their auditor, leading to
Given the two possible shareholder
we
services and perceived audit quality,
H2: There
is
services
is
from
one
their auditors,
from
in firm value.
between non-audit
following hypothesis, stated in null form:
test the
of non-audit
to the disclosure
fees.
reduced audit quality and thus penaHze firms purchasing non-audit
that the equilibrium carmot
services
firms disclose the purchase of non-audit
interpretations of the relation
no share price reaction
If investors infer
when
may
question whether this
be sustained,
i.e.,
However, even
their auditor.
if
(i.e.,
a viable equilibrium.
One answer
firms will not continue to purchase non-audit
such an equilibrium cannot be sustained, a stock
price response to the disclosure of non-audit fees can
equilibrium can be sustained
is
still
occur. Alternatively, such an
firms will continue to purchase non-audit services from their
auditor despite the negative share price effects) if choosing another firm to provide the services
would be
costlier or
produce fewer benefits. Non-audit services
may be more
costly and produce
fewer benefits when purchased from a firm that has less specific knowledge of the company
the auditor.
More
generally,
managers
may continue
to
engage low quality auditors even
market penalizes them for doing so because such a penalty
high quality auditor (Simunic 1984; Becker
et al.
is
tlian
if the
lower than the cost of engaging a
1998).
m. RESEARCH DESIGN
Sample description
Table
1
,
Panel
A summarizes
the sample selection procedure.
of 4,701 definitive proxy statements on the SEC's
February
5,
2001 and June
1
5,
2001
.
From
Our
initial
sample consists
EDGAR database with a filing date between
this total,
we
exclude 1,124 proxy statements
filed
fmancial institutions (SIC codes 6000-6999) because of the unique procedures required to
by
estimate discretionary accruals for these firms. In addition,
do not
relate to an annual
we
meeting of shareholders because the
disclose auditor fees in these proxy statements. Finally,
we
exclude 51 proxy statements that
SEC
does not require fums
exclude 71 proxy statements
indicating that the firm changed auditors during the year because the change
amount of fees
billed.
components of auditor
Of the remaining
fees.^
to
is
likely to affect the
3,455 observations, 218 do not disclose the required
We match
the firms with fee data to Compustat, resulting in a final
sample of 3,074 firms.
Panel
B
of Table
1
details the fi-equency
of proxy filings by month. Firms must
proxy statement prior to the annual shareholders meeting which typically occurs three
months
after the fiscal year end.
statement in
March
or April, corresponding to a
the time-period clustering, Panel
sample
is
The majority of the firms
C of Table
1
(median) rafio of audit fees
statistics
A
reports statistics for the
by the new auditor independence
ranges from a
rule.
minimum
The mean
of 0.02
to a
of 1.00. Only eight percent of sample firms engaged their auditor to provide fmancial
(IS).
Although the mean (median)
0.02 (0.00), untabulated fmdings indicate that this ratio
incurred these fees.
^
or January fiscal year end. Despite
of the fee data. Panel
to total fees is 0.51 (0.49), but
information systems design services
is
our sample filed their proxy
in
reveals that the industry composition of our
three categories of fees required to be disclosed
fees
to four
similar to that of the firms on Compustat in 2000.
Table 2 presents descriptive
maximum
December
file their
The noncompliance
The mean (median)
rate for our
17 percent reported by Abbott et
sample
al.
is
ratio
of other fees to
thus 6 percent
(i.e.,
218
-^
is
ratio
of IS fees
to total
0.28 (0.17) for firms that
total fees is 0.47 (0.48),
3,455), well
below
the
with 96
noncompliance
rate
of
(2001). However, their sample includes only proxy statements filed between
2001 and March 16, 2001. We find that the noncompliance rate for our sample dunng this time period is
noncompliance rate for proxy statements filed after March 16, 2001 is 5 percent. Abbott et al.
(2001) report that firms failing to disclose auditor fees are significantly smaller and more likely to have a non-Big
Five auditor. As shown below, small firms and firms with a non-Big Five auditor report lower fees for non-audit
February
5,
13 percent, while the
services.
10
percent of the sample reporting fees in this category. Finally, the
audit fees
the
(i.e.,
sum of IS and
other fees) to total fees
mandatory disclosures of fee
In addition to the
mean (median)
ratio
0.49 (0.51).
is
data, approximately one-third
sample (1,052 firms) voluntarily disclose information on the composition of other
provide a parsimonious description of this data,
related, e.g., audits
(iii)
of the
To
fees.^
these fees into four categories:
(i)
audit-
of employee benefit plans, regulatory audits, and preparation of registration
statements and other
consulting,
we code
of non-
SEC
combined
filings, (ii) tax, e.g., preparation
audit-related
and
tax, consisting
and
of tax forms, and tax-related
filing
of fees fi^om both of the previous two
categories, and (iv) other advisory, e.g., general consulting services,
and information technology
consulting for systems not associated with the financial statements. '°
B of Table
Panel
2 reports descriptive statistics of the components of other fees. Audit-
related fees are disclosed
by 13 and
1 1
percent of the sample, respectively, while 6 percent of the sample reports combined
audit-related and tax fees.
ranging from a
combined
by 21 percent of the sample. Tax and other advisory fees are disclosed
The
voluntarily disclosed fees are a substantial portion of other fees,
mean (median) of 46
audit-related
and tax
fees.
(45) percent for other advisory fees to 95 (100) percent for
Untabulated regression results confirm that the likelihood
of voluntary disclosure increases significantly with the magnitude of other fees relative
fees.
Moreover, conditional on the
Young
are
more
likely to disclose
ratio
of other fees
to total fees, firms audited
components of other
fees,
to total
by Ernst
&
while firms audited by the other
Big-Five auditors and by non-Big Five auditors are less likely to disclose.
^
The SEC's proposed
auditors.
rule on auditor independence contains a representative list of non-audit services provided by
These services include non-financial statement accounting and auditing, business controls and re-
engineering,
ta.x,
financial, information
distribution, legal services,
^°
The majority (60
generally include
and
systems technology, employee benefit, corporate finance, marketing and
litigation support.
percent) of firms disclosing other advisory fees are clients of Ernst
ta.x
fees with other advisory fees.
11
& Young, and these firms
Table
summarizes the fee data
3
Among
audit firms combined.
the
of the Big Five audit firms, and for
for each
Big Five,
total fees billed to the
differ in the
composition of total
fees.
one-third of total fees billed for any of the Big Five auditors,
by non-Big Five
The composition of fees
auditors.
& Young earns
Ernst
auditors.
KPMG.
In contrast, total
sample firms audited by non-Big Five auditors are only $64 million. The Big Five
and non-Big Five also
billed
other
sample firms range from
$1.77 billion for PricewaterhouseCoopers to just over $700 million for
fees billed to
all
Audit fees comprise no more than
compared
to
58 percent of total fees
Big Five
also varies across the
only six percent of its revenue from IS services, but
bills
more
for
other services (66 percent) than any other auditor. Non-audit fees range from 67 percent of total
fee
revenue for Arthur Andersen
to
75 percent for PricewaterhouseCoopers.
