You are on page 1of 14

On Applying AgencyTheory in Historical

Accounting Research
Robert

Bricker

Department
ofAccountancy
CaseIVestern
Reserve
University
Nandini

Chandat*

Department
ofAccounting
Rutgers
University

Accounting
theoryandaccounting
historyhavelargelyevolvedastwo
independent
fieldsof study.
Accounting
research
is closely
alliedwithcontemporarymodelsin economics
and finance,whichemphasize
rigorin logical
construction
andstatistical
analyses.
Studies
in accounting
history,in contrast,
applyeconomic
theoryveryinfrequently,
andare predominantly
descripfive.
There has been litfie communicationbetween accountantsand historians,

thougha tremendous
synergistic
potential
exists.
Theimportance
of combining
historicaland statistical
approaches
in advancing
economictheoryand the
complementary
natureof the two approaches
havelongbeenemphasized
by
scholars
suchasKeynes[1890]andSchumpeter
[1954].
Accounting
is concerned
withinformation
flowsandthekorganization,
whicharecentralto business
operations,
managerial
decision-making,
andthe
natureandefficiency
of capital
markets.
Hence,thedevelopment
of accounting
is embeddedin issuesrelatingto the development
of the marketsand
corporation.
Tools of historicalscholarship
are thereforeuniquelysuitedto
incorporatingcontextualfactorsand path dependencies
in evaluating
accounting
theory.
Accounting
theoryalsohasthepotential
to enrichourunderstanding
of
business
history.As Lamoreaux,
Raff, and Temm [1997,p. 77] suggest,
historians
wouldbenefitfromtamingto economic
theory"bothfor useful
ideasandfor the lighta coherent
perspective
shedson an otherwise
untidy
past."Thisessay
provides
an illustration
of potential
synergies
in combining
historical
andtheoretical
research
in accounting.
We focusontheuseof agency

* We aregratefulto PaulMirantiandGaryPreyitsfor manyhelpfulconversations.


The
comments
of theparticipants
at the 1998Business
HistoryConference
werealsoextremely
beneficial.
Any errorsare,of course,ourown.

BUS1NESS
AND ECONOMICHISTORY,Volume
Twenty-seven,
no.2,Winter
1998.
Copyright
1998 bytheBusiness
History
Conference.
ISSN0894-6825.
486

AGENCY THEORY IN HISTORICAL ACCOUNTING

RESEARCH / 487

models of capital marketsrelationships,


which are the cornerstoneof
contemporan/accounting
research.
Current mainstreamaccountingresearchis based extensivelyon
economicmodelsof agencythat representthe operatingcompany(firm)
manager
as"agent"andthe individual
investoras"principal."
Thisprincipalagentmodelhasalsobeenimplicitly
adopted
in theregulation
of accountancy,
which focuseson the needsand welfareof a diversegroupof individual
investors
whoentrusttheirwealthto thecontrolof managers.
In this paper,we challenge
this traditional
characterization
of capital
marketsagency
rehfionships
in an accounting
context.In particular,
we focus
on the role phyed by fund managersand other managersof capitalwho
aggregate
the wealthof individuals
and makeinvestmentdecisions
on their
behalf.We performanhistorical
analysis
of thenatureof controlandinfluence
(de factopropertyrights)relationships
as theyexistedat the turn of the century,whenthe corporate
economy
wasstillin its infancy,andcomparethese
relationships
with thosethat haveevolvedamongindividual
investors,
fund
managers,
andfirmmanagers
overthepasttwodecades.
Thesetwoperiodsin
the evolutionof the Americancorporateeconomyare especially
interesting
because
theysignifytwo erasin whichmoneymanagers,
fundmanagers,
and
othermanagers
of capitalhavephyedimportantrolesin capitalmarkets.
Whileouranalysis
takesintoaccount
fundamental
differences
in context
acrossthe two time periodswe study,someinteresting
similarities
are also
apparent.
We findthat,likethemoneymanagers
andbankersat the turnof the
century,fundmanagers
andotherprofessional
investment
managers
of today
playa fundamental
partin transforming
thenature,structure,
andvaluationof
property
rightsin capitalmarkets.
Theirrole,therefore,
goeswellbeyondtheir
traditional
characterization
asflow-through
entitiesor financial
intermediaries.
We suggest
an expanded
framework
of capitalmarkets
agency
rehfionships
in
whichinvestment
managers
arerepresented
as agentsof individualinvestors

andprincipals
of firmmanagers
asa richerbasisfor framingandconducting
accounting
research,
andin policydeliberations.
Our approach
in thispapermaybeviewedasanswering
increasing
calls
for the adoptionof a theoreticalbasisfor conducting
historicalresearch

[Ternin,
1991;Baskin
andMiranti,1997].We concur
withtheviewof Douglas
North [June1994,p. 359]whenhe states:
"Economic
Histoo/is aboutthe
performance
of economies
throughtime.The objective
of research
in thefield
is not onlyto shednewlighton theeconomic
past,but alsoto contribute
to
economic
theov/byprovidingan analytical
frameworkthatwill enableus to
understand
economic
change."
AgencyResearchin Accounting

Whiletheissueof separation
of ownership
fromthecontrolof property
had beencommented
uponby AdamSmith[1976],Veblen[1923],and the
PujoCommittee
[1913],amongothers,the chssicworkof BerleandMeans,

