Вы находитесь на странице: 1из 15
A Transaction Cost Perspective on Vertical Contractual Structure and Interchannel Competitive Strategies Author(s): F.

A Transaction Cost Perspective on Vertical Contractual Structure and Interchannel Competitive Strategies Author(s): F. Robert Dwyer and Sejo Oh Source: The Journal of Marketing, Vol. 52, No. 2 (Apr., 1988), pp. 21-34 Published by: American Marketing Association

Accessed: 30/12/2010 12:23

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless

you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

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.

information about JSTOR, please contact support@jstor.org. American Marketing Association is collaborating with JSTOR

American Marketing Association is collaborating with JSTOR to digitize, preserve and extend access to The Journal of Marketing.

http://www.jstor.org

A

F. Robert Dwyer & Sejo Oh

TransactionCost

Perspective

on

Vertical

Contractual

Structure

and

Interchannel

CompetitiveStrategies

Despite potent theory to contrast transactional governance

in "markets and hierarchies," the variety of

extant channel systems strains this dichotomy. The authors examine three contractually integrated chan-

nel systems in the hardware industry: wholesale

dents. They extrapolate from a transaction cost perspective to frame hypothesized

making

vey

and indepen-

voluntary chains, dealer cooperatives,

across relational forms.

differences in decision

structures and

competitive strategic postures

are generally supportive.

Results from a sample sur-

of retail informants

FOR

marketershave underscoredthe

fact that interorganizational coordinationin dis-

and fundamentalsur-

Guiltinan 1974; Little 1970;

and El-Ansary 1982). Typi-

several years

tributive systems is a potential

vival requirement(cf.

Ridgeway 1957; Ster

cally

jointly allocate responsibility

channel members eliminate redundant roles,

for key channel func-

tions, and establishcommunicationand control sys-

tems. Such coordination inevitablyrequires a firm to understandand manage its dependence on channel

partners. Membersof a coordinatedchannel system often

invest in dedicatedorder processing equipment, cul-

tivate selective demand

and mar-

ketingapproaches, and orchestrateextensive training

programs for effective role

partners.Thus, channel coordinationand efficiency

commonly require investment in

by special displays

performanceby channel

business assets

F.Robert Dwyer isAssociateProfessorof

cinnati. Sejo Ohis AssistantProfessorof BusinessAdministration,

Kangweon National University,

Lagace, AnnWelsh,andthe anonymous JMreviewersforcomments

on previous drafts. They alsothankJames Cory of Hardware Age for

sharing hisobservationsof the

by

University ofCincinnati.

the College of BusinessAdministration FacultyDevelopmentFund,

Marketing,University ofCin-

Korea.Theauthorsthank Rosemary

industry. Partial funding was provided

(physical or human) that promise efficiency but may not be easily redeployable(Williamson 1975, 1985). High levels of interfirm coordination,then, are sup- ported best by assurancesof relationshipcontinuity,

durability, and integrity (Macneil 1980; Williamson

1983, 1984). Several avenues can be taken to ensure sustained

participa-

tion in upstream

tions-provides

associations. Vertical integration-actual

or downstream marketing func-

continuity of performance and some

measure of

control

or governance

over

marketing

functions. For example, Sherwin Williams not only

manufactures paints but distributesand sells end users in its own retail outlets. Likewise,

them to

Allstate

insuranceis sold directlyby Allstate employees rather

than

independentagents.

Vertical

for

integration has its own challenges of utilization. The functional responsibil-

inventorymanagement,billing system perfor-

physical asset

ity

mance, and distributionand sales efficiency is ines-

capable. Humanasset

training,supervision,

may accrue, however, from reduced expenses

ciatedwith compensating resellersand monitoring their

performance(cf.

Phillips 1982). Alternatively, exchange relations may be sus-

managementrequiresemployee

andmotivation.Overall

savings

asso-

AndersonandWeitz 1986;Etgar1976;

Journal of Marketing Vol. 52 (April 1988),

21-34.