The remaining columns of Table
ratios
of fees billed
substantially
for the
in
each category
from non-Big Five
3 report,
auditor, the
Once
to total fees.
auditors.
by
The median
median of the
client-specific
again, Big Five auditors differ
ratio
of non-audit fees to
non-Big Five auditors, while for Big Five auditors the
total fees is 0.30
ratio is at least 0.50.
Thus,
at least
half of the Big Five's audit clients pay non-audit fees in excess of audit fees. However, note that
the
median
ratio
of non-audit fees
to total fees is
lower than the proportion of total fee revenue
derived from non-audit fees. Thus, a relatively small
disproportionate
The
the
SEC
in
data
amount of each of the Big Five
summarized
for a
auditor's non-audit fee revenue.
Tables 2 and 3 differs markedly from 1999 data considered by
formulating the rule on auditor independence
that non-audit fees
clients
in
number of clients account
(SEC 2000)." The
earlier data suggests
comprise ten percent of total revenues, with almost three-fourths of audit
purchasing no non-audit services. In contrast, our findings indicate that non-audit fees
" The SEC
obtained their data from the "Special Supplement: Annual Survey of National Accounting Firms
Public Accounting Report (March 3
1
,
2000) and from reports
12
filed
with the AICPA's
SEC
Practice Section.
-
2000,
comprise approximately 70 percent of total fee revenue (Table
purchasing non-audit services (Table
compile the 1999 data
this
comparison
may not be
2,
3),
with 96 percent of audit clients
Panel A). Although the categorization of fees used to
comparable
strictly
illustrates that the actual
to that required
under the
new SEC
magnitude and frequency of non-audit services
rule,
is
considerably greater than prior estimates.
Model
specification
Measures of non-audit service fees
We use the auditor fee disclosures to develop two measures of non-audit services.
first
measure
is
(FEERATIO). The SEC's purpose
the ratio of non-audit fees to total fees
requiring the fee disclosures
is
The
in
to provide investors with information to evaluate "whether the
proportion of fees for audit and non-audit services causes them to question the auditor's
independence" (SEC 2000, Section
Moreover,
III.c.5).
this ratio
has been used as a proxy for
auditor independence in prior research (Scheiner 1984; Glezen and Millar 1985; Beck, et
al.
1988; Parkash and Venable 1993; Barkess and Simnett 1994; Firth 1997). However, this
measure
is
invariant to the scale of the fees, and thus does not capture the relative financial
importance of the chent to the auditor. In addition,
which cross-sectional variation
in the ratio
is
it
is
not possible to determine the extent to
driven by the level of non-audit fees as opposed to
audit fees.
DeAngelo (1981) argues
that auditor
quasi-rents specific to a particular client,
dependent on each client
studies
have used the
is
ratio
independence
and suggests
is
inversely related to the value of
that the relative
magnitude of fees
a potential surrogate for independence. Lacking fee data, prior
of client sales to
total sales
Lys and Watts 1994; Reynolds and Francis 2001).
13
audited by the auditor
(e.g.,
Stice 1991
Om second measure of non-audit services
is
the percentile rank,
(RANKNON).
by
Firms
of the amount of non-audit fees disclosed by each firm
auditor,
in the highest percentile receive a
percentile receive a rank of
revenue to mitigate error
rank of 100 while firms in the lowest
We use percentile ranks rather than scaling by total audit
1.
in the
measurement of the
relative
firm
magnitude of client-specific quasi-
rents (Johnston 1984).
Although our study focuses on the provision of non-audit services, prior
generally does not
1997). Indeed,
make
some
between audit and non-audit fees
studies suggest that audit fees can also create
between the auditor and
may
a distinction
client
(e.g.,
literature
Hansen and Watts
economic dependence
(DeAngelo 1981; Simunic 1984). Alternatively, higher
audit fees
increase the auditor's investment in reputational capital, thereby increasing the loss from an
impairment of independence (Arrunada 1999). Several prior studies document a positive
correlation
between audit and non-audit
1995). Thus, in addition to
amount of audit
Our
RANKNON, we
fees disclosed
tests include
DeAngelo (1981) argues
fees (e.g.,
by each firm
Simunic 1984; Palmrose 1986; Craswell
include the percentile rank, by auditor, of the
(RANKAUD).
two additional auditor variables discussed
that
surrogate for audit quality.
even when quasi-rents vary across
The most common proxy
in the prior hterature.
clients, auditor size serves as a
for auditor size is an indicator variable for
Big Five auditors. Prior research fmds that Big Five auditors are less likely
management than non-Big Five
1998; Francis et
al.
auditors
(e.g.,
(e.g.,
Beck
et al.
to
allow earnings
DeFond and Jiambalvo 1991,1993; Becker
1999). Prior research also suggests that independence
length of auditor tenure
et al.
is
et al.
decreasing in the
1988; Lys and Watts 1994). Thus, in addition to the
fee-based measures discussed above, our analysis includes a Big Five indicator variable
(BIGFIVE) and
a variable measuring the
statements of the
number of years
company (AUDTEN).
14
the auditor has audited the financial
Estimation equations
We use two complementary approaches to identify firms managing earnings.
research
(e.g.,
documents
Burgstahler and Dichev 1997; Burgstahler and
companies
to
meet
levels, consistent
SEC and the POB
simple benchmarks. Both the
for public
Eames 1998; Degeorge
First Call,
between actual eamings per share and the
announcement of annual eamings, and
we
last
we
equity. 1'
calculate
in net
eamings surprises
eamings and small
relevant
benchmark during
(SURPRISE) and
'-
for the
at
zero in the distribution of
in contrast to prior research,
all
we
find
to
be a
focus the analysis on small eamings surprises
(INCREASE). '3
(1999), the bin width
variable times the negative cube root of sample size.
we
no
non-financial firms
sample firms. Because zero eamings does not appear
the sample period,
al.
by scaling net
income by beginning of the year market value of common
small mcreases in eamings
Consistent with Degeorge et
and Dichev
profits
evidence of a discontinuity in the distribution of eamings levels, either for
2000 or
as the difference
Similarly, following Burgstahler
eamings surprises and earnings changes. However,
in
2000).
meet or beat analysts' expectations
Untabulated results confirm a significant discontinuity
on Compustat
POB
(SEC 2000;
available consensus (median) forecast prior to the
identify firms reporting small increases in
income and the annual change
meet
and project a smooth eamings path creates
identify firms that just
as those reporting a surprise of 0.00 or 0.01.
to
expressed particular concern that the pressure
analysts' expectafions
Using data obtained from
1999)
eamings
with firms managing earnings
pressure on auditors to permit their clients to meet those objectives
(1997),
et al.
a significantly higher than expected frequency of firms with shghtly posifive
eamings changes, and eamings
surprises,
Prior
is
The
twice the interquartile range of the scaled eamings
resulting bin width
is
0.02 for both the change in eamings
and the level of eamings.