488 / ROBERT BRICKER AND NANDINI

CHANDAR

TheModern
Corporation
andPrivate
Property
[1932]hashad a pivotalimpacton
accounting
research
andregulation.
Combining
legalandeconomic
perspectives,BedeandMeansput fortha provocative
thesisthatthe separation
of the
risk-bearing
functions
of ownership
andthe controlfunctionof management
createdconditions
in whichprofessional
managers
couldtakeactionsto the
dement of the ownerandto theirown personalgain.One consequence
of
this, they argued,was the urgentneed for new and reliablechannelsof
communication
in orderto protectshareholder
interests
by enablingthemto
judgemanagerial
performance
in theirroleasstewards
of corporate
resources.
The BedeandMeans'view,comingasit did on the heelsof the Crash
of 1929 foundfavorwith securities
regulators
lookingfor answers.
The New
York Stock Exchange(NYSE) was quick to implementseveralof their
recommendations,
andin 1933,the NYSE in conjunction
with the American
Instituteof Accountants
beganto promulgate
accounting
standards
for companiesthatwerelistedon the NYSE. Thismayhavebeena caseof "toolitfie
too late"for the accounting
profession,
asit did not stemthe publicpressure
for governmental
intervention.
2 The Berle and Means thesisbecamethe
foundationfor the passage
of the Securities
Acts of 1933 and 1934,which
established
legalresponsibilities
in connection
with the agencyrelationship
betweenshareholders
andmanagers.
Oneimportantoutcomeof theseactswas
that accounting
reportscontaining
informationaboutthe financialconditions
andresultsof operations
hadto be madeavailable
by managers
of companies
interested
in accessing
publicsecurities
markets.
The Bede and Means [1932] work becamethe foundationfor subsequentcapitalmarketsagency
modelsin accounting.
In particular,
Jensenand
Meckling[1976] synthesized
earlierworks by Bede and Meanswith the
propertyrightsandcontracting
literature
developed
by Coase[1937,1960],and
AlchianandDemsetz[1972].Theydeftnedthe firm as a "legalfiction"that
maybe characterized
asa "nexusof contracts."
Their analysis
focuseson the
agencyrelationship
betweenshareholders
of a fLrm(principals)
andmanagers
(agents).
The principals
contractwith the agentto performsomeservices
on
the principal'sbehalf.Thesecontractsrequirethe agentto exert effort and
makedecisions.
The agentis assumed
to maximizehis utility,whichis a
functionof both pecuniary
and nonpecuniary
costsand benefits.In the
process,
the managerhasincentives
to takevalue-reducing
or opportunistic
actionssuchasoverconsuming
perks,shirking,
or stealing.
With competition
and rationalexpectations
in capitalmarkets,suchbehavioron the part of
managers
reducesthe valueof the tom. This reductionin tom valueis termed
"agencycost."Contractsbetweenshareholders
and managers
are writtenin
The 1983Journal
ofl_v andEconomics
symposium
celebrating
the 50thanniversary
of
thepublication
of BedeandMeans[1932]provides
anoverview
of theenduring
influence
of
this work.

2 Historicalstudiesof the development


of accounting
includeMiranti [1989],and
PrevitsandMerino[1979].

AGENCY THEORY IN HISTORICAL ACCOUNTING

RESEARCH / 489

orderto reduce
agency
cost,andthereby,
thedeadweight
lossin firmvalueasa
resultof theseparation
of ownership
fromcontrol.
Accounting
isconsidered
to playanimportant
roleasanintegral
partof
thecontracts
thatdefmea firm.For example,
lendingarrangements
betweena
firm and its creditorsoften containseveralaccounting
basedcovenants.
Accounting-based
bonusplansarefrequenfiy
a component
of executive
compensation
plans.Accounting
measures
arecommonly
usedin theperformance
evaluationof a firm'scostand profit centers.Wattsand Zimmermanargue
[1986,p. 196]:"if accounting
is an importantpartof the firm'scontracting
process
andagency
costs(andhence,firmvalueand/ormanagers'
compensation)varywith differentcontracts,
accounting
procedures
havethe potential
to affectfirm valueand/or the manager's
compensation."
This rationalehas
givenriseto several
hypotheses
regarding
theroleof accounting
information
in
marketvaluation
of firmsandmanagers'
useof accounting
discretion.
Parallel
to thesetheoretical
developments,
methodological
innovations
in fmancelike
the capitalassetpricingmodels(CAPN
0 madeit possible
for researchers
to
subject
thehypotheses
derivedaboveto statistical
empirical
tests.
The hypothesis
that accounting
reportsare demanded
to monitorthe
relationship
betweenmanagersand shareholders
is termedthe stewardship
concept
andhasbeenusedby accounting
historians
to explaintheexistence
of
accounting
[e.g.,Yamey,1962].The principal-agent
modelhasalsogenerated
otherhypotheses
relatingto accounting
methodchoice,rationales
for accountingregulation
andinformativeness
of accounting
reports.
The implications
of
the manager-shareholder
agencyrelationship
for contractdesignhavebeen
studied
usinganalytical
models.
Thisliterature
[e.g.,Ross,1973,1974;Wilson,
1968;Spence
andZeckhauser,
1971;Mixrlees,
1974,1976;Stiglitz,1974,1975;
Holmstrom,1979;andAnfie,1982,1984]addresses
the principal's
problem
thattheagentor manager
will shirkandnottakeactions
thatwillbe in thebest
interests
of the principal.
The problemarisesbecause
the principalcannot
observetheagent's
actions.
Therefore,theoptimalcontractbetweenthe principalandthe agentwill providefor the agentto sharein the outcomeof his
actions.
The focuson designing
thesecontracts
is to providetheappropriate
incentives
to theagentandoptimalrisk-sharing
between
theprincipal
andthe
agent.This normativeagencyliteraturehasevolvedparallelto the positive
agency
literature
derivedfromtheworksof BerleandMeans[1932]andJensen
andMeclding[1976].Theseanalytical
modelsalsoprovideintuitionregarding
theroleof accounting
information
in principal-agent
relationships.
In orderto
preservethe mathematical
tractabilityof theixanalysis,
thesemodelshave
portrayedextremelysimplisticsituations
and have been basedon several
unrealistic
assumptions.
3

The contracting
role of accounting
haslargelybeenviewedin the
contextof simpleagencymodelsof the shareholder-manager
relationship.
Accounting
researchers
haveusedformalmanagement
compensation
plansand
aBaiman[1982]provides
a goodoverviewof thisliterature.