TransactionCost Perspective on VerticalContractualStructure/ 21

tained by informaland tacit

to these relationships have achieved mutual benefits

from quality role performance in previous transactions

(Frazier1983b),

er's preferences and capabilities,

pects

ties are not bound contractually in the classical sense

for continued

beneficial exchange. These

creasinginterdependence.Though both parties are likely

to attendto

dertakesseriousor frenetic

andOh 1987). Between these two extremes are many interme-

that is, struc-

turesfor administering and sustainingexchange sys-

tems.

dealer

cooperatives, wholesalers'

a host of

solidify expectations and obligations

tingent claims, third party mediators, and other ad-

justment mechanisms.

agreements. The parties

have come to understandeach oth-

and see

good pros-

par-

in the future. The

for

satisfyingexchanges

(Macneil 1980), but share expectations

expectations support in-

alternative exchangepartners, neitherun-

searching(Dwyer, Schurr,

governance,

diate forms of relational

They

include

franchising,systemsselling,

voluntary associations, and

specially designed contractsthat attempt to

and featurecon-

Objectives

The purpose of our researchis to compare the effects

of interorganizational form on (1) decision-making

processes

system in the marketplace.

Thus, we heed both Arndt's (1979) call for attention

competitiveposture

to interfirmtransactional governance

Goodman's (1979)

mesticationof marketsdoes not circumventthe mar-

within the

channel and (2) the resulting

processes and

of the

poignantcommentary thatthe do-

ket test of the interfirm system.

Background Arndt's provocative contention (p. 69) is:

The

tions

frameworkof long-term relationships. The emerging

the

domesticated markets call

management of interorganizationalsystems and to the

competitive

are

market

is

eroding

transac-

occurring in internal markets within the

for more attention to

political aspects of economic decision making.

Goodmandoes not deny the

comes from mobility

relationship(Day and Wensley 1983). In harmony with

the aims of our study, however, he brings

backinto focus by

competitiveadvantage that

buyer-seller

barriersinherentin a

underscoring the role

the market of the inter-

organizational

(p. 81).

system as

a competitive entity itself

[T]he

distinction

between

.

"domesticated

tems" and nonintegrated systems is in their degree

and

not in theirexemption from the final

There is at least strong anecdotal evidence suggesting

contractual supply

arrangements,

that

tightness of the internal linkages and flows and

sys-

voluntary groups,

.

and

specification buying con-

22 / Journalof Marketing,April 1988

tracts (e.g.,

to compete with

competition.

tires, appliances)

are structuredin order

alternative systems not to eliminate

In framing these aims we underscoreour intention

empirical

contributionto an areain which

in the

to offer an

significantconceptualprogress has been made

past decade (see especially

Achrol, Reve,

and Stern

1985). We

1983; Anderson 1985; Arndt 1979, 1983; Coughlan

1985; Frazier 1983a; Jeulandand

1984; Stern and Reve 1980; Williamson

Shugan 1983; John

thereforeestablish precedents andbenchmarksin clas-

sificationand

for richer theorizing and followup testing.

operation,

that

and provide inductivefodder

very

little

empirical

research

We

emphasize

has addressedhow relationalformsaffect channel po-

liticalandeconomic processes(Stern andReve 1980).

Brown's (1981) cross-channel

plier-retailer relations,Etgar's (1974, 1976) study

property and

(1980; Reve and

casualty insurancedistribution, andReve's

of

comparison of sup-

Stern 1986) study of interorganiza-

the in-

transactionalform per Ster

tionalstructureand sentimentsin Norway are key ex-

ceptions. Brown's explorative work showed

tensity of channel relationships to vary with the degree

of

and

ren's (1967) typology.

of War-

"formalization," that is,

El-Ansary's (1982, p. 357) adaptation

differences in

Conventional systems had less

standardization, "defini-

intense communicationsthan contractualand corpo-

rate systems. Otherwise, there were no statistically

significant

tional

operation.

reciprocity"(i.e., low centralization), and co-

Using a sample with more intraform homogeneity

support for hypothesized

thanBrown's, Etgar found

efficiencyadvantages("streamlinedintrasystem activ-

ities and decisions")

writing systems over commission-compensated con-

for centrally coordinated,

as

"independent

however, that

direct

tractors, better

known

insurance

there is much

more empirical work ahead. The advantages of cen-

trally

liance on relativelysimple

ordination may be less

coordinated systems may obtain only from a re-

agents."Etgarcautions,

products; "centralizedco-

suitable for

systems which

Indeed, ef-

handle complex

.

