Discussion of:

Do Joint Audits Improve or Impair Audit Quality?

By Mingcherng Deng, Tong Lu, Dan Simunic, and Minlei Ye, 2012.

Joint Audit (Shared):

An audit performed by two or more auditors to produce a single audit report.
Shared responsibility,
Typically, audit planning is jointly performed and fieldwork allocated to the auditors.

Joint Audit (Co-Audit):

An audit performed by two independent auditors issuing their own separate reports.

Source: Le Vourch and Morand (2011)

1
th

Mark Penno, 20 Annual Audit Symposium

Audit Quality : De Angelo (1981)1 defines audit quality as function of the auditors
(Technical) competence
Independence

2
th

Mark Penno, 20 Annual Audit Symposium

Commonly-Held Views of Joint Audits:

PRO: European Commissions (2010) Green Paper argues that joint audits increase audit quality and
mitigate concentration in the audit market.
CON: The joint audit also faces strong resistance, mainly a concern for additional cost. (See
appendix to these slides).

A BALANCED VIEW (Illustrated):

I think added quality, value and independence for say a maximum of 5 % additional cost is a
way good deal compared to other initiatives.
--Patrick de Cambourg, Chairman and CEO of Mazars (French accounting firm), 2011.1

Accessed (10/9/12) http://ec.europa.eu/internal_market/accounting/docs/conference20110209/speech_decambourg_en.pdf

3
th

Mark Penno, 20 Annual Audit Symposium

WHAT FINDINGS MAKE THIS PAPER INTERESTING?

Joint audits may cost less the single auditor audits

Joint audits may suffer from lack of independence more than single auditor audits
Large/small auditor pairings may not be effective for joint audits.

4
th

Mark Penno, 20 Annual Audit Symposium

How were these results achieved? (THE MODEL)

The auditors evidence, y, has the following properties:

Weight applied to the

difference between raw
signal and its expectation

Updated
expectation of
true firm value, x

e
rI E~
x
x
y E~
eh

Raw audit
signal

Expected value of
the audit signal

5
th

Mark Penno, 20 Annual Audit Symposium

SINGLE AUDITOR
Time line:

Auditor obtains a

Auditor
expends

and updates
expected firm

Client may
offer a bribe,
Q, in return for
reporting ,

resources, e0

value to rI

r rI

raw result, y0
Client hires
auditor with
fee, F

Auditor suffers an
expected loss
2
E r ~
x | y

6
th

Mark Penno, 20 Annual Audit Symposium

Other Special Features of the Model

Recall: firm value is x, auditors evidence is y.

1. The single auditor audit produces one signal, y 0 . The client bribes the auditor whenever
y0 E~
x | y0 and the auditor acquiesces. Bribes occur of the time (?).
2. The joint audit produces two conditionally iid signals y1 and y 2 with individual cost
functions identical to the single auditor case. The client bribes the auditors whenever
Maximuny1 , y 2 E~
x | y1 , y2 . That is the client will cherry pick the best signal and the
auditors jointly acquiesce. Bribes will occur of the time (?).
3. Auditor(s) acquiesce because the bribe perfectly compensates her (them) for misreporting.
4. The individual cost function for each auditor working jointly is assumed to be identical to the
cost function of a single auditor working alone (see next slide):

7
th

Mark Penno, 20 Annual Audit Symposium

Cost of audit effort, e

Cost
2.0

1) Auditors cost in a joint audit; or

2) Auditor cost in a single auditor audit

1.5

1.0

0.5

0.5

1.0

a Full
audit

1.5

2.0

Audit
Input, e

Full
audit

Key assumption: Two different auditors can cooperatively do a full audit (by splitting
tasks) more cheaply than a single auditor working alone.
8
th

Mark Penno, 20 Annual Audit Symposium

RESULT 1: (Two equal auditors put in the same total effort as found in a single audit)

e1 e2 e0
Joint Audit
combined
effort

Audit effort
under single
auditor

Intuition
1. The free-riding problem (auditors select effort unilaterally rather cooperatively) is offset
by
2. The cost benefit (the auditors marginal cost is lower at e/2 than it is at e).

