Watts Zimmerman Agency Problems Theory of Firms

The Booth School of Business, University of Chicago
Agency Problems, Auditing, and the Theory of the Firm: Some Evidence
Author(s): Ross L. Watts and Jerold L. Zimmerman
Source: The Journal of Law & Economics, Vol. 26, No. 3 (Oct., 1983), pp. 613-633
Published by: The University of Chicago Press for The Booth School of Business,
University of Chicago and The University of Chicago Law School
Stable URL: https://www.jstor.org/stable/725039
Accessed: 11-01-2020 14:07 UTC
REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/725039?seq=1&cid=pdf-reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
The Booth School of Business, University of Chicago, The University of Chicago Law
School, The University of Chicago Press are collaborating with JSTOR to digitize, preserve
and extend access to The Journal of Law & Economics
This content downloaded from 183.171.185.77 on Sat, 11 Jan 2020 14:07:46 UTC
All use subject to https://about.jstor.org/terms
AGENCY PROBLEMS, AUDITING, AND
THE THEORY OF THE FIRM: SOME
EVIDENCE*
ROSS L. WATTS and JEROLD L. ZIMMERMAN

University of Rochester
I. INTRODUCTION
Recent developments in the theory of the firm emphasize the impor-

tance of monitoring the performance of parties to the firm.' Jensen an
Meckling2 hypothesize that an audit is one type of monitoring activity that
increases the value of the firm. An audit by someone independent of the
manager reduces the incentive problems that arise when the firm manage
does not own all the residual claims on the firm. If Jensen and Meckling'
hypothesis is correct, independent audits are expected in the earliest firm
where the manager did not supply all the capital.
On the other hand, a leading auditing text suggests that the appearance
of independent audits is more recent and is the product of governmen
fiat: "One of the earliest steps in recognition of the need for audits oc
curred in England with the passage of the Registered Companies Act of
1862. The Act required that the financial statements of joint stock com-
panies be audited by a person independent of management, and thereby
greatly enhanced the status of professional auditors as well as the growth
of that profession."3 This opinion, that independent audits arose because
* Associate Professors. The authors gratefully acknowledge the comments on earlier

versions of this manuscript by Eugene Fama, Nicholas Gonedes, Robert Holthausen,
Michael Jensen, Richard Leftwich, Clifford Smith, Lee Wakeman, and Jerold Warner.
1 Armen A. Alchian & Harold Demsetz, Production, Information Costs, and Economi
Organization, 62 Amer. Econ. Rev. 777 (1972); Michael C. Jensen & William H. Meckling
Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure, 3 J
Financial Econ. 305 (1976); Eugene F. Fama, Agency Problems and the Theory of the Firm
88 J. Pol. Econ. 288 (1980); Eugene F. Fama & Michael C. Jensen, Separation of Ownership
and Control, 26 J. Law & Econ. 301 (1983); Eugene F. Fama & Michael C. Jensen, Agenc
Problems and Residual Claims, 26 J. Law & Econ. 327 (1983).
2 Jensen & Meckling, supra note 1, at 338-39.
3 Howard F. Stettler, Auditing Principles 20-21 (4th ed. 1977). Note that the statement
[Journal of Law & Economics, vol. XXVI (October, 1983)]

? 1983 by The University of Chicago. All rights reserved. 0022-2186/83/2603-0005$01.50
613
614 THE JOURNAL OF LAW AND ECONOMICS
they were specifically required by government regulation, is

recent congressional staff report that claims that the U.S. sec
"created a need for . . . independent auditors."4
To provide evidence on the importance of the independent a
firm, this paper investigates the history of the business corpo
organization and use of audits by English and U.S. business c
are examined, from the days of English merchant guilds to the tim
were required by law. We find that the audit existed early in t
ment of business corporations (1200) and evolved gradually int
of audit required by the first English companies act (1844). F
appears that audits were widespread among early business co
This evidence suggests that monitoring of performance is importa
crucial, to the formation of firms. The long survival of auditing s
is a part of the efficient technology for organizing firms.5
The audits of the early corporations were conducted by dir
shareholders. The use of outside professional auditors did not
common until the latter half of the nineteenth century in the Un
dom and the early part of this century in the United States.
when the use of professionals became common, corporate audi
generally required by law in either country. This suggests
professional auditors was due to changes in the market for au
The papers that stress the importance of monitoring in the theo
firm analyze the incentive (agency or control) problems that e
ferent types of firms (partnerships, nonprofit firms, corpora
Firms are sets of contracts among the factors of production,6 and
sets of contractual arrangements (for example, alternative pro
structures) provide different incentives for opportunistic beha
contracting parties. This opportunistic behavior reduces the to
of the firm and hence its value. Jensen and Meckling point
markets characterized by rational expectations the cost of the
tic behavior is borne by the offending party. This provides in
that party to write contracts which restrict his opportunistic
Corporate managers or promoters write contracts (by-laws
indenture agreements, compensation plans) that reduce their
that the 1862 Companies Act required independent auditing is false. That ac
quired audits nor specified that the auditors be independent. That act contain
model set of articles that provided for audits but did not specify they be by
4 Staff of Senate Subcomm. on Reports, Accounting, and Management of th
on Government Operations, 94th Cong., 2d Sess., The Accounting Establishm
Study 1 (Comm. Print 1976).
5 Armen A. Alchian, Uncertainty, Evolution, and Economic Theory, 58 J. P
(1950).
6 Jensen & Meckling, supra note 1.
AGENCY PROBLEMS 615
ism. Enforcement of the contract requires mo