.

. products" (p. 23).

come as a

ficiency gains may

dardizationand reducedcustomerservices.

consequence of stan-

Reve tested competing

theories for the effects of

verticalcoordinationon transactionclimate. Report-

ing mixed empirical

institutional economics (Williamson 1975,

and sociopolitical theory (especially Pfeffer and

Salancik's 1978 resource

complementaryexplanations of interorganizational re-

lations. His study provides important direction for

the structureof interfirmrelationsand their

studying implications for economic conduct (Reve and Stern 1986, p. 99).

evidence, Reve suggests that

1985)

dependence model) are

Specific Aims

We examine channel structuresand strategies in

hardware industry.Though a significant

domestic economy(40,000

volumefrommorethan200 full-linewholesalers), the

hardwarechannel was selected because of its well- documentedchannelstructuresanddominance by three

structuralforms:dealer cooperative(e.g.,

wholesaler voluntary(e.g.,

pendent. Firstwe use a transactioncost

model for the effects of inter-

the

the

sector in

dealers

doing

$8 billion in

Cotter,Ace),

Sentry,

Pro), and inde-

perspective to

develop a conceptual

organizational formon key or bureaucratization.We

expectations for each interorganizational form to em-

phasize unique strategies kets.

facetsof decision structure

then offer more explorative

for serving end-user mar-

Conceptual Framework

The marketing channelis an interfirm system whose

members,by an

roles, areinvolvedin the process of

availablefor consumption.

opment in the study of marketing channels has been

the introductionof

(Williamson1975). TCA blends organizationtheory,

economics, andmodemcontractlaw to offer

explanations of why

by managers and

change. TCA has been summarizedin the marketing lit- erature (cf. John 1984;Ruekert,Walker, and Roering

1985). It pays particular attentionto the costs of run-

ning the system-negotiating,

tion, and monitoringperformance.Very briefly, TCA

presumes that market-based exchange is

preferred on flexibility and efficiency grounds.

ever, whentherearefew candidate exchangepartners, inevitable"frictions" may cause market exchange to

"fail,"especially if innatelyimperfect decision mak-

ers face transactionenvironmentsthat are complex/

uncertain.The

erfulframeworkis opportunism, "self-interest seeking

with guile" (Williamson1975, p. distortion of

information and

shirking of obligations,opportunism is consideredan

inherenthuman tendency which, market structuresor

competitive

performance in predictable environments, will cripple

market-based exchange. Relational contracting or ver-

tical integrationmay

ganizing

of

exchange of outputs and negotiated

making a product

A

relatively recent devel-

transactioncost analysis (TCA)

powerful

differentstructuresare selected

theirenvironmentto coordinateex-

assembling

informa-

generally

How-

key variable in this seemingly pow-

26).

unless bridled by measurable

highly

be more efficient forms of or-

Exemplified by

for the transactionif the inherentfixed costs

the exchange system can be amortized

governing

over frequent transactions. To augment andsubstitutefor

gaining over

negotiation andbar-

termsof sale in market exchange, inter-

nal and relationalforms rely on administrative, bu-

member

reaucratic

arrangements behaviors.Several dimensions of bureaucraticstruc-

turehave receivedattentionin the

organizationtheory

literature as critical

innova-

tion,highmorale, andtheir counterparts(cf. Daft 1986; Marrett 1971; Scott 1981). Reve (1980, Ch. 3; John and Reve 1982) com-

mechanismsof decision-makingefficiency,

variation, essential

to

coordinate

of

planes

bines perspectives offered by

and Warren (1972) to establish central-

Ven (1976),

ization, formalization, and interactionsas "the key structuralfeaturesof an interorganizational relation-

ship"(John and Reve 1982, p.

implement

decisions

Formalization

is "the

Centralization

Aldrich (1979), Van de

518).

to makeand

is concen-

are formu-

for example,

degree to

which power

within the dyadic relationship

(p.