'3
We do not
find significant results
absence of a discontinuity
when we
in the distribution
repeat the tests below for firms reporting small profits, but given the
of earnings
this result is
not siuprising. Doubling the bin width to 0.04
produces a significant discontinuity; however, the interpretation of this finding is not
significant results using the wider bin width to identify firms reporting small profits.
15
clear.
We
also do not find
Prior research identifies several factors that influence firms' incentives to meet or beat
Matsumoto (1999)
earnings benchmarks.
prospects, and institutional ownership are
benchmarks.
We measure
company operates
in a
litigation risk
reports that firms with high litigation risk, growth
more
likely to
(LITRISK)
as
be concerned about missing earnings
an indicator variable equal
high risk industry identified by Francis
et al. (1994).
market-to-book ratio (M/B), and insfitutional ownership (%E<^ST)
held by institutions (reported by Spectrum), hi addition,
less likely to report positive earnings surprises,
indicator variable
(LOSS) equal
to
one
if the
Brown
may be more
Growth
one
is
the percentage of shares
firm reported a loss in 2000.
likely to beat earnings
assets
(CFO) because
benchmarks and may
and financing
and poor
performance are additional determinants of non-audit fees identified by Firth
(1997).''*
model includes an indicator variable (FIN/ACQ) identifying fums
(reported
by
Securities
net
CRSP
value-weighted market index
income divided by average
market value of equity
total assets.
Data Corporation).
(LOGMVE). We
(ANNRET), and return on
Finally,
we
Thus, our
that issued securities or
two measures of firm performance, the percentage compounded monthly return
for the
firms
also be better
activity
company during 2000
the
and thus our regression model includes an
able to afford outside consulting services. Acquisition
acquired another
if the
(2001) finds that loss firms are
We control for operating cash flow scaled by average total
with higher cash flow
is
to
assets
for
We include
2000 adjusted
(ROA), defined
control for firm size using the log of the
estimate the following logit model,
where
BENCHMARK indicates the earnings benchmarks, SURPRISE or INCREASE:
'"*
In the next section,
results
we
discuss the estimation of a
of this analysis are reported
in the
model of non-audit
Appendix.
16
as
fees similar to that of Firth (1997).
The
^fi,
Prob {BENCHM'iRK
)
=
F
+ ^.FEEHiTIO (or /S.^RANKNON + P,,RANKAUD
+ P^BIGFIVE + p,A UDTEN + p.LITRISK + P,
)^
MB
(1)
P^LOGMVE + p, %INST + p.LOSS + p^CFO
V+ P,qFIN/ACQ + p,,ROA + ^,2 ANNRET +u
+
Our second approach
managing earnings uses discretionary accruals
to identifying firms
estimated with a cross-sectional modified Jones (1991) model (see
Becker
et al.
1998; Bartov
2001
et al.
).
j
Discretionary accruals
DeFond and Jiambalvo
(DACC)
are equal
1994;
to:
DA CC = TA-[a + fi, [AREV - AREC] + '^^PPE)
where
all
variables are scaled
by
total assets at the
defined as net income less cash from operations,
the
oc.
change
Pi,
and
industry
in net receivables,
JSi
and PPE
is
beginning of the year, and
AREV is the change
is
in net
TA
identified using two-digit
model
at the
is total
revenues,
gross property, plant, and equipment.
are obtained from estimating the following
membership
(2)
The
accruals,
AREC
is
estimates of
industry level, where
SIC codes:
TA^a + p.AREV + p.PPE + e
(3)
We use the absolute value of discretionary accruals (ABSDACC) to measure the
combined
effect of income-increasing
(Warfield
et al.
1995; Becker et
al.
and income-decreasing earnings management decisions
1998; Reynolds and Francis 2001), and estimate the following
empirical model:
ABSDACC = « + PPEERATIO (or JS^^RANKNON + p.^RANKAUD) + /3,BIGFIVE
+ j3,AUDTEN + P.CFO + J3,ABSCF0 + J3^ACC + p^ABSACC
+ p^LEVERAGE + P^LITRISK + p,,M/B + P,,LOGMVE + /?,, %INST
+ P,,LOSS + P,,FIN/ACO + p,,ANNRET + e
We also present results for a partition
discretionary accruals
of the observations into firms with income- increasing
(DACC~) and income-decreasing
17
discretionary accruals
(DACC~).
(4)
Prior research shows that discretionary accruals models
fail to
completely extract non-
discretionary accruals that are correlated with firm performance. '^ Thus,
measures of firm performance,
all
of which are deflated by average
operations (CFO), the absolute value of cash from operations
and the absolute value of total accruals (ABSACC).
the ratio of total Habilities to total assets
be associated with discretionary accruals
Becker
et al. 1998).
We also
we
control for four
total assets:
(ABSCFO),
cash from
measured
control for leverage,
(LEVERAGE), because
as
prior research finds leverage to
(DeFond and Jiambalvo 1994; DeAngelo
The remaining control
(ACC)
total accruals
1994;
et al.
variables in equation (4) are as defined above.
Descriptive statistics
Table 4 reports descriptive
FEERATIO
is
buy more non-audit
sample firms partitioned
A of Table 2,
median of
findings reveal that
FEERATIO
RANKNON as well as RANKAUD.
In addition, clients
of Big Five
is
0.51.
services, consistent with the findings reported in
of the dependent variables in our regression models,
are significantly higher for firms
The remaining
at the
The
which, as shown in Panel
positively correlated with
auditors
statistics for the
Table
3.
All three
SURPRISE, INCREASE, and ABSDACC,
purchasing more non-audit services.
variables in Table 4 indicate that non-audit fees are significantly correlated
with several economic characteristics of the sample firms. High
FEERATIO
cash flows, but lower total accruals. In addition, firms in the high
significantly higher litigation risk, growth,
engage in financing and acquisition
and
acfivities.
median FEERATIO, but stock performance
significantly larger than other firms.
is
FEERATIO
institutional shareholdings,
Profitability
is
tests that
group have
and are more
also higher for firms
lower. Finally, high
The empirical
firms have higher
FEERATIO
likely to
above the
firms are
follow control for these factors.
'^ For example, Dechow et al.
(1995) show that discretionary accruals are biased for extreme values of earnings and
cash flows. For further discussion of these research design issues, see McNichols (2000).
18
EMPIRIC.4L RESULTS
IV.
management
Association between non-audit fees and earnings
Benchmark tests
The
first test
of the earnings management hypothesis examines the association between
The
firms that just meet or beat earnings benchmarks and the purchase of non-audit services.
findings, reported in Table 5, are mixed.'^
audit fee measures
These
The association between SURPRISE and the non-
FEERATIO and RANKNON
results suggest that firms purchasing
is
more non-audit
analysts' expectations, consistent with the earnings
(2000) and
POB
indicating that
the other
The
(2000).
positive
coefficient estimate
and
significant at the 0.01 level.
services are
more
management concerns expressed by
on
variables,
meet
the
SEC
RANKAUD is negative and significant;
low audit fees are also associated with earnings management
two auditor-related
likely to just
BIGFIVE and AUDTEN,
activity.
are significant.