activities and it is hypothesized by Jensen and
of auditing. Another enforcement mechanism
Meckling is the posting of bonds by managers
below reveals that, like auditing, bonding has
history of the business corporation.
An audit will be successful in changing expec
ing the opportunistic behavior costs (agency co
only if it is expected that the auditor will repor
of contract. The probability that the auditors
breach is effectively the auditing profession's d
So the preceding condition can be rephrased to
reduce the agency costs borne by the manager o
the auditor to have a nonzero level of independe
early business corporations' auditors to be ind
this paper. Further, we analyze the changes in
and in the auditor's incentive to be independen
tion evolved.
The evidence analyzed in this paper is qualitative in nature and does not
lend itself to hypothesis testing using large samples and econometric
methods. Random samples of company charters, by-laws, or minutes
cannot be drawn since what exist are chiefly the records of successful
companies. Although audits existed in every firm whose original docu-
ments we examined (the English Merchant Adventurers, regulated com-
panies, and joint stock companies from the fifteenth to the seventeenth
centuries), there may have been other companies that did not have audits
and whose records were not preserved. In spite of this ex post selection
bias, we believe that this historical evidence provides interesting insights
on the theory of the firm and on the origins of the independent audit.
Section II presents evidence that audits were common in early English
business corporations, the medieval craft and merchant guilds, and regu-
lated and early joint stock companies. It presents evidence also that the
auditors had incentives to be independent. This evidence is consistent
with the hypothesis that efficient, agency-cost-minimizing contractual ar-
rangements (for example, the organizational forms) include monitoririg
and is inconsistent with the explanation that independent corporate audit-
ing is the consequence of companies or securities acts. Section II de-
scribes also how the audit evolved in response to changes in organiza-
tional forms. Early audits were performed by committees of owners
7 Ross L. Watts & Jerold L. Zimmerman, Auditor Independence and Scope of Services
(July 1982) (unpublished manuscript, Univ. Rochester, Grad. School Management).
(members). The composition and size of these committees changed

organization's size and structure changed.
Section III describes the development of the independent profes
audit firm and its substitution for the internal shareholder audit c
tee. In addition the section explores explanations for that substit
Section IV summarizes the conclusions.
II. AUDITING IN THE EARLY CORPORATION
In this section we provide a description of each type of business corpo-

ration to demonstrate that incentive problems existed in those corpora-
tions. Then we give the evidence that auditing and bonding existed for the
type of corporation along with the evidence that the audits were designed
to reduce incentive problems. Finally, we discuss the incentives of the
auditors to be independent.
A. English Merchant Guilds
1. Description. Merchant guilds appeared in England shortly after the

Norman Conquest (A.D. 1066).8 The guild arose to protect the prosperity
of the merchants after the conquest by forming cartels to monopolize
trade.9 They did not exist before the conquest because there was very
limited trade.10 A guild was chartered by the crown and given a monopoly
to trade within its own particular town. Each guild member traded on his
own account or in partnership with other guild members. Members of a
guild were not allowed to enter into partnership with nonmembers. Fur-
ther, a guildman had to share any purchase with other guild members who
wished to participate in the venture, at the same price."
The guilds are among the earliest examples of incorporation. Most
guilds held property and incorporation reduced the transactions costs of
8 1 Charles Gross, The Gild Merchant 4 (1890); 1 W. Ashley, An Introduction to English

History and Theory 71 (1923); 1 William Robert Scott, The Constitution and Finance of
English, Scottish and Irish Joint-Stock Companies to 1720, at 7 (1912).
9 Robert E. Ekelund & Robert D. Tollison, Mercantilist Origins of the Corporation, 11
Bell J. Econ. 715 (1980). Besides functioning as a cartel, guilds performed other functions
including social, religious, and municipal government. Although these other functions for
some guilds might be more important than the cartel-enforcing, mercantile functions, it is
these latter functions that are the focus of the subsequent analysis. Carlo M. Cipolla, Before
the Industrial Revolution: European Society and Economy, 1000-1700 (2d ed. 1980).
1O Gross, supra note 8, at 3-4.
" In the twelfth to fourteenth centuries, the merchant guilds in the larger towns started to
specialize by craft or ware. The organization of these craft guilds (sometimes called mercan-
tile societies or companies of merchants) is similar to that of the merchant guilds. Gross,
supra note 8, at 106-57; and Scott, supra note 8, at 8.
AGENCY PROBLEMS 617
transferring the property.12 If the guild it

there were substantial costs in changing
and new members were admitted. Hence
tion itself was "to continue indefinitely
devised."'3 The idea that the whole body
separate from that of a member was exp
common seal to be used as evidence that
ing.14 Guilds were expressly incorporated
II (1367-1400)."5
The members of the guild provided r
officers of the guild.16 Contractual arra
and by-laws) existed to define and restri
parties, including the officers. The major
man or master and his associates, called
The number of stewards generally varied
places the stewards fulfilled the alder
stewards administered the revenues from
tolls. Many guilds had lands and teneme
often traded on behalf of the members and conflicts of interests between
manager and owners were present. Given that, it is not surprising the
manager's acts were restricted by the contracts. The alderman and stew-
ards had specific duties and roles to fulfill.19 Further, given the restrictions
on the guild officers' actions, we would expect to observe that the
officers' actions would be monitored.
2. The Existence of Auditing and Bonding. The constitution of the
merchant guild at Ipswich in 1200 requires that the alderman "on oath
shall make due return, annually before the bailiffs and coroners (of the
town) of all profits arising during the year" from the guild's monopoly
trading of stone and marble.20 The merchant guild at Bury St. Edmunds
had, by 1304, provision for an annual audit.21
12 Gross, supra note 8, at 99.