518).

of a system

tratedat one vertical level" is the extent to which norms

(Scott 1981, p. 95-9),

lated explicitly

written

contracts. We narrowReve's "interactions"dimension, which

via rules, coded behaviors, and emphasis on

emphasizes "task related flows

sources,

precise concept of participation. Following

(1972) and Hall (1977),

degree of input to decisions, including ideas genera-

tion, decision-making involvement, and goal

lation.

variables"describethe extent to which a marketre-

lationship has been replacedby

bureaucratic relationship"(John and Reve 1982, p.

518).

Characterizing Complex Systems

Though the world of channel systems is difficult to

organize into Williamson's dichotomy of "marketsand

hierarchies," the parsimony of TCA and its clarity for

polar predictions

evaluating intermediateforms. We begin by attempt-

ing

analysis

is the relationship between

portant (highest percentage of purchases) full-line wholesaler.

marketor internalized exchange. Ourunit

the

of

activities, re-

andinformation" (p. 31), in favorof the more

Duncan

we view

participation as the

formu-

Together,

in three-dimensional space, these

an administrativeor

make it a useful startingpoint for

to orderthe threechannel forms characteristicof

hardware industryaccording to their proximity to

of

a retailerand its most im-

Independent.

The form closest to market-based

exchange

is probably the associationbetween retailers

'Though these three facets seem central to understanding

roles in relational analysis

different professional

uniformly trained

that compares systems

vertical

dimensions of

have

with dif-

cooperation and program control in our study, other

bureaucratizationsuch as specialization and professionalismmight

important

ferent numbersof levels (intermediaries) and

driven by

contrasts

which a priori might be

"cultures" (e.g., Palamountain 1955

pharmacists with other retailer coalitions).

TransactionCost Perspective on VerticalContractualStructure /

23

and independent wholesalers. Though it was the back-

boneof

indepen- to whole-

sale buyinggroups for consolidated purchasingpower.

local distrib-

about half the field's dents.About80 of the

1960s. Todaythey do businesswith two of every three

hardwarestores and home centers in the country.

aside from

the two

ownershipdifferences,

same basic way. Members agree to purchase a sub-

stantial portion of their merchandisefrom the orga-

nization. They also agree to standardizecertain op-

showcase their affiliation

eratingprocedures and to

through store signage and advertising. These system-

of the stores a store's logo

for consumers.That

wide

hardware wholesaling 50 yearsago, todayonly

210

wholesalers are

independentsbelong

Stem and El-Ansary(1982) note that,

systems operate in the

The others are primarilysmall, mostly utors.

Though to theirretailers

exceptions,

Stem and

these wholesalers offer some store

(e.g.,

have been

declining

pro-

merchandising aids and

in num-

grams circular programs),they

berand losing retailcustomersfor several

few

harshassessmentof wholesaling seems germane.

programsclarify the total offering

Technically,

is, by recognizing

years. With

El-Ansary's(1982, p. 127)

general, much more

consumersshould understandthe

emphasis,productmix,

dealer's marketing

and service orientation.

the ownership differencebetweenthe

[W]holesaling firms are, in

distinction in the marketsand

two systems is a key

hierarchiesframework.That is, dealer coops are in-

Whenretailers join the system they purchase a minimumvolume of mer-

tegratedsystems.

not only agree to

chandise, but also buy

Ace re-

some

quires dealersto invest $5000, Cotter $1000, and HWI

$800 and to make end-of-year stock purchases with annualrebates.Most dealer purchases areon a simple

cost-plus basis, with typically no

markupcategories.Thoughprogramparticipation can

vary and purchase

tuate, reflecting

investment, vested sup-

plier-based store identity, and end-of-year

purchases

riers.

exit barriersand backwardinte-

congru-

grationimplicitly shouldtend to supportgoal

ence in dealer cooperatives(Ouchi 1980). Hence, the

formal structuring of joint

efficient execution of programs that have ambiguous

performance measuresand

portunism. Also, the auditability

preoccupied with logistical functions than they are

with penetrating markets. The result of this preoc-

cupation has meant that they have been severely out

of position as majorchanges have taken place within

certain industries

"shares"in the cooperative,

Characterized by limited

interdependence and the

voting, some nonvoting. For example,

relatively shallowconcernof each memberfor system

performance, the independentsapproximate market-

relationship is unlikely

to be secured by an administrative hierarchy, a for-

malized system of procedures and rules, or an insti-

tutionalizedconsultativestructure.Instead, the system

works on a more organic basis; authority is shared,

programs are negotiated,

relationship is

bureaucratization developed before, we expect low

centralizationand formalizationand relatively high

levels of

scope of the exchange

relationship is likely to be narrowerthan what could

be supportedefficiently with more vertically inte- gratedsystems (Anderson and Weitz 1986).