Neither of
The
results for
the remaining controls indicate that firms reporting small positive earnings surprises have greater
litigation risk
The
and
institutional
ownership, are
basic premise of the
benchmark
missing benchmarks. Earnings management
benchmark but not
for
fmns
that just
larger,
analysis
is
thus
and are more
is
that firms
presumed
likely to report a profit.
manage earnings
to exist for firms that just beat the
miss the benchmark. Therefore,
we
estimate the
firms that just miss analysts' expectations (forecast errors of -0.01 or -0.02).
results indicate that
FEERATIO
and
0.01) in the "just miss" regressions;
'^
The
to avoid
model
The untabulated
RANKNON are negative and significant (p-values
RANKAUD
results are presented after the deletion
of statistical
is
iasignificant.'^
outliers.
However,
all
These
less
than
results provide
inferences are qualitatively similar
if outliers are retained.
'"^
We
also estimate the
specification, both
coefficient
model
FEERATIO
on RANTCA.IID
is
for firms that beat analysts' expectations (forecast errors of 0.02 or 0.03).
and
RANKNON
are insignificant (p-values of 0.70
negative, with a p-value of 0. 12.
19
for
In this
and 0.71, respectively). The
additional evidence consistent with a correlation between non-audit fees and earnings
management
to beat analysts' expectations.
Statistical inference in the regression containing
affected
by
multicollinearity because there
two independent
variables.'^ Thus,
untabulated results confirm that
surprises (p-value
=
significant (p-value
auditor,
0.09).
=
The
0.19).
we
is
RANKNON and RANKAUD could be
a high degree of correlation (+0.72) between these
estimate the
model using each variable
The
separately.
RANKNON is positively associated with small positive earnings
coefficient
hi addition,
on
we
RANKAUD is negative but not statistically
estimate the
model with
of the amount of total fees disclosed by each firm
the percentile rank,
(RANKTOT). Reynolds and
by
Francis
(2001) argue that larger clients create greater economic dependence regardless of the source of
fees.
However, we find
that the coefficient
on
RANKTOT is
suggesting that the composition of fees paid to the auditor
Contrary to the robust evidence for the
SURPRISE
is
msignificant (p-value
=
0.62),
relevant.
model, there
is
no evidence
that firms
paying higher non-audit fees are more likely to report small increases in earnings. Specifically,
both
FEERATIO
and
are consistent with
RANKNON are insignificant in the INCREASE regressions.
Gore
et al.
(2001)
who
find a positive association
Our results
between the purchase of
non-audit services and meeting analysts' expectations for U.K. firms with a Big Five auditor, but
no association between non-audit services and reporting small earnings increases or small
profits.'^
the
The
results for the control variables in the
SURPRISE
regressions are consistent with
regressions with two exceptions; firms reporting a small increase in earnings
have poor market performance and do not
'^
INCREASE
attract institutional
Correlations between the auditor fee meastires and auditor proxies
'^
owners.
BIGFIVE and AUDTEN
are less than
±
0.30.
Although our sample includes clients of non-Big Five auditors, 96 percent of the firms included in the Table 5
regressions are audited by a Big Five auditor. We also report below that our results are robust to exclusion of firms
with a non-Big Five auditor.
20
Discretionary accruals
The second
test
tests
of the earnings management hypothesis examines the relation between
discretionary accruals and the purchase of non-audit services. Table 6 reports
for the absolute value
of discretionary accruals in Panel
summary
statistics
A and for the separate partitions of
income-increasing and income-decreasing discretionary accruals in Panel B. The positive and
significant coefficients
more non-audit
on
services
FEERATIO
and
manage earnings
AUDTEN are negative and significant,
RANKNON in Panel A indicate that firms purchasing
to a greater extent
indicating that
than other firms.
low audit
fees
RANKAUD and
and longer tenure of the
auditor are associated with greater earnings management. Additionally, discretionary accruals
are correlated with firm performance, leverage, firm size,
Following Reynolds and Francis (2001),
model separately
results,
for income-increasing
presented in Panel
B
of Table
6,
we
and
institutional shareholdings.
also estimate the discretionary accruals
and income-decreasing discretionary accruals. These
reveal that
FEERATIO
significant for firms with positive discretionary accruals,
RANKNON are positive and
and
and negative and significant
for firms
with negative discretionary accruals. Assuming that companies with positive (negative)
discretionary accruals are attempting to
finding that firms purchasing
manage earnings up (down),
more non-audit
these results confirm the
services are better able to achieve their earnings
objectives. In contrast, the coefficient estimate
on
RANKAUD is negative (positive) and
significant for firms with positive (negative) discretionary accruals, suggesting that firms paying
high audit fees manage earnings less than other firms.
for firms
AUDTEN is
also positive and significant
with negative discretionary accruals. The results indicate that
total accruals (cash
flows) are positively associated with income-increasing (income-decreasing) discretionary
accruals. In addition, litigation risk
is
higher for firms with
21
more
positive discretionary accruals.
To examine
we
the robustness of our results to alternative
measures of discretionary accruals,
repeat the analyses in Table 6 using discretionary working capital accruals, defined as total
accruals less depreciation expense, and discretionary operating accruals, defined as
The untabulated findings
before special items less cash firom operations.
reported in Table
6.
Our
results are also robust to the inclusion
are correlated with discretionary accruals.
much of the
are similar to those
of two additional variables
We include the Altman Z-score
fmancial distress, and the change in inventory because
as a
we
estimate the
model separately
Consistent with the resuhs discussed above,
significant, but that
we
proxy
Thomas and Zhang (2001)
cross-sectional variation in discretionary accruals can be explained
inventory. Finally,
for each of the
find that
income
that
for
suggests that
by changes
in
ranked fee variables.
RANKNON and RANKAUD are
RANKTOT is insignificant.^^
Additional sensitivity analysis
Estimations for several alternative specifications indicate that the findings are generally
robust.
First, to further
ensure that the results are not driven by cross-secfional variation between
we
firms with a Big Five and non-Big Five auditor,
repeat the analyses in Tables 5 and 6 using
only the sample of firms with a Big Five auditor. The untabulated results are consistent with
those reported above. Second,
control, with
no change
partitioned at the
and the
ratio
R.\NKTOT
in inferences.
The
We
also estimate the
size,
measured using
results are qualitatively
work by Chimg and Kallapur (2001)
of the
IS
repeat the analyses with the log of total assets as the firm size
median value of firm
log of total assets.
20 Concurrent
we
either the log
unchanged, except that
reports no association
client's total fees to the auditor's total fees.
insignificant even though
models separately
for the
sample
of market value or the
FEERATIO
and
between absolute discretionary accruals
This result
is
consistent with our finding that
RANKNON and RANKAUD are individually significant.