13 Scott, supra note 8, at 3.
14 Id.

16 Although it is difficult to assess the typical proportion of a member's wealth adminis-
tered by the guild officials, some guilds appear to have owned substantial property. Scott,
supra note 8, at 6; and Gross, supra note 8, at 14, 151, 159-60.
17 Gross, supra note 8, at 26-27.
18 Id. at 28.
19 Id. at 23-35.
20 Id. at 25.
21 This reference is to the original medieval Latin text of documents supplied by 2 Gross,
Several craft guilds and companies of merchants were audi

by committees of members.22 The records of the Worshipfu
Grocers of the City of London and the Worshipful Company
of the City of London indicate, in 1346 and 1546, respectiv
accounts of those two companies were audited annually. Sim
accounts of the retiring Warden of the Worshipful Compan
ters were audited each year from the fifteenth century on by a
of past wardens and other members and based on their rep
ance, if any, of his bond23 returned to the retiring Warden.24
suggests that the audits were not superficial and were not m
of cash or assets on hand. Expenditures were examined in d
cites several examples of the auditors refusing to certify an
various charges.
The audit of the guild appears designed to monitor the ma
tracts. It came at the end of the managers' tenure and was
check for unauthorized expenditures. Further, it also appea
been designed to check for other breaches of contracts. In
case, the auditors of a craft guild explicitly fined the master
for breaching certain ordinances (that is, breaching their con
prevalence of guild audits26 and bonding arrangements is co
the existence of conflicts of interest creating incentives for
monitoring technologies to be devised.
Guild audit committees usually consisted of four guild m
occasionally public officials.27 Although these auditors did n
in auditing as a full time career, they still had the responsibility
certain managerial functions and to check for breaches of c
3. The Auditors' Incentives to Be Independent. An audi
heavily fined for not completing the audit in due time.28 In add
supra note 8, at 34, to support his statements in vol. 1. We obtained the de

by having the Latin translated.
22 Edward Boyd, History of Auditing, in History of Accounting and Acc
(Richard Brown ed. 1905).
23 Bonding is hypothesized by Jensen & Meckling, supra note 1, at 325
reduce the costs that arise from the conflicting of interests.
24 Boyd, supra note 22, at 86-88.
25 Id. at 81.
26 The development of the incorporated borough is closely related to the
the merchant guild; the officers of the borough tend to be drawn from guil
Gross, supra note 8). Observing borough audits similar to those of the gu
supra note 22) suggests the type of audit described above was a widespread
the fourteenth, fifteenth, and sixteenth centuries.
27 Initially, the auditors were compensated in goods (for example, ale) a
paid. In the Spanish Company (a regulated company circa 1600) the eight au
twenty shillings. Pauline Croft, The Spanish Company 79-80 (1973).
28 Boyd, supra note 22, at 81.
AGENCY PROBLEMS 619
cost, lack of performance and independenc

tion and, in the extreme, caused loss of h
share of the guild's monopoly profits.
auditors had to own property.29 Presuma
easier to recover damages against them if
thereby providing the auditors with furth
if one occurred.
The use of an audit committee rather than a single auditor also en-
couraged auditor performance and independence by making collusion of
the manager and auditor more difficult. It is more costly to bribe an entire
committee than a single individual.
B. Merchant Adventurers/Regulated Companies
1. Description. Prior to the latter half of the thirteenth century the

English export trade was generally conducted by the German Hanse Mer-
chants.30 However, from that time until the seventeenth century the ex-
port of English raw products was conducted through English and foreign
merchants who were given a monopoly over the export of raw products
(principally wool). Certain towns were specified for the shipping and sales
of those products, companies of merchants were eventually formed, and
the system facilitated the collection of the royal customs.31 Each member
of the company provided his own capital (inventories and ships) and
traded on personal account (or in partnership).
In the fourteenth century, under Edward III's auspices, the cloth indus-
try grew in England and manufactured goods, primarily cloth, began to be
exported. In the late fourteenth and early fifteenth century, English mer-
chants involved in this trade applied for charters for organizations to
monopolize the trade. In 1391, 1407, and 1408 charters were granted to
English merchants trading with Prussia, the Netherlands, and Scan-
dinavia, respectively. The latter two companies became known as the
Merchant Adventurers of England and the Eastland Company.32 In the
sixteenth century, companies of merchants trading with other countries
were incorporated, including the Spanish Company (1577), the Turkey
Company (1581), the Venetian Company (1583), and the Levant Com-
pany, a union of the Turkish and Venetian Companies (1592).33

30 Id. at 140.
31 Id., at 144.
32 Scott, supra note 8, at 8-9.
33 In exchange for granting and enforcing the cartel, the Crown was often made a partner
or given low interest "loan," Ekelund & Tollison, supra note 9. The regulated companies
passed by-laws governing commerce with the foreign countries. Their rules were specific
There is evidence that some of the members of the earl

companies were also members of companies of merchants
the minutes of the Merchant Adventurers were kept in th
those of the Mercers' Company of London and the Mer
Adventurers' headquarters until 1666.34 These facts sugges
merchants who began the Merchant Adventurers came from
Company. Given this relationship, it is not surprising tha
organization of the regulated company is similar to that of
merchant guilds.
By the fifteenth century some guilds had developed a sys
istration consisting of a governor and a number of associ
multiple of twelve. For example, the alderman and four as
Ipswich guild were replaced by a governor and twenty-fou
This change is reflected also in the organization of the Mer
turers of England which in its 1505 charter called for th
governor and twenty-four associates.36
2. The Existence of Auditing and Bonding. Most of t
companies appear to have been audited.37 Before 1632 the
pany elected auditors as required (for example, when accou
sented). After 1632 auditors were elected annually.38 The
of the Spanish Company includes an entry39 that at the m
general court (shareholders' meeting) on September 7, 1604
"brought in his account" and eight auditors were appointe
it.40 Six were assistants (directors) and two were ordinary
Eastland Company also had auditors. A letter from the Com
don to the Company at York in 1674 refers to the au
dissatisfied because the details were not supplied for cert
and governed details of family and social life, Scott, supra note 8, at 10. Br
could lead to a member's losing his right to trade with the foreign count
36 Id. at 9.
37 References to six English regulated companies, two of which later mer
Of the remaining five companies, four were definitely audited. Histori
company trading with Prussia were sketchy and contained no mention o
38 Mortimer Epstein, The Early History of the Levant Company 68 (1
39 Croft, supra note 27, at 11 and 79.
40 Apparently, accounts were also presented annually to the company bef
Spain. Croft, supra note 27, at xxix, notes that neither the last treasure
George Hanger, nor his immediate predecessor, Sir John Watts, bother
counts (because the company broke up due to the war). The accounts pre
ber 7, 1604, were Hanger's apparently before the war, because the register
a resolution warning Watts to present his account with all convenient sp
41 Maud Sellers, The Acts and Ordinances of the Eastland Company 97
AGENCY PROBLEMS 621
The codification of the ordinances and