terms of trade. The scale or

based exchange. The channel

more than a few

fluc-

volume and

consistency can

the essence of recurrent spot con-

rebateson

and decision making for the

dimensions of

tracts, the moderate up-front

collaborative. On the

combine to make for nontrivial exit bar-

participation as a result of bargaining over

Relatively high

stakes should enable more

would otherwise risk

of

op-

is

performance

Cooperative and wholesaler voluntary. Two larger

of wholesalers represent a higher form of

independents.By

categories

vertical coordinationthan the

listing a numberof independently owned dealersin a

voluntarygroup (hereafter called

taries"), a wholesalercan deliver goods and support

andeffectiveness than

the dealers acting on their

sored cooperative (hereafter called

but the initiativecomes from retailersratherthan the

"coop") is similar,

own. The retailer-spon-

serviceswith greaterefficiency

enhanced by ownership

via rationalized reporting and

en- control systems.

However,

plicate

if we considerthe historicalaccountsof

as well as a source

the wholesaler. Thus, these presumed

advantages of reduced intrasystem

command

by akin to inter-

"wholesalervolun-

these two channel systems, the literature tends to im-

as the more hi- the wholesaler is

erarchically administered. That is,

wholesaler voluntary groups

authority and expertise,

a locus of

of leadership. Memberretailers relinquish some of their

autonomy to

facetsof the wholesaler voluntarysystem tendto pro-

vide the efficiency

haggling (Williamson1985, Ch. 3) that make it nalizationin some respects.

is dif-

fused throughout the retail membership and therefore

role specification

sources are more difficult to accomplish"(Stem and

distributor.This retailer-organized and owned whole-

sale

servicesfor its members.

about65 wholesalers participate in one of

company is operateddemocratically to perform

Though

dealer

coops ac-

The latter have evolved from playing a

in the industry(less than 20%) in the early

and a latitude of decisions

the

Auto, Sentry, Trustworthy,Pro),

count for nearly 48% of wholesale sales (Hardware

Age 1984).

major wholesaler voluntarygroups(e.g., Western

In a retailer-sponsoredcooperative "power

and concomitant allocation of re-

minorrole

24 / Journalof Marketing,April 1988

El-Ansary1982, p. 325). Thoughtechnically

is an

years ago Hollander

(1968, p.

with manufacturers, diffused power

seemed to preclude thematiccoordinationof market-

ing programs in dealer coops.

a

coop

integratedsystem, ownership

is open to dealers

who purchase shares. Twenty

133) speculated that despite buying power

anddiverse

goals

turnover can be high. Coordination is

difficult. Investment in manufacturing or processing

Membership

may be hazardous and inadvisable, since acceptance

Financing prob-

have

little interest in financing

location difficulties tend

al-

of the

lems may inhibit growth, since the members

output may

not be assured.

may

.Risk

.

to deter experimentation.

We caution,however, thatmuchhas changed since

Hollander'sobservations.

Membership in the coops

has increasedfrom 8200 in 1969 to more than 20,000 today. Their growth is creditedto several factors by

industry observers.

The dealer-ownedfirmshave a

is their major pitch to the still-unaffiliated

price advantage-which

indepen-

.

. Today they're not just wholesalers: they're

paint

own windows, etc. They're major na-

They're professional management

manufacturers:Cotter and Ace

.

tional advertisers.

make their own

.

. HWI its

consultants (Hardware Age 1984, p. 60).