Chung and
Kallapur (2001) also find no association berween absolute discretionary accruals and the ratio of client non-audit
fees to the auditor's total fees. We re-estimate the model using this measure instead of R,ANKNON and also find
insignificant results.
The difference
in these findings is consistent
quasi-rents with error, suggesting that our use of ranked variables
22
with the fee data measuring the client-specific
is
appropriate.
RANKNON are insignificant in the upper half of the
Thus,
we
find
management
Third,
fees
some evidence
is
we
FEERATIO
disaggregate
SURPRISE
regressions.
between non-audit fees and earnings
into
two components -
the ratio of other fees to total fees
SEC
services to their audit clients.
ISRATIO and OTHRATIO
for
fees.
to
that
independence, the preliminary
on June 27, 2000 prohibited auditors from providing these
The fmal
component of non-audit
the ratio of IS fees to total
(OTHRATIO). Because of concerns
of IS services poses a particularly strong threat
independence rules issued by the
the IS
in the
not significant for larger firms.
(ISRATIO) and
the provision
that the association
sample
rules relaxed the ban, but require separate disclosure of
We repeat the analyses
in Tables 5
FEERATIO. Only OTHRATIO
is
and 6 substituting
consistently significant in the
untabulated results, possibly because relatively few firms report IS fees.
To
control for systematic industry differences,
identified in Table
results
1
.
We
we
also repeat the analyses using a
include fixed-effects for the industries
common
from both of these specifications are consistent with our primary findings.
Market reaction
to the disclosure of
non-audit fees
The second hypothesis concerns
hypothesis using an event study.
information became public,
we
To
the valuation effects of non-audit fees.
identify the event date,
search the
the necessary fee data then
we use
i.e.,
We test this
the earliest date at
which the
fee
EDGAR database to identify frnns that filed a
preliminary proxy statement disclosing auditor fees.-' If we
-'
sample of 1,943 firms. The
fmd
a preliminary
proxy containing
the release date of the preliminary proxy as the event date; if
Approximately 10 percent (314 firms) of the definitive proxy statements
release of a preliminary proxy statement.
23
in
our sample were preceded by the
not,
we
To
use the release date of the definitive proxy.
we
eliminate observations if the proxy
We
obtain returns data fi-om the 2001
events,
market model prediction
errors.
was
mitigate the effects of confounding
filed within
CRSP
tape.
two days of 10-K, 10-Q, or S-K.^^
Abnormal
Market model coefficients
returns are event date
are estimated fi-om firm-level
regressions of raw retiims on Nlc'SE size-decile portfolio returns over the 100-day period ending
30 days prior
to the event.
Because stock prices
will react to the
new
information contained in
the auditor fee disclosures, tests of the shareholder wealth hypothesis require a proxy for the
unexpected component of non-audit
fees discussed
above
to estimate the
The
fees.
first set
of tests uses the measures of non-audit
following regressions;
ARET = a, +
/?,
FEER4 TIO + £,
(5)
ARET = j3,+/3, MNKNON + p.RANKNON + e
The
fee measures in equations (5)
if non-audit fees
and
assume uniform prior expectations, and
(6)
contain a large unexpected component relative to
component. Anecdotal evidence suggests
audit fees (e.g.,
We
proxy for unexpected non-audit
assumes
fees.
that non-audit fees vary with
Appendix provides
assumes
details
(FEERESEDUAL)
This prediction error
large, positive
unexpected shock.
using previously released data
is
the residual fi-om a
model
to
that
agency costs (Parkash and Venable 1993; Firth 1997). The
that the publicly available data items in our
when
expected firm-specific
were surprised by the magnitude of non-
we
model
are those used
by investors
to
form
also estimate the following model:
Approximately 20 percent (614 firms) of the sample firms were deleted for
similar
its
are appropriate
on the empirical estimation of the model. This alternative proxy
expectations of non-audit fees. Thus,
^^
that investors
Weil and Tannenbaum 2001), consistent with a
also derive a prediction error
(6)
these observations are included in the analysis.
24
this reasoti.
Results are qualitatively
ARET = j8,+j3, FEERESID UM + £
Overlapping event dates
effects
may produce dependence
of cross-sectional dependence on significance
coefficients
by estimating
the event date.
Because
all
(7)
across firm-events.
levels,
we
Non-event dates are taken from the
control for the
create distributions of regression
models on 200 non-event-dates beginning
the
To
intervals [r-141, ?-2]
at
date /-141, where
and
/ is
[t+2, t+6l].
firm-observations are shifted to non-event dates as a block, firm-observations in each
non-event date regression are temporally related as they are in the event date regression.
Summary
reported in Table
statistics
7.
The
from the empirical estimation of equations
coefficient estimates on
FEERATIO
and
(5)
through (7) are
RANKNON are negative but
not significant after controlling for cross-sectional dependence (two-tailed bootstrap p-values are
0.14 and 0.15, respectively). In contrast,
expected portion of non-audit
These
results provide
fees, is negative
is
that non-audit fees
explicitly controls for the
and significant (bootstrap p-vsLlue=0.02)P
modest evidence consistent with investors reducing valuations
compensation for a perceived reduction
analysis
FEERESIDUAL, which
may have
in audit quality.
However, an important caveat
as
to this
implications for the market's expectations of future cash
flows unrelated to audit quahty. For example, high non-audit fees
quality or that the firm faces difficulties previously
unknown by
may
signal poor
investors.
management
Distinguishing this
"performance" explanation from the "audit quality" explanation could be resolved by an
^'*
examination of the subsequent accounting performance of sample firms.
23
One
observation (Viador Inc) was found to be influential and deleted. The
of an interim
stock price.
company announced
the appointment
on May, 1, 2001, the same date its proxy statement was filed, leading to an 127% increase in
Because this company had relatively low non-audit fees, its inclusion would increase the significance of
CEO
the results.
2'*
One
is that over 99
of the proxy statement, thus pre-empting
factor that suggests that investors are responding to a perceived reduction in audit quality
percent of our sample had released fiscal
least a portion
2000 earnings
prior to the fihng
of the performance-related information content of the fee data.
25
at
V.
SUMMARY AND CONCLUSION
This paper provides evidence on the association between non-audit fees and earnings
management, as well as evidence on
fees.
We
to just
find that firms purchasing
meet or beat
accruals.
more non-audit
find
we
ratio.
likely to
by implication, lower
engage
in earnings
the extent that earnings
management
is
is
one
attribute
of non-audit
We also find evidence that the stock market
decreasing in the unexpected component of the
disclosure requirements
may
quality earnings.
that firms
paying high (low) non-audit (audit) fees are
management, we do not address the
magnitude of non-audit and audit fees
SEC
likely
This result indicate that investors associate non-audit fees with lower
Although our evidence indicates
more
more
to report larger absolute discretionary
interpret these results as suggesting that the purchase
reaction to the disclosure of non-audit fees
quality audits and,
services from their auditors are
and
services is associated with lower quality earnings.
non-audit to total fee
announcement of non-audit
to the
no association between the purchase of non-audit services and the
meet prior year's earnings. To
of earnings quality,
market reaction
analyst earnings expectations
However, we
ability to just
the stock
is
related issue of whether the
associated with financial statement fraud. Additionally,
alter the existing
equilibrium in the audit services market. For
example, the disclosure of fee data could increase the competitiveness of the audit market by
reducmg
the cost to firms of making price comparisons and negotiating fees. In addition,
irrespective of the relation
between non-audit services and earnings
quality, firms
purchase of non-audit services to avoid the appearance of non-independence.