OVERALL RESULT: The joint audit with equal auditors cost less than the single auditor audit, but
has more misrepresentation (i. e, cherry picking).

9
th

Mark Penno, 20 Annual Audit Symposium

TWO UNEQUAL AUDITORS IN A JOINT AUDIT (Big/Small instead of Big/Big)

The first auditor is identical to one of the previous auditors examined

The second auditor is inferior (its costs are higher)
The first auditors share of the liability > , in part, caused by the weaker auditor.
Result: The total effort for an unequal joint audit may be DIFFERENT than total effort for EQUAL
joint audit. This may result in less overall effort, potentially making this option inferior.

10
th

Mark Penno, 20 Annual Audit Symposium

GENERAL OBSERVATIONS

1. Model consistency: While the reports are made independently (no coordination) to the client,
the bribe is based jointly on y1 and y 2 (suggesting coordination).
2. Aggregation: If, instead, the average of y1 and y 2 used, then the superiority of the single
audit (versus equal joint auditors) would seem to disappear. Now, the average will exceed rI
only of the time.
3. Negotiation: To the extent that both auditors in a joint audit must sign off on a single report,
and are ambiguity averse, two auditors (in a modified setting) may jointly agree to a bribe only
Miny1 , y2 E~
x | y1 , y2 . In this case the probability of bribery falls to .
4. Quasi-rents: The bribe is referred to as a quasi-rent. Quasi-rent is an analytical term in
economics, for the income earned, in excess of post-investment opportunity cost, by a sunk
cost investment. [Because the auditor here breaks even the quasi-rent is really zero.]

11
th

Mark Penno, 20 Annual Audit Symposium

France?

France (Mandatory)
Francis et al (2009) found that despite the dominance by Big 4 firms in terms of overall
revenues, only 11.5 percent of companies are audited by two Big 4 auditors, which means
most listed companies in France have either one or two French accounting firms as their
auditors (Francis et al.)

The joint audit requirement serves to make the French audit market less concentrated
(France has one of the least concentrated audit markets in Europe).

12
th

Mark Penno, 20 Annual Audit Symposium

The joint audit requirement seems to have enabled midtier and small audit firms to corner a
relatively significant share of the market (Le Vourch and Morand, 2011):

Type of joint auditor pairs for French companies listed on regulated

markets (by number of mandates, 2009, detail)
Type of auditor pair
1 Big 4, 1 small audit firm
2 Big 4
1 Big 4, 1 midtier audit firm
1 midtier audit firm 1 small audit firm
2 small audit firms
2 midtier audit firms
Number of companies with 2 auditors

Number of
companies
171
90
87
57
51
13
469

%
35%
19%
18%
12%
11%
3%
97%

Source: Le Vourch and Morand (2011)

The Green Paper suggested the use of joint audits with at least one smaller audit firm (e.g., not
a Big 4 audit firm) for the audits of large companies to mitigate the concentration and enhance
the market structure.
13
th

Mark Penno, 20 Annual Audit Symposium

Consider, an alternative cost function (not in paper):

Cost

Single Auditors in
Joint Audit (sum)

Coordination
costs 1

Single Auditor
Audit
0.5

1.0

1.5

2.0

Audit
Input
E.g., The President of Ernst & Young France and Southern Europe, stated that joint
audits cost more (in the order of 20%) since it doubles the number of people at
meetings (referenced Andr et al., 2012)
14
th

Mark Penno, 20 Annual Audit Symposium

CORRUPTION INDEXES AND MANDATORY JOINT AUDITS

Countries with a joint audit requirement include France, Kuwait, Saudi Arabia, South Africa
(financial services), Tunisia, Morocco, Algeria, the Ivory Coast, , and Congo.