Hence, mounting evidence suggests that dealer

their apparent structural

plausible

coops have been overcoming

limitationsas catalogedby Hollander.It is

that, muchlike relatively autonomous faculty at a public

university, membersof a dealer coop will administer

themselves through reasonedcodes of conduct. Though

dealer coops arerun by well-paid,experiencedprofes-

sionals (cf. Cory 1983), these managers work within

with explicit

norms, and are approvedby the membership.Thus,

in a sense, coops seem to have opted for a rule-by- law ratherthan a

tion andcentralizationhave been framed

collinearfacets of bureaucratization (cf. John 1984),

but we should

stitutability in transactional governance(cf. Daft 1986,

a formalized administrativestructure,

rule-by-manhierarchy. Formaliza-

implicitly as

more considerationto their sub-

give

Ch. 5; Reve and Ster

1986, p. 92-7).2

Furthermore,cooperatives have strategic advan-

inherentin their ownership structure.Because

wholesaler-retailer goal

mit-

relativelyhigher levels of

tages

of

congruence and performanceauditability,coops

2AsReve (1980, p.

253) notes, interpretations of Williamson's (1975)

boundaries applied to "markets"

in our study, preservation

"integrated"

solidify

transactional forms into

of contractual relations plainly in-

on the

frameworkdiffer in the definitional

and "hierarchies." As we aim to

of the dichotomy by grouping

and "all other" or "spot market" and "all other" distorts key distinc-

tions within groups. "The study

volves more than an examination of discrete markets on the one hand

greater attention to

and hierarchical organization

transactionsof the middle

ing of complex economic organization" (Williamson 1985, p. 83-4).

range will help to illuminate an understand-

igate opportunistic tendencies and therefore support

investmentin

paints and windows illustratesuch (upstream) trans-

down-

in insur-

anceallowseffectivemass merchandising(Etgar1976),

the dealer cooperativesupports

wholesaler. In both in-

point of sale typically

stances,

regional

stream

action-specific investments supportedby coops'

transaction-specific assets. Privatelabel

cohesion. Just as direct

writing

nationaland

goal

advertising efforts by the

seller influence at the

can overridemass media messages (Porter1974); in-

dependents could rely on the advertising for traffic

generation, then opportunistically switchcustomersto

competingsuppliers'products.

We do

not suggest

that coops were formedto mit-

forms are the

environmentalselec-

igateopportunism.Today's governance

net of

priorstrategic choice and

tion. That is, fit of business

organizational survivalflows from the

or

conduct and market contingencies

otherenvironmental imperatives.Early motivationsfor

independent retailersto join a coop

untary included improved retail practices(accounting,

merchandise

(leaders and privatelabels)(Duncan and Phillips1946),

and "escaping chain store

services

1939). Seem- model of evo-

lutionarychange in channelsis the fact that coops and

voluntaries have achieved cost efficient functional

structuresand significant control and integration of

activities. It is from their

current stage of evolution that we posit their bureau-

craticstructureand competitive advantages(cf. Nel-

son and Winter 1982 on the evolution of distinctive

skills).

wholesaling and retailing

ingly consistentwith Guiltinan's (1974)

to customers" (Stewart and Dewhurst

greaterflexibility in prices, hoursand special

taxation and maintaining

buying, display, advertising), lower cost

or wholesalervol-

Internal Administration Hypotheses

The preceding deductive reasoning from TCA (Wil-

liamson

descriptive accounts

literature supports the following hypotheses.

H1: Wholesalervoluntariesexhibit the high-

1975) coupled

with limited induction from

in the trade press and channels

est,

independents the lowest,

cooperativesmidrange levels

ized wholesaler

authority.

and dealer

of central-

H2: Dealer cooperative groups the

exhibit the

highest, independents

wholesalervoluntaries midrange levels of

and

lowest,

formalizationin wholesaler-dealerrela- tions.

H3: Wholesaler-sponsoredgroups exhibit the

and

levels of

lowest, independents

dealer cooperatives midrange

wholesaler-dealer sion making.

the

highest,

participation in deci-

TransactionCost Perspective on VerticalContractualStructure / 25

Hypothesized

Implications of Interorganizational Form

We expect the threechannelstructuresto play out dif- of

management is an integratedorganizationalemphasis

strategic

ferent marketingstrategies. "Theessence

Competitive Strategy

on

securing and sustaining a competitive advantage withinthe marketsserved by individualbusinessunits"

(Day and Wensley 1983,

p. 80). Given a set of dis-

tinctive

competencies

arising from their relationship

emphasize those

and differen-

withthe supplier, retailmemberswill

advantages in their marketingprogramming.Key ele-

mentsof retail strategy include store location, target

market selection, low price emphasis,

tiationon such facets as assortment,service, consis-

tency, and quality (cf. Arnold, Capella, and Smith

1983). Assuming that locational strategies are gen-

erally outside

tions, we considerchannelstructure consequences on

the other key strategic variables. Governancestructuresarebelieved to vary

in their

ability to support different types of strategic aims (cf.