ers of Bristol in 1618 includes a requir
statement of receipts and payments at
his accompt before Auditors assigned
As in the guilds, the monitoring me
mented by the bonding of officials. T
venturers of Bristol,43 the Spanish C
and the Eastland Company46 were requ
bonds were substantial (the Levant Com
the Spanish Company's five hundre
pany's one thousand marks sterling; t
specify an amount). In addition, offici
in foreign towns were required to pos
The form of the audit of the regulated
merchant and craft guilds and compan
members was elected to audit the treasurer's accounts. One difference
was the increase in the size of the committee of auditors, which was
probably due to the larger scale of the regulated companies. Not only the
general accounts were audited, but also the accounts of the company in
other towns.
3. The Auditor's Incentives to Be Independent. As in merchant and
craft guilds, the auditors could be fined for refusing to act,48 and the
committee form made collusion with the manager more difficult.
It appears from the existence and the similarity of the requirements
across all the companies whose records we can obtain that auditing of
regulated companies was the usual if not the universal practice. Further,
the similarities of audit arrangements (for example, the use of bonding of
the officers and the audit to determine how much of the bond to return)
and the commonality of membership with the guilds suggests the audit
was adopted from the guilds. This in turn suggests that the same mecha-
nisms provided the auditor with incentives to be independent (for ex-
ample, the use of committees and penalties-including loss of reputa-
tion).
42 John Latimer, The History of the Society of Merchant Venturers of the City of Bristol
69 (1903).
43 Id.
44 Croft, supra note 27, at 79.
45 Epstein, supra note 38, at 73.
46 Sellers, supra note 41, at 24.
47 Id. at 25.
48 Id. at xxiii.
C. Joint Stock Companies

1. Description. In 1553, one and a half centuries after the appearance
of the first regulated company, another form of corporate firm appeared in
England, the joint stock company. The first joint stock companies, the
Russia Company and the African Adventurers, were formed for overseas
trade just as the regulated companies were. However, a major difference
between the two was the method of financing the trade. In the regulated
company each member supplied his own capital and traded on his own
account or in partnership, using his own ships.49 In the joint stock com-
pany the officers of the company traded on behalf of all the members or
shareholders. Initially, capital was raised to finance each separate voyage
and the proceeds were distributed after the voyage was completed. Sev-
eral hypotheses can explain this changed method of financing, which
occurred while regulated companies continued to be formed after the
emergence of the joint stock company (for example, the Turkey Company
in 1581, the Venice Company in 1583, and the Levant Company in 1592).50
One hypothesis is that the scale of the joint stock company voyages,
per se, explains the change. Either the voyages required too large an
investment for a single merchant (or a partnership of a few merchants) or
the scale of the voyages of regulated companies was too small to allow
separate specializations in management and risk bearing.5' However, it is
unlikely that this hypothesis alone can explain the emergence of the joint
stock company. The first two joint stock companies were the Russia
Company and the African Adventurers, both of which had their first
voyages in 1553.52 It is unlikely that the value of the cargoes of each
voyage of the two firms exceeded 5,000 pounds.53 At the same time, the
regulated company, the Merchant Adventurers, shipped 60,000 pounds a
year in woolen goods.54 It would seem a partnership of Merchant Adven-
turers could have financed a 5,000-pound voyage.
Another hypothesis is that the magnitudes of the voyages combined
49 Thomas S. Willan, The Early History of the Russia Company 1553-1603, at 19 (1956).
50 Epstein, supra note 38.
51 George J. Stigler, The Division of Labor is Limited by the Extent of the Market, 59 J.
Pol. Econ. 185 (1951), discusses the effect of the extent of the market on the degree of
specialization.
52 Scott, supra note 8, at 17-22. The original names of the companies were (for the Russia
company) "the Merchants Adventurers of England for the discovery of lands, territories,
isles, dominions and seignories unknown, and not before that late adventure of enterprise
commonly frequented" and (for the African Adventurers) "the Adventurers to Guinie."
53 Willan, supra note 49, at 22.
54 Id.
AGENCY PROBLEMS 623
with the greater risk of the voyages u

panies led to risk sharing of the claim
regulated companies evolved to contro
with relatively familiar countries. The
of the trade, and so on, were known. O
of the Russia Company sought "to dis
Indies."56 The route and the nature of
unknown. Similarly, the African Adv
with Guinea-a risky undertaking. The
tempt to stop such trade and the trade
Portfolio theory suggests the efficien
investors being well diversified. Thus,
the voyages of the Russia Company and
spread across many investors.58
The majority of members of the Ru
chants already engaged in foreign trad
ers).59 Hence, it is not surprising tha
Company was hard to distinguish from t
joint stock character of the company w
charter.61 Like a regulated company o
Company had a governor and twenty-
The African Adventurers operated wi
chief adventurers who in turn had par
made calls on its shares to finance eac
proceeds in accordance with the sha
sought a charter because it wished to
difficulties with the Portuguese.63
55 Fama, supra note 1.