Country

Sweden
United Kingdom
United States
France
Kuwait
Saudi Arabia
South Africa
Tunisia
Morocco
Algeria
Ivory Coast
Congo Republic

Transparency
International
Corruption
Perceptions
Index 2011*
9.6
7.8
7.1
7
4.6
4.4
4.1
3.8
3.4
2.9
2.2
2.2

15
th

Mark Penno, 20 Annual Audit Symposium

Summary
Because the auditing environments are complex, it is challenging to provide a
comprehensive economic model of an audit market.
Political factors may also be at work (e.g., long-term concerns with market
concentration, or poor national business infrastructure).
Careful modeling is useful however in isolating critical forces that may have an
impact on auditing practice.
This is a thought-provoking paper that identifies issues that need to be considered
by the accounting community.

16
th

Mark Penno, 20 Annual Audit Symposium

Additional notes (appendix):

Team problem: The assumed 50% liability sharing represents proportionate liability (e.g., France).
But if two auditors are jointly responsible, then could one firm be held liable for the actions of the
other? If instead, the liability were joint and several, the auditors would have more incentive to
cooperate. [Holmstrom (1982) suggests penalizing the whole team when free-riding is a problem.]
Many countries have joint and several liability, however. In Sweden, firms are jointly liable for
opinions issued (Haapamki et al., 2012).
Assuming that e1 and e2 are hidden actions seems equivalent to assuming that there is no
documentation of work in the working papers.

17
th

Mark Penno, 20 Annual Audit Symposium

Negotiation: The authors view the pair of (Q, r) as representing the give-and-take between the
company and its auditor. However, there doesnt seem to be any give-and-take in the model, in that
the client has all of the bargaining power.

18
th

Mark Penno, 20 Annual Audit Symposium

Loss Function: The authors assume the an audit firms loss to be r x , where represents
2

the firms share of the overall loss, r x . If we consider De Angelos (1981) notion of
2

collateral, however, the loss would be better represented by r x , where is increasing in

the size of the audit firms portfolio of clients. This embellishment alone mitigates against the hiring
smaller audit firms.
2

19
th

Mark Penno, 20 Annual Audit Symposium

Audit Evidence: The authors justify disclosing y Ex | y by arguing that the assumption is justified
by the observation that the audit evidence documented in the audit firms working papers is the only
admissible evidence in court. Normally we must deflate the absolute value of y appropriately when
interpreting them. It is therefore not clear why the raw y would be so privileged. In contrast, the
authors assume that when the case is brought to trial, the courts are not influenced by the raw data,
because if they were, they might (ex post) accept a report of r y as non-negligent, and (in the
2
limit) ( y x) 0 for all y, as long as y is documented in the audit. This is confusing.

20
th

Mark Penno, 20 Annual Audit Symposium

Benefit Function: The market (pricing) function M r r appears to be exogenously imposed.

For example, if e 0 , then the report would be uninformative, and the function would be flat. Also,
*

it would appear r , should be endogenous. An exogenous function requires more explanation,

because the capital markets do not appear to play a role.

21
th

Mark Penno, 20 Annual Audit Symposium

Rationale For Cost Function: The authors note that they make standard assumptions about the
auditors cost function. To the extent that (as the authors argue) none of the previous papers have
dealt with joint audits, are the standard assumptions appropriate here (e.g., extra coordination
costs)?