490 / ROBERT BRICKER AND NANDINI

CHANDAR

firmdebtcontracts
to generate
andtesthypotheses
aboutmanagers'
choiceof
accounting
procedures
and the stockpriceeffectsof thesechoices[e.g.,
Zmijewskiand Hagennan1981;Healy, 1985;DeAngelo,1986, 1988;Jones
1991;Gaver,Gaver,andAustin,1995; Holthausen,
Larcker,andSloan1995].
The agency
paradigm
anddevelopments
relatingto the CAPM andthe

efficientmarketshypothesis
have also influencec
accounting
regulation.
Accounting
researchers
in thistraditionhavesupported
theeffortsof accounting regulators
by analyzing
the implications
of theseeconomicmodelsfor
disclosure
regulation.
The FinancialAccounting
Standards
Board(FASB),its
predecessor,
the Accounting
Principles
Board(APB),and the Securities
and
ExchangeCommission
(SEC) haveutilizedstudiessuchas Beaver[1973],
BeaverandDemski[1974],Benston[1969,1973],etc.asthebasisof regulating
thepracticeof accountancy.
The principal-agent
literaturehasthushad a tremendous
impacton
accounting
researchand regulation.Yet, this researchhas been narrowly
focused,and has been predominantly
basedon analyticalor statistical
approaches
employing
largestandardized
databases.
Suchexclusive
relianceon
a narrowbasisis inappropriate
in advancing
knowledge
in an interdisciplinary
subjectlike accounting,
whichexistswithina complexinterplayof political,
socialandeconomic
settings.
Strikingly
absentfromcontemporary
accounting
research
isa modeling
or evenanunderstanding
of thesecontextual
factors.
Historicalscholarship
in uniquelysuitedto bridgingthis void by
providingthe necessary
toolsto understand
factorsthat havecontributed
to
economic
change.
A fewstudies
haveapplied
contemporary
agency
theoryand
relatedmodelsin information
economics
to historiceventsor periods.
Watts
andZimmerman
[1983]useda principal-agent
setting
fromwhichto arguethat
auditors'
reputation
served
asa bondforindependence
evenin earlymerchant
guilds.De Long[1991]investigated
theroleof the Houseof Morganin the
earlycorporateeconomyin America,suggesting
that the need to resolve
principal-agent
problems
engendered
by the corporate
form of organization
results
in thecreation
of institutions
thatwouldnotexistin a perfectly
competitive world.He arguedthatJ.P. Morganservedthe interests
of individual
investors
anddid not behavein a self-serving
mannerbecause
of theneedto
preserve
thevastamountof reputafional
capitalat stake.Ramirez[1995]used
contemporary
financemethodology
to providean empiricalevaluation
supporting
the proposition
that J.P. Morganand Companyresolvedthe
principal-agent
problemby alleviating
informational
asymmetries
between
investors
andmanagers.
He portrayed
J.P.Morganasa financial
intermediary,
who facilitated
a smoother
functioning
of the primaryagencyrelationship
between
individual
shareholders
andmanagers.
Thesestudiesare primarilyconcerned
with applyingcontemporary
agencyandrelatedeconomic
modelsto historical
timeperiods.The issueof
howrelevant
thesemodelsarein thecontextof particular
timeperiods
in businesshistoryhashowever
not beenaddressed.
In thisessay,
we usehistorical
research
in a mannerthatis fundamentally
differentfromthestudies
described

AGENCY THEORY IN HISTORICAL ACCOUNTING

RESEARCH / 491

above.Ratherthanapplythe existingstylized
modeldirectlyto the past,we
providea dynamicassessment
of the descriptive
validityof the model.Our
analysis
focuses
on the turnof thecentury,
whenthe corporate
economy
was
still developing,
and on the pastquartercenturywhichhaswitnessed
the
tremendous rise of the institutional investor. We demonstrate the limitations of

characterizing
a directandsimpleprincipal-agent
rehtionship
betweenindividualshareholders
andfirmmanagers
in capitalmarketsthataredominated
by
thepresence
of theseinstitutional
investors.
Capital Markets Relationshipsin the Early CorporateEconomy

An analysis
of theevolution
of agency
relationships
in capitalmarketsis
animportantperspective
fromwhichthedevelopment
of thecorporation
and
itsenvironment
maybe addressed.
Agencytheoryprovides
a particularly
useful
lensto studythepastwitha viewto understanding
theroleof accounting
informarionin the development
of marketsandorganizations.
Historicalevidence
thusobtainedmaybe utilizedto sharpen
the focusof contemporary
models.
The approach
alsohasthe advantage
that economic
modelsdeveloped
by
contemporary
researchers
arenot appliedin a mechanistic
mannerto studythe
past,thuslessening
thedangers
of presentmindedness
[Previts
andBricker1994].
America'sdrive to industrialism
initiatedby the developmentof
railroads
in the secondhalf of the eighteenth
centurygaveimpetusto the
growthof the corporateform of organization
[Chandler,Bruchey,and
Galambos,1968]. Until the 1880s,however,the typicalmanufacturing
company
wassmallandclosely
heldandlargely
served
localmarkets,
existing
merelyasextended
versions
of soleproprietorships
orparmerships.
Ownership
andmanagement
in thesecaseswereeithersubstantially
the sameor were
closelyrelated.Externalfinancial
reporting
wasthereforeconsidered
unnecessary
andevenunwise.
A corporation
hadthe rightto privacyregarding
its
financialinformation,just as any privatecitizendid. The regulationof
accountancy
duringthisperiod,as reflected
in statecorporation
laws,also
reflected
thissentiment
against
publicdisclosure.
Thesestatelawsgenerally
required
two setsof reports:
onesubmitted
to publicauthorities,
whichwere
consideredto be confidential,and the other, summarizedfinancialinformation

submitted
to stockholders,
butnotto thepublicatlarge[Hawkins,
1963].
In the 1880slargeindustrial
combines
started
to emerge
in the United
States.
TheAmerican
economy
wasfastchanging
froma primarily
undifferenfiated,agrarian,
localeconomy
intoa differentiated,
urban,industrialized
one.It
wasalsoexperiencing
enormous
growth.Whilein the 1850s,the industrial
outputof the U.S. was far belowthat of England,by 1894the valueof
American
outputalmostequaled
thecombined
outputof theUnitedKingdom,
France,andGermany,
andbyWorldWarI, America
wasproducing
morethan
one-third
of theworld'sindustrial
goods[Chandler,
Bruchey,
andGalambos,
1968].Suchenormous
growthmadestructural
changes
inevitable,
and gave
impetusto the development
of a corporate
economy.
The integration
of