58 Fama, supra note 1.
59 See Willan, supra note 49, at 21. This evid
advanced by Ekelund & Tollison, supra note 9,
as the efficient method whereby owners of the
valued users-the most efficient managers. Whil
the regulated companies, involving the same me
adopt the more efficient joint stock organizatio
observed at the same time, in the same country
62 Id. at 21, 30.
63 Id. at 21.
1566,64 and it was not until 1588 that another company w

trade with Africa.65
2. The Existence of Auditing and Bonding. The first join
panies were audited. Sources external to the company indi
early as the 1580s, and afterwards, the Russia Company
annually.66 The Court of Minutes of the East India Company
years of its existence (September 1599-August 1605) includ
for the appointment of auditors every time accounts were pres
general court (that is, the shareholders' meeting). The first
is dated December 31, 1600, and all four auditors appointe
tors.67
The accounts of the early joint stock companies were audited by a
committee of shareholders (members) and/or directors. This practice con-
tinued into the eighteenth and nineteenth centuries. Some companies had
provision for the annual appointment of a committee of shareholders to
inspect the accounts.68 Forrester69 describes the accounting and control
system of a canal company from 1768 to 1816 and that description in-
cludes sureties or guarantees provided by officers combined with an audit
at the end of an officer's term of employment. Ma and Morris examined
the records of Australian and British joint stock banks prior to the 1844
Companies Act and found that those banks "had their accounts audited
by the directors or provided for auditors to be appointed from the general
body of shareholders."70
U.S. corporations also used committees of auditors.71 The Bank of
Commerce in New York in the annual report dated May 1850 reports:
64 Id. at 34.
65 2 Scott, supra note 8, at 10.
6 See Willan, supra note 49, at 23. The history of the Russia Company before 1666 has to
be traced from sources outside the company, since like many of the other companies the
Russia Company's own records were destroyed in the great fire of London. Consequently,
the statement above is based on outside sources and it is possible that the accounts were
audited before 1580.
67 Henry Stevens, The Dawn of British Trade to the East Indies 107 (1967). In 1621 the
East India Company appointed two paid officials (later one) with the title "Auditor." Sir
William Foster, The East India House 9 (1924). This practice appears to be unusual.
68 Armand B. DuBois, The English Business Company after the Bubble Act 1720-1800, at
300 (1938).
69 D. A. R. Forrester, Early Canal Company Accounts: Financial and Accounting As-
pects of the Forth and Clyde Navigation, 1768-1816, 10 Accounting & Bus. Res. 109 (1980).
70 Ronald Ma & Richard D. Morris, Disclosure Practices of British and Australian Banks
in the Nineteenth Century 26 (July 1980) (unpublished paper, Univ. New South Wales).
71 See Gary J. Previts & Barbara D. Merino, A History of Accounting in America 3-4
(1979), who report that the Massachusetts Bay Company in 1629 used audit committees and
that this practice continued through the 1870s.
AGENCY PROBLEMS 625
"The usual quarterly examinations, wit

have been made during the year by Co
tors." The 1781 Charter of the Bank of
every quarterly meeting of the board t
the books every day. The Philadelphia N
appointed a committee of directors to e
"audit the bank's assets monthly."72 Al
Pennsylvania issued annual reports from
the accounts were examined by a comm
3. The Auditors' Incentives to Be Indep
served that provide auditors with incen
the early joint stock companies and reg
reputation affected the probability that
auditor or even admitted to the compan
to be on a one-man-one-vote basis.73 Ye
bers of the committees or courts of dir
panies.74 Further, merchants frequently
of several companies in that period.75
The survival of the committee of auditor
suggests it was an efficient monitoring
that the method was inefficient on the basis of the frauds which occurred
in the seventeenth to nineteenth centuries and that government regulation
was necessary for effective control.76 However, they only consider the
benefits of removing observed abuses. Ignored are the costs of regulation
to remove these abuses. They fail to consider the number of firms in
which the contracting and monitoring system between managers and
shareholders worked. Scott makes this point in replying to Adam Smith's
criticisms of joint-stock companies:
In fact, while the methods of control and of internal organization were far from
perfect, they were much better than might have been expected, considering the
72 Nicholas B. Wainwright, History of the Philadelphia National Bank 20 (1953).

73 See Croft, supra note 27, at 78.
74 Of the eighteen directors named in the 1605 Charter of the Levant Company, at least
nine also served on the committees of the East India Company and at least three also served
as directors of the Spanish Company. Epstein, supra note 38, at 165; Croft, supra note 27, at
3, 4, and 247; and Stevens supra note 67, at 6, 7, 12, 13, 121, 179, and 237.
75 For example, William Harryson served on audit committees of the East India Company
in 1600 and 1601 while he was a director of that company, Stevens, supra note 67, at 107,
156, and 166, and on an audit committee of the Spanish Company in 1605 while an ordinary
member. Croft, supra note 27, at 37. Nicholas Lyng and Richard Wyche also served on audit
committees of both the East India Company and the Spanish Company in 1601 and 1605,
respectively. Id., and Stevens, supra note 67, at 156.
76 Edwin M. Dodd, American Business Corporations until 1860 (1954), at 291-307.
times and how undeveloped the joint stock system was in the seven
tury. Despite some instances of fraud, carelessness and profligacy o
agents abroad, numerous instances can be quoted of a remarkable
duty, while amongst the directors or assistants there was a large-h
terestedness, united to a careful supervision of business, which is
mendable. It is noteworthy that out of the great number of com
officers have been investigated in this work, the allegations offraud
ment are comparatively rare.77
D. Observations on the Development of Auditing

The evidence suggests the practice of having a committee
was not imposed on the merchant guilds, regulated compani
stock companies by law. The auditing was voluntary. Typica
were no references to auditing in the charters of the guilds
companies, and joint-stock companies examined. The auditin
order of the general court (meeting of members of sharehold
court of assistants (directors' meeting). Thus, when the U.K
Act of 1844 required directors to keep accounts and require
counts to be audited by persons other than the directors (or their
Parliament was merely incorporating into the law a version o
that had existed for six hundred years.
The hypothesis advanced here is that committees of audito
because they were an efficient method of monitoring contra
managers and those supplying capital. However, auditing pr
not constant over time, it changed as the business corporatio
Two dimensions in which auditing changed are the compositi
tive size of audit committees. For example, the committee of
the regulated and first joint-stock companies (for example,
Company and the East India Company) included assistants (th
tors). By 1844 this was a committee of shareholders. The inc
assistants on the early audit committees is, at first glance, p
sumably, the assistants were managers whose actions the audi
monitor.
The answer appears to be that assistants were not exactly eq
the directors of a modern corporation. In both the Spanish
1605 and the East India Company in 1600 the ratio of director
to shareholders (ordinary members) was very large by today'
The Spanish Company had sixty-one assistants out of a total m
of 310.79 The East India Company had twenty-four members
See Scott, supra note 8, at 451-52. Emphasis added.