Ruekert,Walker, and Roering 1985). Hence, it may be possible to match structureswith strategy in dis-

criminatingways.

gional

their relatively low capability for vertical coordina-

orientationof the independent wholesalersand

the purview of retailer-supplier rela-

Because of the small size and re-

be to em- the bound-

tion, the independents' best strategy would

phasizetarget marketselection,

ary

of information exchange, negotiation, and monitor-

Thus, theirdealersseem likely to concentrateon

ing.

small marketniches and

ularservice or product line.

lackthe buyingpower

low price strategy.

not seem

comprehensiveprogramming aids. Thus, largely be- cause of the lack of alternative strategic courses, we

suspect independents

marketsor

wide assortmentsand

specifying

of the

product marketfor maximizing

efficiencies

perhapsemphasize a partic-

Independent wholesalers

to support theircustomersin a

The structureof the relationdoes

capable of providing

will opt to focus on specific

merchandise categories.

In comparison with the other

levels

structures, whole- are likely to achieve

commitmentsto fol-

Never-

saler-sponsoredvoluntarygroups

of merchandising coordinationand pro-

high

centralized authoritypro-

retailershave

providedrelatively open-ended low wholesaler programs. This relationalfacetcan lead

to greaterstrategicemphasis on store format, private

labels, and adaptivepromotionalcampaigns.

motional harmony. Their

vides an adaptabilityadvantage in that

these commitmentsmust be enforced by the

theless,

wholesalerand hence confined to activities that are

readily measured by inspectingoutput

effort. Finally, the buying power of wholesaler vol-

untariesis generally less thanthatof the cooperatives

or monitoring

(HardwareAge

1984). This fact plus theirvested au-

26 / Journalof Marketing,April 1988

thority to differentiateon thebasesoutlinedbeforemake wholesalervoluntaries unlikely to emphasize a low price strategy. Dealer cooperatives seem capable of gaining a

competitiveadvantageby following two principal

numbersof members

and

sale buyingpower

make possible a low price strategy. However, there

aretwo reasonsnot to expect

becausethe tradehas an inherentassortmentand ser-

vice emphasis, large

price strategy. Second and

unlikely to pursue the low

members are

re-

tail strategies. First, the large

sophisticated distribution system support whole-

with manufacturers, which

should

such an emphasis. First,

numbersof

coop

strategy

against the variety of intertype

fact thatsuch a

perhaps more important is the

is difficult to sustain

competitors suchas discounthouses, mailorder chains,

and warehousehome centers.

A second approach is more promising and consis-

tentwith coops' structural advantages over competing

channel systems. Dealer coops

commitmentsbased on

cause the ownership structureof

tem mitigates opportunism in the channel,

have a competitiveadvantage in mass merchandising

and differentiation through

strategic actions yield higherpayoffs otherchannel structuresbecause of

monitoring costs and goal

flected in diminishedfree-ridertendencies. Thus, ex-

are able to extract

it should

goal congruence. That is, be-

the cooperativesys-

store advertising. These

to

coops

thanto

inherently lower

congruence benefits re-

ecuted with less "friction," wholesaler affiliation in advertising and use of sales promotions are the dif-

ferentiatingstrategies

summarize, for interorganizational form ef-

we offer the following ex-

of

coop

retailers.

To

fects on retail strategies, In

plorativehypotheses. nel forms:

comparison with other chan-

H4: Independents have a greater inclinationto

target

size specialtyproducts in theirretailstrat-

egy.