492 / ROBERT BRICKER AND NANDINI

CHANDAR

ownership
andcontrolthatwasinherentin the nineteenth
centutyftrrnand
evenearlyclosely-held
corporations
gavewayto theirseparation.
The developmentof railroads
furtherhastened
thegowthof thecorporate
economy
by
expanding
thepossibilities
forcommerce.
Suchenormous
expansion
withina relatively
shortspanof timerequired
thedeployment
of hugeamounts
of capital.
Capitalmarkets,
aswe knowthem
today,didnotexist.Investment
bankers
of thistimelikeJ.P.Morganplayeda
centralrolein fillingtheneedforcapital
to fuelthisexpansion
[DeLong,1991]
dueto theirabilityto mobili.e
largeamounts
of capital.
The combinations
andmergermovement
of thelatenineteenth
centuty
resulted
in the formationof several
publiclyheldcorporations.
In theabsence
of adequate
financial
information
or publicdisclosures
fromthesecorporations,
investors
boughttheirsecurities
primarily
on thebasisof theirconfidence
and
trustin theinvestment
firmsmarketing
thesecurities.
The investment
bankers'
involvement
constituted
a stampof qualityandan implicitguarantee
of the
security
[De Long,1991].Corporate
managers
alsodepended
on theseinvestmentbankers
to raisecapitalandmakea marketfor theindustrial
securities
of
their firms.

Therewerethustwofiersof capitalmaxkets
agency
relationships
during
thisearlycorporate
economy:
onebetweenindividual
investors
andinvestment
bankers and the other between investment bankers and industrial firms. The

Berleand Meansworld of powerful,concentrated


managers
who directly
controlledthe wealthof individual
investors
wasnot yet in place.Individual
investors
wereeffectively
investing
in thereputation
of theinvestment
bankers,
whotheybelieved
hadinformation
to makegoodinvestment
decisions.
Financial
reporting
in thiserawasrarelythoughtnecessary,
prudent,or
evendemanded
[Hawkins,
1963].Themeager
financial
statements
thatdidexist
werecompletely
inadequate
for purposes
of individual
investorvaluationof
securities.
The accounting
profession
wasstillin its infancy.Therewasno
established
bodyof accounting
theory,andthelackof uniformity
in accounting
practices
renderedfinancialstatements
inadequate
asa basisfor comparative
evaluation
of firms.Auditingpractices
were still considered
unusual.The
functionof publicaccountants
andtheirreportswasgossly
misunderstood,
and most accountants
"neithercouldnor desiredto modifymanagement's
desirefor corporate
secrecy"
[Hawkins,
1963,p. 145].
Financial
relationships
in capitalmarkets
weremadepossible
therefore
by the structureof propertyrights.Investorsduringthis periodexpected
regulardividends
fromthe corporations,
andaslongas theyreceived
these
dividends did not care about financial statements. The nature of control and

influence
relationships
thatexistedat thistimealsomadeit possible
for the
organization
andfinancing
of corporations
evenwiththeabsence
of structured
capital
markets,
adequate
information
flows,securities
regulations
anda general
social
acceptance
of thecorporate
formof organization.
Investment bankers were at the center of these control and influence

relationships,
as is well-documented
in business
historyresearch.These

AGENCY THEORY IN HISTORICAL ACCOUNTING

RESEARCH / 493

bankers,
particularly
J.P.Morgan,
wereinvolved
initially
withrailroadfinancing.
They actedas managers
of largeunderwriting
syndicates
whichagreedto
purchase
stocks,
convertible
bondsandothersecurities
not purchased
by the
railroadcorporation
in anoffering,assoleunderwriters,
or asdirectsubscribers
of securities
throughprivateplacements
[Chandler
andTedlow,1985].These
activities,of centralimportancein an economicsystemwith relatively
undeveloped
capitalmarkets,increased
the bankers'controlover the raikoad
industry.The systemworkedon faith, reputation,and characterof these
bankers,
allof whichwereessential
factors
in theearlydevelopment
of corporations.During the economicdepression
of the 1890s,the industrywent
througha seriesof reorganizations,
againfinanced
by theinvestment
bankers.
Theserequiredraisingof cashthroughtheissueof newsecurities,
realigning
fixedcharges
by exchanging
new securities
for the old, and the creationof
voting trusts which were vestedwith full corporateauthorityby the
shareholders.
The votingtrustswerecontrolled
by investment
bankerswho
usedthemasvehicles
to wieldsignificant
controlovertherailroads.
The trusts
servedto separate
thevotingandownership
fightsinherentin commonstocks.
The mergermovementthat startedwith the railroadsspreadto other
industrial
concerns
in thelatenineteenth
century,
resulting
in a numberof large
pubticlyownedmanufacturing
companies.
Somewell known companies
of
today,like GeneralElectric,AmericanTelephoneand Telegraphs,
Federal
Steel,UnitedStatesSteel,andInternational
Harvester
wereorganized
at the
turnof thecentury.
The needfor capitalandthefragmentation
of theindustry
(particularly
in thecaseof AT&T), resulted
in a closerelationship
between
the
companyand its bankers.The bankersfosteredcombinations
in industrial
concerns
withthepurpose
of bringing
themarketundercontrol.For example,
by 1900,Standard
Oil througha seriesof horizontalandverticalcombinations
that wereorchestrated
by Morgan,controlled90% of the domestic
industry
[Galambos
andPratt,1988].In the restructuring
of elevenmajortomsof the
steelindustryinto the United StatesSteelCorporation,the first billion-dollar