78 Ananias C. Littleton, Accounting Evolution to 1900, at 289 (1933).
79 See Croft, supra note 27, at xxxvi. The extra assistant beyond the mult
was the secretary.
AGENCY PROBLEMS 627
eral committee (assistants) out of a total

ratios of assistants to members (one to f
were likely to be representative of the
number of assistants would make it more difficult for the assistants to
collude against the membership. Hence, explicit monitoring of the assis-
tants in general would not be efficient. On the other hand, officials with
direct personal control over resources (for example, the treasurer) would
in the absence of monitoring, be able to convert resources to their own
personal use. Consequently, it is not surprising that the treasurer was
audited by assistants and that assistants were not audited by members, in
general.
If the inclusion of assistants on the audit committees of the early joint-
stock companies can be explained by the high ratio of assistants or direc-
tors to shareholders, the movement to a shareholder committee can be
explained by the large drop in that ratio. In the eighteenth and nineteenth
centuries, despite the Bubble Act of 1720, use of the joint stock form of
organization grew, and the number of directors for each company and the
ratio of directors (assistants) to shareholders (members) dropped substan-
tially, particularly in the early eighteenth century.81 This suggests that it
would be less costly for the directors to collude against the shareholders.
Consequently, there was the increased tendency to use committees of
shareholders, not directors, to audit the accounts.
While the evidence suggests that from the time of the guilds to the mid-
nineteenth century auditing was used as a monitoring device for contracts
between managers and equity holders, we found no direct evidence that
audits were used to monitor debt contracts in that period. The secondary
sources examined contained no references to the auditors' reporting to
debt holders or that the debt holders even received the audit reports. We
would not expect such evidence for the guilds, because before the Refor-
mation loans were discouraged by the church.82 However, in the eigh-
80 Stevens, supra note 67, at 58-63.

81 Two observations support this contention. First, the increase in the capitalization of
joint stock companies suggests a greater number of shareholders. In 1560 the capital of the
only two joint stock companies in existence was less than ?10,000 or .013 percent of the
national wealth of the United Kingdom. Scott, supra note 8, at 439. By 1695 the capital of
joint stock companies was 1.3 percent of national wealth and by 1720 it was 13 percent.
Scott, supra note 8, at 439. Also, the membership of the regulated companies and the large
majority of the members of the first joint-stock companies were merchants. However, in the
seventeenth and eighteenth centuries nonmerchants became the suppliers of capital to the
joint-stock companies, Scott, supra note 8, at 440-43, suggesting an increase in the number
of shareholders. Second, the average number of directors of a company appears to have
dropped during this period. Dodd, supra note 76, at 291, describes the directors in the
eighteenth century as a "small group."
teenth century companies issued numerous bonds.83 Compa

under which the bonds were issued used accounting number
ample, profits) to restrict dividends84 and one would expec
straints to be monitored. There is some indirect evidence to
auditing was used to monitor debt contracts. It is possible t
were available to bondholders and the audit by the committ
holders served to control the shareholder-bondholder conflict of interest.
The proposition that the accounts were available to debt holders is
strengthened by the inclusion in the 1862 U.K. Companies Act of a re-
quirement that a statement of mortgages be available to creditors.85
III. THE DEVELOPMENT OF THE PROFESSIONAL AUDIT FIRM
In this section we provide brief descriptions, of the development of the

independent professional audit firm for the United Kingdom and the
United States. Then explanations for the firms' development are inves-
tigated.
A. Development in the United Kingdom
The U.K. company acts from 1844 to 1900 did not require outside
auditors. The 1844-45 acts required the directors to keep accounts and
required those accounts to be audited by persons other than the directors
or their clerks.86 Further, the auditors were required to be shareholders.
The 1856 act dropped the compulsory audit requirement.87 The 1862 act
included an optional model set of articles that provided for audits. While
the provision did not require the auditor to be a shareholder, it also did
not require the auditor be a professional firm. A series of miscellaneous
acts (Railway Companies Act, 1867-68; Banking Companies, 1879; Water
Companies, 1871) required audits, but again not by outsiders.88 The 1900
Companies Act reestablished compulsory audits. However, by this time
"the accounts of most of them (public companies) were not only audited
but were in fact audited by chartered accountants. Indeed, practice has
generally outrun legal minima."89 Chartered accountants are professional
accountants accredited by professional accounting societies.
83 Dodd, supra note 76, at 112.