26
may reduce
the
APPENDLX
Estimation of Unexpected Non-Audit Fees
In this appendix,
we
The purpose of this model
market reaction
tests.
develop a model
is
to estimate the
We base
examine a company's decision
We
quality audits
FEERATIO.
unexpected portion of non-audit fees for use
in our
our model on Parkash and Venable (1993) and Firth (1997)
to
and
They argue
restrict the
estimate the following
who
purchase non-audit services from their auditor, conditional on
the level of auditing services purchased.
demand high
to explain cross-sectional variation in
that
companies facing high agency costs
purchase of other services from their auditor.
model using
all
observations with the necessary data:
13
FEEMTIO = ^aJND,
+ fi.BIGFIVE + jS.AUDTEN + p.ROA + P.LOSS
1=1
+ p.ANNRET + p, %INST + p,CFO + p^LEVERAGE + p^INVREC
+ P^FIN/ACQ + P.LOGiVIVE + j3,,PI/B + v
We include industry intercepts, IND„ for each of the
Panel
C
of Table
1
.
.An indicator variable
auditors are able to provide
AUDTEN,
the
company, controls
may proxy
for
for the possibility that
services and/or charge a
premium
agency costs as well as the potential
We include two measures
defined as net income divided by average total assets, and
equal to one if the firm reports a net loss.
the
to 13 industries identified in
compounded CRSP monthly
Big Five
for their services.
of the
of the relationship between the auditor and chent.
from the purchase of outside consulting.
ROA,
1
that the auditor has audited the financial statements
for the length
Performance
(BIGFIVE) controls
more non-audit
number of years
/=
(Al)
We also
for the firm to benefit
of accounting performance,
LOSS, an
indicator variable
include stock return performance,
return in calendar 2000. Institutional ownership
percentage of shares held by institutions (reported by Spectrum), and
total liabilities to total assets.
27
ANNRET,
(%INST)
LEVERAGE
is
is
the
the ratio of
We include
ability to
operating cash flow scaled by average total asset
purchase outside consulting services. In addition,
INVREC
complexity of services performed by the auditor.
as a percent
thus
we
of total
assets.
include, and thus
Rapidly growing companies
we
include
FIN/ACQ, an
company
securities or acquired another
in 2000,
statistics
is
to control for firms'
control for the extent and
inventory plus accounts receivable
may demand more
consulting services
indicator variable for firms that issued
and the market to book
control for firm size with the log of market value of equity
Summary
we
(CFO)
ratio
(M/B). Finally,
(LOGMVE).
from the estimation of equation (Al) are reported in Table Al
model explains 31 percent of the cross-sectional variation
we
in
FEERATIO, comparable
.
The
to the
explanatory power of the models in Parkash and Venable (1993) and Firth (1997). The results
indicate that firms in three industries, food, extractive,
services, while firms in the
retail,
purchase fewer non-audit
computer industry purchase more non-audit services. Consistent with
Parkash and Venable (1993) and Firth (1997),
poorly performing firms, as measured by
However, neither
and
institutional
find that firms with Big Five auditors and
ROA and ANNRET, purchase more non-audit services.
ownership nor leverage
activities also contribute to significantly
fees are increasing in firm size
we
is
significant. Acquisition
higher non-audit fees. Finally,
and decreasing
in auditor tenure
28
we
and growth.
and financing
find that non-audit
TABLE Al
Estimation of Unexpected Non-Audit Fees
13
FEEHA no =
X
(=1
«, ^^D,
+ A BIGFIVE + p._ A UDTEN + p,ROA + P,L OSS + p, ANNRET
-
+ P^ %INST + p, CFO + P.LEVERAGE + P.INVREC + P,^FIN/ACQ + p,,LOGMVE + P.^MB + v
Coefficient Estimate
Variable
t-statistic
-1.29
Agriculture
-1.09
-2.20
-0.94
-0.72
-0.37
-3.79
-0.30
0.74
-1.62
-2.40
-1.77
6.08
4.46
-3.62
-2.74
1.32
^.69
-0.63
1.16
0.99
-1.08
8.08
20.01
-5.87
Industry
membership
is
determined by SIC code as follows: agriculture (0100-0999),
1999, excludmg 1300-1399), food (2000-21
1
1),
textiles
minmg
& pnnting/'publishing (2200-2799),
&
construction (1000-
chemicals (2800-2824,
2840-2899), pharmaceuticals (2830-2836), extractive (2900-2999, 1300-1399), durable manufacturers (3000-3999,
excluding 3570-3579 and 3670-3679), transportation (4000-4899),
utilities
(4900-4999),
(7000-8999, excludmg 7370-7379), and computers (7370-7379, 3570-3579, 3670-3679).
if
the firm's auditor
is
a
Big Five
firm,
and zero otherwise;
retail
(5000-5999), services
BIGFIVE
is
equal to one
AUDTEN is the number of years that the auditor has
ROA
is net income divided by average total assets; LOSS is equal to one if
and zero otherwise; CFO is cash from operations, deflated by average total
assets; ANNRET is the percentage compounded monthly stock return for calendar 2000 minus the CRSP valueweighted market index; %INST is the percent of shares held by institutions (reported by Spectrum); LEVER^AGE is
the ratio of total assets to total liabilities; INVREC is inventory plus accounts receivable as a percentage of total
audited the firm's financial statements;
the firm reported a net loss in fiscal 2000,
assets;
is
FIN/ACO
is
equal to one
if
the firm issued securities or acquired another
the log of market value of equity;
and M/B
is
the market-to-book ratio.
29
company
in fiscal
2000; LOGArfVE
REFERENCES
Abbott, L.,
G. Peter, and D. V. Rama. 2001. Audit, non-audit and information
S. Parker,
technology
fees:
some empirical evidence. Working paper, University of Memphis.
Abdel-khalik, A. R. 1990. The jointness of audit fees and
analysis.
Antle, R. 1982.
Contemporary Accounting Research
The auditor
as an
economic
6:
demand
for
MAS:
a self-selection
295-322.
agent. Journal
ofAccounting Research 20: 503-527.
Antle, R. 1984. Auditor independence. Journal ofAccounting Research 22: 1-20.
Antle, R., P. A. Griffen, D.
J.
Teece, and O. E. Williamson. 1997.
An economic
analysis of
auditor independence for a multi-client, multi-service public accounting firm
prepared for the
AICPA by The Law & Economics
.