company,
investment
bankers
onceagain
werecentral.
Theunderwriting
circular
issuedin thisconnection
notedthat "the entirePlanof Organization
and
Management
of theUnitedStates
SteelCorporation
shallbedetermined
byJ.P.
Morganand company"
[Chandler
andTedlow,1985,p. 282].J.P. Morgan
orchestrated
thecreation
andfinancing
of U.S.Steel,
handpicked
themanager,
obtained
a 200%returnon fundsadvanced
duringthemerger,andreceived
hugefees.
The meansfor financing
corporations
wereconcentrated
in a fewhands.

Thesebankers
became
important
investment
agents
at theturnof thecentury.
In earlycapital
markets,
thereexisted
agency
relationships
between
a comparativelysmallnumber(fromtoday's
perspective)
of largecorporations,
anda
relatively
smallgroupof investment
managers
andindustrial
tom managers,
whoin turnhada fiduciary
relationship
with theindividual
investors
whose
resources
theymobilis.
ed.The BedeandMeansthesisignores
thevitalroleof
theseinvestment
managers.
In analyzing
thecorporation,
BedeandMeansdid

494 / ROBERT BRICKER AND NANDINI

CHANDAR

not address
the roleof bankersandunde-ritersasagentsof capitalism
and
therefore
didnotadequately
portrayproperty
controlrelationships.
Therewere,of course,severalvociferous
criticsof big business
and
moneymanagers[e.g.,Brandeis,1914].The ?ujo heatingson the "money
trust"reportedin 1913thattherewas"a highdegreeof financial
concentration
...anda closealliance
between
theheadsof a fewN&vYorkcitybanksandthe
principalusersand suppliersof capital"[Carroso,1973].De Long [1991,
p. 217] assertsthat "[t]he forty-fiveemployees
of Morganand Company
approvedand vetoedproposed
top managers,
decidedwhat securities
they
wouldundevrite,andthusknplicidydecided
whatsecurities
wouldbe issued
andwhatlinesof business
shouldreceive
additional
capital."
Thecontroloverinvestor
resources
waseffected
through
representation
on corporate
boards,
controlling
individual
savings
in lifeinsurance
companies
and banks,and by creatinga syndicate
system,
whichconsisted
of various
techniques
of originating,
undenvfig,anddistributing
newsecurities.
In the
1920s,otherformsof institutional
investment,
particularly
investment
trusts,
evolved.The absence
of regulation
separating
commercial
and investment
banking
aswe havetoday,together
witha healthy
U.S.economy
stimulated
the
rapidgrowthof thesetrusts.
4 Smallinvestors
invested
in thesetrusts,whichin
turninvested
in portfolios
of securities
of industrial
firms.Themarketvaluesof
thesecurities
of thetruststhemselves
weretypically
fargreater
thanthesumof
thevalues
of theproperty
thatthese
trustsowned,consisting
solely
of common
andpreferred
stocks
anddebentures,
mortgages,
bonds,
andcash.It isapparent
thatin earlycapital
markets
a pzemium
waspaidfor financial
entrepreneurship.

Theinvolvement
of investment
bankers
in theearlycorporate
economy
wentfarbeyondtheircontemporary
economic
conceptualization
as"financial
intennediafion."
BedeandMeansdid not address
thispatternof property
controlandinfluence
thatcharacterized
earlycapital
markets
in arguing
thatthe
balance
of powerhadshiftedentirely
in favorof theprofessional
corporate
manager.
Theirfailureto address
insfitufionalized
formsof corporate
control
limitsthe descriptive
validityof their analysis
in the contextof the early
American
corporate
economy.
Theredidnotexista directagency
relationship
between
individual
investors
andcorporate
managers.
Investment
bankers
and
trustmanagers
as agentsof individual
investors
provedto be knportant
monitoring
agentsagainst
opportunisfic
behavior
by corporate
management
[Carosso,1973].Agencymodelsthat do not considerthesetwo tiers of
relationships
do not accurately
portraycapitalmarketsrelationships
in early
American markets.

4 By 1927,Wall Streetinvestment
trustssold$400millionof securities,
andby the fall
of 1929,total assetsof thesetrustwereestimatedto exceed$8 billion,an elevenfoldincrease

[Galbraith,
1955].

AGENCY THEORY IN HISTORICAL ACCOUNTING RESEARCH / 495

CapitalMarketsRelationships
in Contemporary
CapitalMarkets

It hasbeensuggested
thatwe are returning
to an era of financial
capitalism
[Hawley,
1995].Investment
companies
todayplayanimportant
role
in themanagement
of capital,
asindividual
investors
areincreasingly
employing
professional
capital
managers.
Asof 1992,morethanhalfof thecommon
stock
outstanding
in theUnitedStates,
amounting
to over$2 trillion,wasownedby
pension
funds,mutualfunds,etc.,compared
with40%in 1980,andlessthan
15% in 1950) Institutionsaccountfor about80% of all tradingactivity.The
averageNew York StockExchange
transaction
now exceeds
2,000 shares,
whichis nearlysixtimesthefigurefor 1974,andhalfthedailytrading
volume
takesplacein blocks
of 10,000shares
or more.Meanwhile,
individual
investors'
directholdings
of common
stockrepresent
only16%of theirfinancial
assets,
down from 44% in the late 1960s,and transactions
of lessthan 100 shareshave