84 Ross L. Watts & Jerold L. Zimmerman, The Demand for and Supply of Accounting
Theories: The Market for Excuses, 54 Accounting Rev. 273, 277-78 (1979).
85 Lawrence R. Dicksee, Auditing 159-66 (1892).
86 Littleton, supra note 78, at 289.
87 Bishop C. Hunt, Auditor Independence, 59 J. Accountancy 453 (1935).
88 Francis W. Pixley, Auditors (1881); and also Dicksee, supra note 85.
89 Hunt, supra note 87, at 454.
AGENCY PROBLEMS 629
The substitution of professional auditors

audit committees appears to have occurr
evidence of professional firms' being appoi
ever, it is likely that professional firms we
committee because the 1845 Companies Act
outside experts at company expense." The o
that the auditor be a shareholder from the o
suggests pressure to appoint professionals
tendency to employ professional accountant
tendency appears to be connected to the fl
would be expected if an independent audit
promoters). Pixley reports the tendency th
tuses of new Companies now include amon
professional Accountants as their Auditors
are gradually replacing the Shareholders
one."91
B. Development in the United Stat

The 1933 Securities Act required that cor
have audits by independent or certified pub
the 1920s most companies listed on the N
(NYSE) were already audited by professiona
that 82 percent of NYSE companies had pr
The substitution of professional auditors
curred later in the United States than in th
by 1900 most traded U.K. companies were
tants, but in that year only a minority of
audited by professional auditors. A random
listed on the NYSE in 1900 whose annual
Baker Library or the Columbia University
companies with professional auditors.93
Many of the U.S. audit firms existing in 1

91 Pixley, supra note 88, at 165.
92 George J. Benston, The Effectiveness and Effects
Requirements, in Economic Policy and the Regulation
G. Manne ed. 1969).
93 Professor David Fehr of the Harvard Business School provided us with the findings
from the annual reports in the Baker Library.
chartered accountants who came to the United States to audit American

companies selling securities in London.94
C. Explanations for the Professional Firm

The substitution of professional auditors for shareholder auditors oc-
curred in both Britain and America in periods when a professional auditor
was not required by law. This suggests that the substitution was the result
of market forces. There were two major market developments in the
period 1844-1900 in the United Kingdom that can explain the shift from
shareholder to professional auditors: (1) an increase in the demand for
audits and (2) the introduction of a low-cost mechanism for certifying
auditor competence and independence.
1. An Increase in the Demand for Audits. The demand for audits in
the United Kingdom in the 1860s and 1870s increased because the com-
plexity of the accounts, the legal liability of directors, and the size and
number of corporations all increased. Accounts became more complex in
the latter half of the 1800s in the United Kingdom because of government
regulation of railroads and utilities and taxation that created the incentive
to write off a portion of the capital stock each year.95 In the same period
courts began holding directors liable if the company was not using more
complex accounting procedures (for example, provisions for doubtful ac-
counts).96
Another important factor shifting the demand for auditing was the tre-
mendous expansion in the number of companies. The nominal value of
listed securities on the London Stock Exchange increased fourteenfold
between 1853 and 1893.97 While most of that increase was caused by the
enormous growth in foreign securities traded (thirty-eight-fold increase),
private domestic security values increased fourfold.
The increased complexity encouraged specialization in auditing and
hence the growth of professional firms. The growth in the scale of the
capital markets increased the fixed cost of an auditor's establishing a
reputation that would serve as a bond for the auditor's independence.
This led to the development of large professional audit firms.
2. The Introduction of a Low-Cost Mechanism for Accrediting Au-
ditors. The first professional society of accountants was formed in Scot-
94 C. A. Moyer, Early Developments in American Auditing, 26 Acc. Rev. 3 (1951); John

L. Carey, The Rise of the Accounting Profession 27 (Amer. Inst. of Cert. Pub. Accnts. 1969-
70); Chester W. DeMond, Price, Waterhouse & Co. in America 10-12 (1951).
95 Watts and Zimmerman, supra note 84, at 290-95.
9 Dicksee, supra note 85, at 251-53; and Littleton, supra note 78, at 309.
97 Edward V. Morgan & William A. Thomas, The Stock Exchange (1962), at Table 5.
AGENCY PROBLEMS 631
land in 1854 and in England in 1870.98

to provide information on the accountan
rather in bankruptcies.99 The first ban
fessional accountants to be appointed tr
1772 and in England in 1825. Later Eng
tants' role and created a demand for in
accountants. The 1869 act so increased
competing for appointment as trustee
"The whole affairs in bankruptcy have
set of men called accountants, which wa
introduced into the law." 100 The profe
names, thereby providing information o
integrity. Brand names were established
dards of conduct, examinations for admi
"chartered accountant." Once the mech
certifying quality in bankruptcies was
certify quality and independence in aud
The rapid substitution of professional
shareholder auditors in the United King
can be explained by (1) an outward shif
and professional auditing being a declini
a downward shift in the supply curve
crediting accountants were already incur
bankruptcy work.
The development of professional audit
States than in the United Kingdom is co
nation for the substitution of professio
United Kingdom had similar rates of gr
98 Richard Brown, A History of Accounting an

208, 235, 237.
9 Littleton, supra note 78, at 271-84. The origin
not even list auditing as one of the public accoun
207-08.