H5: Affiliates of wholesaler-sponsoredgroups

promotional

specials, assortment, and service. H6: Membersof dealer cooperativesempha- size affiliationin advertising.

and serve small marketsor empha-

differentiateon the basis of

Method

Our sample frame was the Yellow Pages listings of

hardwarestoresin 10 cities

graphic areasacrossthe UnitedStates.

piled

and deleted stores in

two large

Fromourcom-

representing a mix of geo-

list of

approximately 400 stores, we eliminated

branchstoresof local multiples

cities on an nth-name,systematic basis to

ensurea maximumof 40 stores per city.

Then we telephoned 212 hardwareretailersto (1)

forthcoming mail survey. not answer or had dis-

participate. Hence

questionnaires to 186 dealers.

Dealer informantswere

given a $5 token of ap-

interested respondents,

resultswas prom-

a return envelope was

mistargeted re-

left the hardwarebusi-

five weeks after mailout was

verifyaddresses,(2) identify the manager of each store,

and (3) call attentionto the Of the 212 retailers, 19 did

connected numbers, four had gone out of the hard-

ware business, andthreerefusedto

we mailed

preciation for their anticipatedcompletion of the mul-

tipagequestionnaire. A summary of

ised to

provided, and a reminder postcard was sent 10 days

aftermailout.With adjustment for one

cipient who reportedhaving

ness, our response rate 71% (133/185).

BureaucratizationMeasures

Indicantsof bureaucratic structuring have evolved sig- nificantly in the last decade of channels research.

organizationtheory,

John (1984) developed

Drawing from empirical

measuresof centralizationand

formalizationthat evidenced

Inspiredby his scales, we customized 5-item mea-

suresof centralizationand formalizationin the hard-

ware industry. Centralizationitems includedthe need

for permission, freedom to make program adapta-

tions, andthe

work in

reliability and validity.

force of supplier recommendationsand

Formalizationincludedstandardized pro-

suggestions.

cedures,specifiedresponsibilities, relianceon written

contracts, and order policy.

Participation had more eclectic origins(Dwyer and

Welsh 1985;

JohnandMartin 1984;Spekman andSter

1979). Therefore our instrumentcontained "extra"

participation item candidates,

ment of ideas, opinions, and suggestions; involve-

ment;consultation; and appreciation of concerns.

LISREL X2 sta- in

including encourage-

items,

a

single-factor

Table 1

gives

illustrative

tistics showing item convergence

andcorrelations among

the item-summedvariables.The scales show conver-

gence termsof reverse-scaleditems to correlatein the

par-

ticipationscale, as previous

another study revealeda contextualor methods bias;

wordeditems were not the reverseof their

and

Anderson 1984). Alpha coefficients are sufficient for

our

variablesand an acceptable measurementmodel in-
dicatediscrimination.

model, reliabilitycoefficients,

with minimalitem deletion. We allow the error

work with the scales in

Gerbing

negatively positively wordeditem counterparts(cf.

purposes. Low

correlations among item-summed

Business Strategy Measures We used Porter's (1980, Ch. 2) genericstrategies

common typologies of retail strategies (cf. Arnold, Capella, andSmith 1983;McCarthy andPerreault 1984,

p. 403)

statementsof

With constraintson

overall

and

to develop a 12-item battery of Likert scale

strategicemphasis. we includedthree items

questionnairelength,

by

othermeansof "differentiating"

each for whatwe held as the two most easily grasped

strategies,(1) market niching and (2) low cost (price) emphasis. We developed two items we anticipated

would measuredealers' reliance on sales service

store personnel. Finally, four items were developed

to measurethe use of

the store: assortment, wholesaler affiliation, promo-

TABLE1

Summary of Content, Internal Consistency, and Discrimination of Bureaucratization Measuresa

Scale

No. of

Items

Initial/Final

Centralization

5/4

Formalization

5/4

Participation

7/7

Sample Item

For many

this store

recommendations of my

primary w/s My relationship with

w/s

by written contracts

facets of running

I

yield

to

our

is

governed primarily

My ideas

for ordering,

selling, and servicing are

welcomed by our primary

w/s

x2 (d.f.)

1.13(2),

.60(2),

p =

p =

.57

.74

15.32(11),b p = .17

Correlations Among Item-Summed Scales

Alpha Centralization Formalization

.72

.68

.79

.33*

.13

.01</