fallen to less than 2% of total volume.Bernstein[1992] suggests


that
"individuals
whobuyandsellfortheirownaccount
area disappearing
breed."
To be sure,thereare fundamental
differences
betweenthe earlyand
moderncorporateeconomies.
Followingthe stockmarketcrashof 1929,
regulatory
measures
like the Glass-Steagall
Act restricted
investment
activities
of banks.The federalgovernment
throughthe newlycreatedSecurities
and
Exchange
Commission
(SEC)playeda directrole in requiringfuller,more
reliableandmoreusefiddisclosures
by the managers
of industrial
finns.The
SEC supportedthe effortsof the AmericanInstituteof CertifiedPublic
Accountants,
andaccounting
rulescalled"generally
accepted
accounting
principles"werepromulgated.
Sincethisperiodtherehasbeena constant
pressure
on the part of finn managers
to improvethe qualityof their accounting
disclosures.
The measures
separating
investment
andcommercial
bankingand
enhanced
regulation
of corporatedisclosures
encouraged
individualinvestor
participationdirectlyin capitalmarkets.For a few decadesafter the Great
Crashof 1929,the emergence
of a retailmarketfor securities
renderedthe
Berle and Means model most closelyrepresentative
of capitalmarkets
relationships
astheyexisted.
During the prosperous
periodfollowingWorld War II, however,a
numberof nonbanking
and nondepository
institutions
suchas insurance
companies,
pensionfunds,and mutualfundsbecameimportantsecurities
holders.
6 As reportedin TheEconomist
[1990],Americanprivateinvestors
reduced
the net valueof theirequityholdings
by about$550billionbetween
theendof 1983andtheendof 1989,whichis about40%of theirportfolios
in
1983."Werethesetrendsto continue,
thelastAmerican
to holdshares
directly
wouldsellhislastonein theyear2003"[TheEconomist,
1990].

Statistics
on institutional
holdings
andtradingarecontained
in Bernstein
[1992].
The phrase"pensionfundsocialism"
hasbeenemployed
to characterize
theextentof
institutionalized
capital.

496 / ROBERT BRICKER AND NANDINI

CHANDAR

The growingeconomicimportanceof contemporary


havestment
managers
hasincreased
theixproperty
controlcapabilities.
Manylargepension
fundslikeCalpetshavedefinedandstructured
shareholder
activism
programs.
7
Thereareseveral
highlypublicized
hastmaces
of shareholder
activism.
In 1990,
shareholders
targetedITF for excessive
executivecompensation.
Pressures
from hastitutional shareholders resulted in the ouster of chief executives at several

companies
suchas GeneralMotors,AmericanExpress,IBM, Westinghouse
Electric,AppleComputer,
Eli Lilly,Kodak,ScottPaper,andBorden.Pension
fundsare also dixecfiyhavolved
in corporategovernance
throughboard
representation,
andin regulatory
areas.
In 1992,theSEC,asa resultof pressure
from institutionalinvestors,adoptednew regulations
virtuallyeliminating
restrictions
on communication
amongshareholders
of a company.Mutual
fundshavebeenlessprominently
involvedin corporate
governance.8
However,
investment
companies
like mutualfundshavea hugeamountof economic
power.9
Capitalmarketstructures
have also evolvedin a mannerfavoring
institutional
investors.
Tradingis now conducted
virtuallyroundthe clockin
after-hours
electronic
tradingaccessible
onlyto institutional
investors.
Large
institutionalinvestorscan now tradeblocksharesoff the floor. Further,some

privatelyplacedsecurities
(underRule 144A)maybe tradedonly among
institutional
investorsin privatetransactions.
Thesemeasures
"furtherthe
development
of a two-tierstockmarket"[Torres,1992].
In today's
American
economy,
institutions
exertenormous
controlover
theproperty
of individual
investors
andactastheiragents
in making
decisions
regarding
havesting
in operating
companies.
w By investing
haa mutualfund
ratherthanin anoperating
company,
anindividual
investor
haspropertyand
information
rightsagainst
theinvestment
fundandnot against
theoperating

conapmay.
Theinvestment
fundin turnhasthesefightsagainst
theoperating
company.
This fundamental
role of fundsis ignoredwhentheyare viewed
withinthe traditional
principal-agent
framework
that represents
operating
company
managers
asagentsof investors.

7SeeWahal[1994]fora description
of someof thelargerprograms
andtheprocess
of
activism.

sResearchers
haveargued
thatthishasbeena consequence
of U.S.Securities
lawsthat
aredesigned
to promote
market
liquidity
ratherthangoodgovernance
03hide,
1993].
9At theendofJune1994,mutualfundscontrolled
morethan$2 trillion,upmorethan
100%in just3 years[Henriques,
1994].
0Whileour analysis
focuses
on the U.S economy,
the propertycontrolcapabilities

exerted
byinvestment
managers
in theU.S.isechoed
in manyothercountries
likeGermany,
Japan,
andSouthAfrica.Fora discussion
of capital
markets
relationships
in thesecountries,
seeProuse[1995],Bartet al. [1995],andWalter[1993].

AGENCY THEORY IN HISTORICAL ACCOUNTING

RESEARCH / 497

Conclusion

Our analysis
suggests
thatthe simpleprincipal-agent
modelthatis the
basisfor muchof contemporary
accounting
researchand regulationis not
descriptive
of markets
in whichthereisa highproportion
of institutional
investors.A two-tiersetting
thatwouldrepresent
investment
managers
asagents
of
individualinvestors
and principals
of corporate
managers
wouldprovidea
moreaccurate
portrayal
of property
controlagency
relationships
in thesemarkets.
While we recognizethat there are potentiallyseveral agency
rehtionships
amongdifferentcapital
marketparticipants,
u weconsider
thatthe
elaboration
thatwe suggest
is directly
usefulin accounting
theory,practice,
and
regulation.
Our analysis
suggests
a needfor broadening
the scopeof agency
research
in accounting.
The two-tiermodelmoreclearlyillmates contracting
and finandalreporting
issues
involvinginvestors,
fundmanagers,
and corporatemanagers,
andwouldenableus to raiseand considertheoretical
issues
that we havenot yet done.The agencymodelalsoprovidessomeunique
perspectives
on thedevelopment
of accounting
thought,whichis grounded
in
theevolution
of capital
markets
agency
relationships.
Researchers
in accounting
historyandaccounting
theoryhavea lot to sayto oneanother.
References

Alchian,A.A., andH. Demsetz,"Production,


InformationCostsandEconomicOrganization,"American
Economic
Review,
62 (1972),777-795.