Report
Consulting Group, Inc.
Arrunada, B. 1999. The Economics of Audit Quality: Private Incentives and the Regulation of
Audit and Non-Audit Services. Norwell, MA: Kluwer Academic Publishers.
Baiman, S. J.H. Evans HE, and
Research 29: 1-18.
J.J.
Nagarajan. 1991. Collusion in auditing. Journal ofAccounting
The provision of other
Accounting and Business Research
Barkess, L., and R. Simnert. 1994.
pricing issues.
C, and
Barry,
S.
services
by
auditors: independence
and
24: 99-108.
Brown. 1985. Differential information and security market equilibrium. Journal
ofFinancial and Quantitative Analysis
Bartov, E., F. A. Gul, and
20: 407-422.
2001 Discretionary accruals models and audit
qualifications. Journal of Accounting and Economics 30: 421-452.
Bazerman, M.
H., K. P.
J.
P. J., T.
J.
.
Morgan, and G.
independence. Sloan
Beck,
S. L. Tsui.
Frecka, and
F.
Loewensein. 1997. The impossibility of auditor
Management Review (Summer):
I.
Solomon. 1988.
MAS involvement and auditor tenure:
Accounting Literature
M.
7:
An
89-94.
empirical analysis of the relationship between
implications for auditor independence. Journal of
65-84.
DeFond, J. Jiambalvo, and K. R. Subramanyam. 1998. The effect of audit
quality on earnings management. Contemporary Accounting Research 15: 1-24.
Becker, C. L.
Beeler,
J.
D.
,
,
and
L.
J.
E.
Hunton. 2001. Contingent economic rents: precursors to predecisional
distortion of client information.
Working paper, University of South
Brown, L. D. 2001. A temporal analysis of earnings
Accounting Research 39: 221-241.
30
Carolina.
surprises: profits versus losses.
Journal of
I. Dichev. 1997. Earnings management to avoid earnings increases and
Journal ofAccounting and Economics 24; 99-126.
Burgstahler, D., and
losses.
Burgstahler, D., and
M. Eames.
1998.
Management of earnings and
analyst forecasts.
Working
paper, University of Washington.
Chung, H., and
S.
Working
Kallapur. 2001. Client importance, non-audit services, and abnormal accruals.
paper, Purdue University.
Coffee, J.C. 2001.
The acquiescent gatekeeper:
reputational intermediaries, auditor
independence, and the governance of accounting. Working paper, Columbia University
School of Law.
Crasw^ell, A. T.,
J.
name reputations and
and Economics 20: 297-322.
R. Francis, and S. L. Taylor. 1995. Auditor brand
industry specializations. Journal ofAccounting
Davis, L. R., D. N. Ricchiute, and G. Trompeter. 1993. Audit effort, audit fees, and the provision
of nonaudit services to audit chents. The Accounting Review 68: 135-150.
DeAngelo,
L. 1981. Auditor size
and audit
quality.
Journal ofAccounting and Economics
3:
183-
^199.
DeAngelo, and D. J. Skiimer. 1994. Accounting choice
Journal ofAccounting and Economics 10: 193-225.
DeAngelo, H.,
Dechow,
P.
Dechow,
P.
L.
in
troubled companies.
M., R. G. Sloan, and A. P. Sweeney. 1995. Detecting earnings management. The
Accounting Review 10: 193-225.
M., R. G. Sloan, and A.
P.
Sweeney. 1996. Causes and consequences of earnings
manipulation: an empirical analysis of firms subject to enforcement actions by the SEC.
Contemporary Accounting Research
13: 1-36.
DeFond, M. L., and J. Jiambalvo. 1991. Incidence and circumstances of accounting
Accounting Review 66: 6^3-655
DeFond, M.
L.,
and
J.
errors.
The
Jiambalvo. 1993. Factors related to auditor-chent disagreements over
income-increasing accounting methods. Contemporary Accounting Research
9:
4 II -43
DeFond, M. L., and J. Jiambalvo. 1994. Debt covenant violation and manipulation of accruals.
Journal ofAccoimting and Economics 17: 145-176.
F., J. Patel, and R. Zeckhauser. 1999. Earnings management
Journal of Business 12: 1-34.
Degeorge,
to
exceed thresholds.
Dopuch, N., R, King, and R. Schwartz. 2001. Auditors' independence: appearance
Working paper, Washington University.
31
vs. fact.
1.
Feroz, E. H., K. Park, and V.
S.
The
Pastena. 1991.
financial
and market
effects
of the SEC's
accounting and auditing enforcement releases. Journal ofAccounting Research 29: 107142.
Firth,
M. 1997. The provision of nonaudit
services by accounting firms to their audit clients.
Contemporary Accounting Research
Francis,
J.
D., D. Philbrick,
disclosures. Journal
Francis,
R., E. L.
J.
and K. Schipper. 1994. Shareholder
ofAccounting Research 32: 137-164.
Maydew, and H.
C. Sparks. 1999.
A
reporting of accruals. Auditing:
Glezen, G. W., and
J.
A. Millar. 1985.
disclosures required by
Gore,
14: 1-21.
P., P. F.
and corporate
role of Big 6 auditors in the credible
Journal of Practice
An
ASR No.
The
litigation
&
Theory
18: 17-34.
empirical investigation of stockholder reaction to
250. Journal of Accounting Research 23: 859-870.
Pope, and A. K. Singh. 2001. Non-audit services, auditor independence, and
earnings management.
Working
paper, Lancaster University.
Hansen, S. C, and J. S. Watts. 1997. Two models of the auditor-client interaction:
United Kingdom data. Contemporajy Accounting Research 14: 23-50.
Johnston,
Jones,
J.
J. J.
1984. Econometric Methods. 3rd. Edition.
1991. Earnings
management during import
New York, NY:
tests
McGraw-Hill,
relief investigations.
with
Inc.
Journal of
Accounting Research 29: 193-228.
Levitt,
A. 1998. The Numbers Game. Remarks delivered
at the
NYU Center for Law and
Business, September 28.
Lys, T., and R. L. Watts. 1994. Lawsuits against auditors. Journal of Accounting Research 32:
65-93.
Magee, R. and M. Tseng. 1990. Audit pricing and independence. The Accounting Review
65(2):
315-336.
Matsumoto, D. 1999. Management's incentives
University of Washington.
to guide analysts' forecasts.
McNichols, M. F. 2000. Research design issues in earnings management
Accounting and Public Policy, forthcoming.
Working
studies.
Journal of
Palmrose, Z-V. 1986. The effect of non-audit services on the pricing of audit services:
evidence. Journal of Accounting Research 24: 405-41
paper,
fiirther
1.
Palmrose, Z-V. 1999. Empirical research on auditor litigation: considerations and data. Studies
Accounting Research #33. American Accounting Association: Sarasota, FL.
32
in
Parkash, M., and C. F. Venable. 1993. Auditee incentives for auditor independence; the case of
nonaudit services. The Accounting Review 6S: 113-133.