101 Nicholas A. H. Stacey, English Accountancy 21 (1954).
102 A non-mutually exclusive hypothesis for the establishment of professional societies is
that they were designed to restrain trade. However, it does not appear that this motivation
was important, because the professional societies in the United Kingdom did not take any
action to restrict entry until 1893, when the Institute of Chartered Accountants in England
and Wales had a bill introduced in Parliament "to restrain all persons from practising who
are not Chartered Accountants." Brown, supra note 98, at 243. Moreover, this bill was
withdrawn and other societies competed with the Institute for the next seventy years.
103 Stigler, supra note 51.
the years from 1853 to 1903,104 but the U.K. capital mark
larger than in the United States in 1853. Thus, the absolute in
scale of the capital markets and in the demand for auditing
1844-1900 was larger in the United Kingdom. In addition
States did not experience the same reduction in start-up co
crediting professional accountants.
Unlike the United Kingdom, the United States did not ex
change in bankruptcy laws that gave creditors the powers
countants as trustees. Hence, there was less demand in the U
for information on an accountant's reputation for handling
and consequently, professional accounting societies were no
to accredit accountants.
The first U.S. professional accountants society was formed in 1887,105
thirty-three years after the first Scottish society and seventeen years after
the first English society. At that time most of the accountants' work in the
United States involved audits and investigations, and little or none in-
volved liquidations and bankruptcies.106 The start-up costs of the Ameri-
can societies were borne largely by British accountants who came to
America to audit firms raising capital in London and stayed to start their
own firms. Evidence of the value of the British auditors' brand name
capital is provided by the British auditors' urging the American society to
adopt the title "certified public accountant" to designate their members
instead of the title of "chartered accountant." 107
The first U.S. society began accrediting members in 1896, when the first
certified public accountants law in New York state was passed. Following
this law, nonaccredited auditors were rapidly replaced, so that, as noted,
by the 1920s most NYSE companies were audited by professionals.
The change from shareholder auditors to professional auditors was not
a switch from nonindependent to independent auditors, if independence is
interpreted as the likelihood that an auditor will report a discovered
breach of contract. An important incentive to be independent, the effect
of failure to report a breach on the auditor's reputation and hence his
future business, existed for both the shareholder and professional audi-
tors.
104 Ross L. Watts & Jerold L. Zimmerman, The Markets for Independence and Indepen-
dent Auditors 30 (June 1981) (unpublished manuscript, Univ. Rochester, Grad. School
Management).
105 Carey, supra note 94, at 36-39.
'06 Brown, supra note 98, at 278-79.
107 James D. Edwards, The Antecedents of American Public Accounting (1956), reprinted
in Contemporary Studies in the Evolution of Accounting Thought 53 (Michael Chatfield ed.
1968).
AGENCY PROBLEMS 633
An interesting example of the importance

to the auditor is provided by the expa
Company to the United States in the
Waterhouse and Company did not allow
they sent to America (Jones and Caesar)
damage to Price Waterhouse's reputation
IV. CONCLUSIONS
The survival of the bonding and auditing practices from the Ipswich
merchant guild in 1200 to the British joint stock banks of 1836 and U.S.
banks and canal companies of the mid-nineteenth century is consistent
with the existence of agency problems and the use of bonding and moni-
toring devices to reduce agency costs.109 Moreover, the size and composi-
tion of audit committees are observed evolving in response to changes in
the size of the firm and the nature of the agency costs faced by the
contracting parties.
The pervasiveness of voluntary (costly) auditing in the precursors to
the modern corporation is consistent also with auditors' developing
quality-assuring devices-in particular, mechanisms that increase the
probability the auditor will report a breach in a contract he is to monitor
(that is, be independent). We observe mechanisms being devised that
supplied the incentives for auditors to maintain their independence in the
guild and regulated companies (committees and penalties, including loss
of reputation), in the joint stock companies, and finally in the develop-
ment of professional societies.
Overall, the evidence suggests that the existence of the independent
auditor is not the direct result of government fiat. The appearance of the
professional independent auditor was encouraged by changes in U.K.
bankruptcy laws, but the United States' evidence suggests that even with-
out those bankruptcy laws, economies of scale in auditing would have led
to the development of the professional independent auditor.
108 DeMond, supra note 94, at 16.

109 T. A. Lee, The Historical Development of Internal Control from the Earliest Times to
the End of the Seventeenth Century, 9 J. Accounting Res. 150 (1971), traces the use of audits
in government back to ancient times.

Watts Zimmerman Agency Problems Theory of Firms

Uploaded by

Copyright:

Available Formats

You might also like

Watts Zimmerman Agency Problems Theory of Firms

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Watts Zimmerman Agency Problems Theory of Firms

Uploaded by

Copyright:

Available Formats

The Booth School of Business, University of Chicago

ROSS L. WATTS and JEROLD L. ZIMMERMAN

Recent developments in the theory of the firm emphasize the impor-

* Associate Professors. The authors gratefully acknowledge the comments on earlier

[Journal of Law & Economics, vol. XXVI (October, 1983)]

they were specifically required by government regulation, is

ism. Enforcement of the contract requires mo

(members). The composition and size of these committees changed

II. AUDITING IN THE EARLY CORPORATION

In this section we provide a description of each type of business corpo-

A. English Merchant Guilds

1. Description. Merchant guilds appeared in England shortly after the

8 1 Charles Gross, The Gild Merchant 4 (1890); 1 W. Ashley, An Introduction to English

transferring the property.12 If the guild it

12 Gross, supra note 8, at 99.

15 Gross, supra note 8, at 99.

Several craft guilds and companies of merchants were audi

supra note 8, at 34, to support his statements in vol. 1. We obtained the de

cost, lack of performance and independenc

B. Merchant Adventurers/Regulated Companies

1. Description. Prior to the latter half of the thirteenth century the

29 Gross, supra note 8, at 34.

There is evidence that some of the members of the earl

The codification of the ordinances and

C. Joint Stock Companies

with the greater risk of the voyages u

55 Fama, supra note 1.

1566,64 and it was not until 1588 that another company w

"The usual quarterly examinations, wit

72 Nicholas B. Wainwright, History of the Philadelphia National Bank 20 (1953).

D. Observations on the Development of Auditing

See Scott, supra note 8, at 451-52. Emphasis added.

eral committee (assistants) out of a total

80 Stevens, supra note 67, at 58-63.

teenth century companies issued numerous bonds.83 Compa

III. THE DEVELOPMENT OF THE PROFESSIONAL AUDIT FIRM

In this section we provide brief descriptions, of the development of the

A. Development in the United Kingdom

83 Dodd, supra note 76, at 112.

The substitution of professional auditors

B. Development in the United Stat

9 Littleton, supra note 78, at 296.

chartered accountants who came to the United States to audit American

C. Explanations for the Professional Firm

94 C. A. Moyer, Early Developments in American Auditing, 26 Acc. Rev. 3 (1951); John

land in 1854 and in England in 1870.98

98 Richard Brown, A History of Accounting an

100 Littleton, supra note 78, at 283.

An interesting example of the importance

108 DeMond, supra note 94, at 16.

You might also like