Ante, R., "The Auditorasan EconomicAgent,"Journal


ofAccounting
Research,
20 (Autumn
1982),Part II, 503-527.

, "AuditorIndependence,"
]ournalofAccunting
Research,
22 (Spring1984),1-20.
Baiman,S., "AgencyResearch
in Managerial
AccountmA Survey,"
Journal
ofAccounting
Literature,
(1982),154-210.

Bart,G., J. Getson,andB. Kantor,"Shareholders


asAgentsandPrindpals:
The Casefor
SouthAfrica'sCorporateGovernance
System,"
Journal
of AppedCo,orateFinance,
(Spring1995),18-31.

Baskin,
J.B.,andP.J.Miranti,
Jr.,A History
ofCo,orate
Finance
(NewYork,1997).
Beaver,
W.H., "WhatShould
be theFASB'sObjectives?"
Journal
ofAccountanff,
136(August
1973),49-56.

--,

andJ. Demski,"The Natureof Financial


Accounting
Objectives:
A Summary
and
Synthesis,"
in Stuces
onFinandalAccunting
Octives:
1974,supplement
to Vol. 12 of the
]ournal
ofAccounting
Research,
(1974),170-187.

Benston,
G.J.,"TheEffectiveness
andEffectsof theSEC'sAccounting
Disclosure
Requirements,"in HenryG. Manne,ed.,Ecnoraic
PoffandtheRegulation
ofCo,orate
Securities
--,

(Washington,
DC, 1969),23-79.
"RequiredDisclosure
and the StockMarket:An Evaluationof the Securities
Exchange
Act of 1934,"American
Economic
Rev/ev,
63 (March1973),132-155.

Berle,A.A.,andG.C.Means,TheModern
Cooration
andPrivate
Pro)erly
(NewYork,1932).
Bernstein,
P.L.,Capital
Ideas:
TheImprobable
Offgins
ofModern
lVallStreet
(NewYork,1992).
Bhide,A., "TheHiddenCostsof StockMarketLiquidity,"
Journal
ofFinancial
Economics,
34
n As discussed
in McCraw[19841
andMiranti[1990].

498 / ROBERT BRICKER AND NANDINI

CHANDAR

(1993),31-52.
Brandeis,
L., OtherPeople's
MonyandHolytheBankers
Useit (NewYork,1914).
Carosso,
V.P., "The Wall StreetMoneyTrustfromPujothroughMedina,"Bnsiness
History
Rev/e:v
(Winter1973),421-437.

Chandler,
A.D. Jr., S. Bruchey,
andL. Galambos,
TheChanging
Economic
Order.Readings
in
American
Business
andIndustry
(NewYork,1968).
Chandler,
A.D. Jr.,S.Bruchey,
andL. Galambos,
andR.S.Tedlow,TheComing
ofManagerial
Capitasm:
a Casebook
ontheHistory
ofAmerican
Economic
Im#tntions
(Homewood,Illinois,
1985).
Coase,R., "TheNatureof theFirm,"Economica,
4 (November1937),386-405.
, "TheProblemof SocialCost,"]ournalofLav
andEconomics,
3 (October1960),1-44.
DeAngelo,L.E., "Accounting
Numbersas Market ValuationSubstitutes:
A Studyof

Management
Buyouts
of PublicStockholders,"
TheAccounting
Reviev,
61 (July1986),
400-434.

DeAngelo,L.E., "Managerial
Competition,
Information
Costs,andCorporate
Governance:
The Useof Accounting
Performance
Measures
in ProxyContests,"
Journal
ofAcconning
andEconomics,
10 (1988),3-36.

De Long,J.B., "Did J.P. Morgan'sMen Add Value?An Economist's


Perspective
on
Financial
Capitalism,"
in Ternin,ed.,Inside
theBrainess
Enterprise,
205-236.
Editorial,"A Survey
of Capitalism:
Punters
or Proprietors,"
TheEconomist,
May5, 1990.
Fama,E., "Agency
Problems
andtheTheoryof theFirm,"Journal
ofPoal Economy
(April
1980),288-307.

Galambos,
L., andJ. Pratt,ThePaeoftheCorporate
Commomvealth
(NewYork,1988).
Galbraith,
J.K., TheGreatCrash1929(Boston,
1955).
Gaver,J., K. Gaver,and J. Austin,"AdditionalEvidenceon BonusPlansand Income
Management,"
Journal
ofAccounting
andEconomics,
19 (1988),3-28.
Hawley,J.P.,"Political
Voice,Fiduciary
Activism,
andthe Institutional
Ownership
of U.S.
Corporations:
The Role of Publicand Noncorporate
PensionFunds,"Sodological
Perspectives,
38 (1995),415-435.
Hawkins,
D.F., "TheDevelopment
of ModemFinancial
Reporting
Practices
amongAmericanManufacturing
Corporations,"
Business
History
Reviev
(Autumn1963),135-168.
Healy,P.M.,"TheEffectof BonusSchemes
onAccounting
Decisions,"
Journal
ofAccounting
andEconomics,
7 (1985),85-107.

Henriques,
D.B., "A TrustBetrayed:
Questions
of ConflictStingMutualFunds,"TheNev
YorkTimes,
August7, 1994.
Henriques,
D.B.,"A TrustBetrayed:
To SomeFundManagers,
RedFlagsDon'tMeanNo,"
TheNearYorkTimes,
August8, 1994.
Hendques,
D. B.,"A TrustBetrayed:
Seeking
Dataon Funds,Investors
andRegulators
find
Frustration,"
TheNevYorkTimes,
August
9, 1994.
Holmstrom,B., "Moral Hazard and Observability,"
The BellJournalof Economics
and
Management
Salerice,
(Spring1979),74-91.