Pitt,
Birenbaum. 1997. Serving the Public interest: A New Conceptual
Framework for Auditor Independence. New York; NY: American Institute of Certified
H.
and D.
L.,
E.
Public Accountants.
Public Oversight Board. 2000. Panel on Audit Effectiveness: Report
and Recommendations.
Stamford, CT: American Institute of Certified Public Accountants.
Reynolds,
J.
K., and
J.
R. Francis. 2001.
Does
The influence of large clients on
ofAccounting and Economics 30: 375-
size matter?
office-level auditor reporting decision. Journal
400.
Scheiner,
J.
H. 1984. The impact of
SEC
independent auditors and their
Research 22: 789-797.
nonaudit service disclosure requirements on
clients:
an empirical investigation. Journal of Accounting
and Exchange Commission. 2000. Final Rule: Revision of the Commission's Auditor
Independence Requirements. Washington D.C.
Securities
Simunic, D. 1984. Auditing, consulting, and auditor independence. Journal of Accounting
Research 22: 679-702.
D. Using financial and market information to identify pre-engagement factors associated
with lawsuits against auditors. The Accounting Review 66: 516-513.
Stice,
J.
Teoh,
S.
H., and T.
coefficient.
J.
Wong.
1993. Perceived auditor quality and the earnings response
The Accounting Review
6S:
346-367
Thomas, J. K, and H. Zhang. 2001. Inventory changes and future
Columbia University.
returns.
Working
paper,
Trompeter, G. 1994. The effect of partner compensation schemes and generally accepted
accounting principles on audit partner judgment. Auditing:
Theory
A Journal of Practice and
13: 56-69.
M
Wallman, S.
.H. 1996. The future of accounting,
Accounting Horizons 10: 76-97.
Warfield, T. D.,
J. J.
part EQ: Reliability
Wild, K. L. Wild. 1995. Managerial ownership, accounting choices, and
informativeness of earnings. Journal of Accounting
Watts, R. L., and
J.
and auditor independence.
L.
Zimmerman. 1986.
and Economics
Positive Accounting Theory.
Prentice-Hall Inc.
33
20: 61-91.
Englewood
Cliffs,
NJ:
Weil,
J.,
and
J.
Tannenbaum. 2001. Big companies pay
Wall Street Journal, April
10:
CI.
34
audit firms
more
for other services.
The
TABLE
1
Sample Description
Panel A: Sample selection
Observations
Selection criteria
Definitive proxies filed between 2/5/0 1 and 6/ 1 5/0
Less:
4,70
Financialinstitutions (SIC codes 6000-6999)
(1,124)
Proxies not relating to annual meeting of shareholders
(51)
Change
(71)
in auditor
during fiscal year 2000
3,455
Missing fee data
(218)
Observations with fee data available
3,237
Observations not available on Compustat
(163)
Sample Observations
3,074
Panel B: Distribution of observations by month
Month
N
February
March
April
May
June
Tot^^
3,074
100.00
TABLE
1 -
continued
Sample Description
Panel C: Distribution of observations by industry
Sample
Industry Description
Compustat
TABLE
2
Descriptive Statistics of Auditor Fees Disclosed in Definitive Proxy Statements
Filed Between February
5,
2001 and June
15,
2001 (N=3,074)
Panel A: Mandatory disclosures offee data
Standard
Variable
Audit
Audit/Total
Mean
Third
Firs!
Deviation Ouartile
Median
Ouartile
Minimum Maximum
%>
s
i
OS
u
in
wJ
a,
<
^
0)
£
g
5:
TABLE
4
-
continued
Sample Descriptive
FEEIL4 no
is
Statistics
the ratio of financial information system and other fees to total fees;
RANKNON is
the percentile rank
by auditor; BIGFIVE is equal to one if
the firm's auditor is a Big Five firm, and zero otherwise; AUDTEN '\s the number of years that the auditor has audited
the firm's financial statements; SURPRISE is equal to one if the firm just meets or beats the consensus analyst
forecast (i.e., forecast error of 0.00 or 0.01) for fiscal 2000, and zero othenvise; INCREASE is equal to one if the
firm reports a small increase m earnings relative to the prior year, and zero otherwise; ABSACC is the absolute value
of total accruals, equal to net income minus cash from operations, deflated by average total assets; CFO is cash fi-om
operations, deflated by average total assets; ABSCFO is the absolute value of cash fi-om operations deflated by
average total assets; ACC is total accruals, equal to net income minus cash firom operations, deflated by average total
assets; LITRISK is equal to one if the firm is in a high litigation risk industry identified by Francis et al. (1994)
(SIC's 2833-2836, 3570-3577, 7370-7374, 3600-3674, 5200-5961), and zero otherwise; M/B is the market-to-book
ratio; %INST is the percent of shares held by institutions (as reported by Spectrum); LOSS is equal to one if the firm
reported a net loss in fiscal 2000, and zero otherwise; FIN/ACQ is equal to one if the firm issued securities or
acquired another company in fiscal 2000; LEVERAGE is the ratio of total assets to total liabilities; ROA is net
income divided by average total assets; ANNRET is the percentage compounded monthly stock return for calendar
2000 minus the CRSP value-weighted market index; Ml'^E is the market value of equity (in millions); and ASSET is
the book value of total assets (in millions).
of non-audit fees, by auditor; IL4NIC4UD
is
the percentile rank of audit fees,
40
+
^
p
o
V
Si
O
§
cs
B
u
c
+
Si
ca
e)
i
+
+
g
5
^
exi
_c
'c
«
OX)
I
•D
<
-s
p.
-a
i
o
o
i
^
^
+
^+
o
OX)
=1.
o
II
S
-^
ON
+
—
o
'
VO (N Tt
cs <N r-~
~
o
oooooo'ooooooo
V
V
TABLE
Summary
Statistics
Panel A: Dependent variable
is
6
from Discretionary Accruals Regressions
absolute value of discretionary accruals
ABSDACC = a + P.FEERATIO (or P^^RANKNON + p^.RANICWD) + P.BIGFIVE
+ j3,AUDTEN + /3,CF0 + jS.ABSCFO + jS.ACC + J3,ABSACC +
+ P,,M/B + fi,,LOGMVE +
^,,
^.LEVERAGE + J3,LITRISK
%INST + p,,LOSS + /3,,FIN/ACQ + /3^,ANNRET + £
ABSDACC
+
I
^
^+
e
M
V
u
-
+
Q
g
0:;
+
u
+
^
^ +
§•2
^
oq
ta
f
I
^
£
•a
E
+
+
^
-Q ^
< S
"-^
c«
•£?
^ ^+
5
cci,
-Si
^
S o
g
QC^
II
"^
r^
+
"^
5 i ^
£
c?
^
Q
^+
00
§•
E
.o
3
CO
c
e
E
o
s
Si
975
h^
m
Lib-26-67
MIT LIBRARIES
3 9080 02246 1211