22
th

Mark Penno, 20 Annual Audit Symposium

The Current Status of Joint Audits:

In Denmark in 2009 (four years after joint audit was dropped), only 19 companies out of 182
had more than one auditor (Le Vourch and Morand, 2011).
The Proposal for a Directive of the European Parliament and of the Council amending
Directive 2006/43/EC on statutory audits of annual accounts and consolidated accounts
issued by the EU in November 2011 did not propose joint audits (dropping the suggestion
made in the Green paper, 2010).
Numerous respondents to the Green Paper (Audit Policy: Lessons from the Crisis) issued in
October 2010, suggested there is little evidence that joint audits add to audit quality but they
would necessarily increase audit fees (Andr at al., 2012).
In a recent survey (PwC) that polled 120 large businesses (excluding France) across Europe
(European Audit Committee Chair and CFO Poll, YouGovStone, June 2011), 99% of
interviewees declared being against mandatory joint audit (page 6).2
Haapamki et al. (2012) note that prior studies tend to provide evidence that joint audits are
associated with higher fees. Andr at al. (2012) provide evidence that audit fees paid by listed
French companies (where a joint audit is mandatory) are significantly and economically higher
than those paid in the UK and Italy.]
2

http://www.pwc.com/en_GX/gx/audit-services/publications/assets/european-audit-committee-chair-and-cfo-poll.pdf
23
th

Mark Penno, 20 Annual Audit Symposium

References
Andr, Paul, Graldine Broye, Christopher Pong and Alain Schatt, Do joint audits lead to greater
audit fees?, working paper, 2012, ESSEC Business School Paris. SSRN
De Angelo, Linda, Auditor Size and Audit Quality, Journal Of Accounting and Economics, (1981)
Vol. 3, pp. 183-200.
EC (2010) Green Paper: Audit policy: Lessons from the crisis, 13th of October 2010, European
Commission, Brussels, pp. 1-21. Accessed 10/9/12 at http://eurlex.europa.eu/LexUriServ/LexUriServ.do?uri=COM:2010:0561:FIN:EN:PDF

EC (2011) Summary of responses Green Paper - audit policy: Lessons from the crisis, 4th February
2011, European Commission, Brussels, pp. 1-36. Accessed 10/9/12 at
http://ec.europa.eu/internal_market/consultations/docs/2010/audit/summary_responses_en.pdf
Francis, Jere, Chrystelle Richard, and Ann Vanstraelen, Assessing Frances Joint Audit requirement:
Are Two Heads better than One?, Auditing: A Journal of Practice and theory, Vol 28, (2009) pp. 3563.
Haapamki, Elina, Tuukka Jrvinen, Lasse Niemi, Mikko Zerni, Do Joint Audits Offer Value for
Money? Abnormal Accruals, Earnings Conservatism, and Auditor Remuneration in a Setting of
Voluntary Joint Audits (2012), European Accounting Review, (2012) forthcoming
24
th

Mark Penno, 20 Annual Audit Symposium

Holmstrom, Bengt, Moral Hazard in Teams, The Bell Journal of Economics, Vol. 13, No. 2
(Autumn, 1982), pp. 324-340
Lesage, Cedric, Nicole Ratzinger-Sakel, Jaana Kettunen, Is joint audit bad or good? Efficiency
perspective evidence from three European countries, accessed (10/9/12)
http://aaahq.org/AM2012/display.cfm?Filename=SubID%5F1711%2Epdf&MIMEType=application
%2Fpdf
Le Vourch, Jolle, and Pascal Morand, Study on the effects of the implementation of the acquis
on statutory audits of annual and consolidated accounts including the consequences on the audit
market, ESCP Europe (2011). Found 10/8/12 at
http://ec.europa.eu/internal_market/auditing/docs/studies/201111-study_en.pdf
Narayanan, V.G. An Analysis of Auditor Liability Rules. Journal of Accounting Research 32(1994,
Supplement): 39-59.
Piot, Charles, and Rmi Janin, External auditors, audit committees and earnings management in
France. European Accounting Review Vol. 16 (2007), pp. 429454.
Taylor, Mark and Daniel Simon. (1999) Determinants of audit fees: the importance of litigation,
disclosure and regulatory burdens in audit engagements in 20 countries, International Journal of
Auditing, 34 (3), pp. 375-388.

25
th

Mark Penno, 20 Annual Audit Symposium