Holthausen,
R.,D. Larcker,
andR. Sloan,
"Annual
BonusSchemes
andtheManipulation
of
Earnings,"
Journal
ofAccounting
andEconomics,
19(1995),29-74.

Jensen,
M.C.,andW. Meckling,
"Theoryof theFirm:Managerial
Behavior,
Agency
Costs
andOwnership
Structure,"
]ournalofFinandalEconomics,
3 (1976),305-370.
Jones,
j., "Earnings
Management
DuringImportReliefInvestigations,"
]onrnal
ofAcconnting
Research,
29 (Autumn1991),193-228.
Keynes,
J.N., TheScope
andMethod
ofPoetical
Economy
(London,1890).
Lamoreaux,
N., D.M.G. Raff,andP. Temin,"NewEconomic
Approaches
to theStudyof
Business
History,"Bnsiness
andEconomic
History,
26 (Fall1997),57-79.
McCraw,T., Prophets
ofRegulation
(Cambridge,
MA, 1984).
Miranti,P.J.Jr.,Accountan.cy
Comes
ofAge:TheDevdopmeut
ofanAccounting
Profission,
1886-1940

AGENCY THEORY IN HISTORICAL ACCOUNTING RESEARCH / 499


(ChapdHill, 1990).

Mirflees,
J.A.,"Noteson WelfareEconomics,
Information
andUncertainty,"
in M. Balch
etal.,eds.,EssqsonEconomic
Behavior
Under
Unctrtainy
(Amsterdam,
1974).
--,
"TheOptimalStructure
of Incentives
andAuthority
WithinanOrganization,"
Bell
]ournalofEconomics,
7 (Spring1976),105-131.
North, D., "EconomicPerformancethroughTime," American
Economic
Review,84
(June1994),359-368.

Previts,
G.J.,andB.Merino,
A Histo{y
ofAccountin&
inAmerica
('NewYork,1979).
--,
andR. Bricker,"FactandTheoryin Accounting
History:Presentmindedness
and
Capital
Markets
Research,"
Contemporary
Accounting
Research
(Spring
1994).

Prouse,
S."Coworate
Governance
in anInternational
Perspective:
A Survey
of Corporate
ControlMechanisms
amongLargeFirmsin theU.S.,U.K.,Japan,
andGermany,"
FinancialMarkets,
Institutions
andInstrnments,
V4, N 1,(1995),1-63.

PujoCommittee.
U.S.Congress.
House.Committee
on Banking
andCurrency.
Minority
Report
oftheCommittee
Appointed
pursuant
toHouse
Resolution
429and504toInvestigate
the
Concertira#on
ofControl
ofMon.ey
andCret(Washington,
DC, 1913).
Ramirez,
C.D., "DidJ.P.Morgan's
MenAdd Liquidity?
Corporate
Investment,
CashFlow,

andFinancial
Structure
at theTurnof theTwentieth
Century,"
TheJournal
ofFinane
(June1995),661-678.

Ross,S.A.,"TheEconomic
Theoryof Agency:
ThePrincipal's
Problem,"
Am#ican
Economic
Review,
63 (May1973),134-139.

Schumpeter,
J.A.,Histo{y
ofEconomic
Analsis('NewYork,1954).

Smith,
A.,An Inqui{y
intotheNature
andCauses
ofthelVealth
ofNations,
vol.2, reprinted
edn.
(Chicago,
1976).

Spence,
M., andR. Zeckhauser,
"Insurance,
Information,
andIndividual
Action,"American
Economic
Review,
61 (May1971),380-387.

Sfiglitz,
J.E.,"Incentives
andRiskSharing
in Sharecropping,"
Review
ofEconomic
Strokes,
41
(April 1974),219-255.

, "Incentives,
RiskandInformation:
NotesTowards
a Theoryof Hierarchy,"
Bell
Journalof
Economics,
6 (Autumn1975),552-579.

Temin,
P.,Inside
theBusiness
Enterprise:
Historical
PerfeX'yes
ontheUseofInrmation
(Chicago,
1991).

Torres,
C. "NYSERuleChange
LetsBigInstitutions
Bypass
FloorTrades,"
ThelVallStreet
Journal,
C1, C13,October26, 1992.

Veblen,
T.,Absentee
OwnershOo
andBusiness
Enteise
inRecent
Times:
TheCase
ofAmerica,
1923
('NewYork, 1964).

Wahal,
S.,"Pension
FundActivism
andFirmPerformance,"
Jouraal
ofFinancial
and,,Quantitative
Analsis,31 (1996),1-23.

Walter,
I., TheBattle
oftheSystems:
Control
ofEnterprises
and
theGlobal
Economy
(Kiel,1993).
Watts,
R.L.,andJ.L.Zimmerman,
Positive
Accounting
Theo{y
(Englewood
Cliffs,
NewJersey
1986).

, "Agency
Problems,
Auditing,
andtheTheory
of theFirm:Some
Evidence,"
Journal
ofLa andEconomics,
26 (October
1983),613-634.

Wilson,
R.,"TheTheory
of Syndicates,"
EconomeMca,
36(January
1968),119-132.
Yamey,
B.S.,"Some
Topics
in theHistory
of Financial
Accounting
in England
1500-1900,"
inW.T.Baxter
andS.Davison,
eds.,
Stues
inAccounting
Theo{y
(London,
1962),
14-43.
Zmijewski,
M.,andR.Hagerman,
"AnIncome
Strategy
Approach
tothePositive
Theory
of
Accounting
Standard
Setting/Choice,"
Jouraal
ofAccountin&
andEconomics,
3 (August
1981),
129-149.