An Examination of the Nature of Trust in Buyer-Seller Relationships

Author(s): Patricia M. Doney and Joseph P. Cannon

Source: Journal of Marketing, Vol. 61, No. 2 (Apr., 1997), pp. 35-51
Published by: American Marketing Association
Stable URL: http://www.jstor.org/stable/1251829 .
Accessed: 24/09/2013 11:48
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
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 [email protected].

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

http://www.jstor.org

This content downloaded from 192.236.36.29 on Tue, 24 Sep 2013 11:48:26 AM

All use subject to JSTOR Terms and Conditions

PatriciaM.Doney and Joseph P.Cannon

An
Trust

Examination
of the
Nature
of
in Buyer-Seller
Relationship

The authors integrate theory developed in several disciplines to determine five cognitive processes through which
industrial buyers can develop trust of a supplier firm and its salesperson. These processes provide a theoretical

frameworkused to identifyantecedents of trust.The authors also examine the impact of supplierfirmand salesperson trust on a buying firm'scurrentsupplier choice and future purchase intentions.The theoretical model is
tested on data collected from more than 200 purchasingmanagers. The authors find that several variables influence the developmentof supplierfirmand salesperson trust.Trustof the supplierfirmand trustof the salesperson
(operatingindirectlythroughsupplierfirmtrust) influence a buyer'santicipatedfutureinteractionwiththe supplier.
However,after controllingfor previousexperience and supplierperformance,neithertrustof the selling firmnor its
salesperson influencethe currentsupplierselection decision.

apidly changing competitive environmentsare forcing business marketingfirms to seek more creative
and flexible means for meeting competition. Many
firms have responded to these challenges by building collaborative relationships with customers and suppliers
(Dertouzos, Lester, and Solow 1989). Such collaborative
relationshipsrely on relationalforms of exchange characterized by high levels of trust (Dwyer, Schurr,and Oh 1987;
Morganand Hunt 1994). The high levels of trustcharacteristic of relational exchange enable parties to focus on the
long-termbenefits of the relationship(Ganesan 1994), ultimately enhancingcompetitiveness and reducing transaction
costs (Noordewier,John, and Nevin 1990).
As business marketersplaced greateremphasis on building long-termrelationships,trusthas assumed a centralrole
in the developmentof marketingtheory (Dwyer, Schurr,and
Oh 1987; Morganand Hunt 1994) and practice (Dertouzos,
Lester, and Solow 1989). Marketingresearch on trust primarily focuses on two targets of trust: supplier firms and
their salespeople. Trustof a supplierfirm and trustof a supplier's salesperson, though related, representdifferent concepts. For example, a long-term relationshipwith a trusted
suppliercould be jeopardized by a company representative
who proves to be dishonest and unreliable (e.g., Kelly and
Schine 1992). Conversely, highly trusted salespeople can
preserve customer commitment during difficult times created by managementpolicies thatappearcontraryto the customer's best interests(e.g., Schiller 1992).
R

M.Doneyis Assistant
Patricia
Professor
ofMarketing,
Florida
Atlantic
Uniis Assistant
Professor
of Marketing,
Goizueta
JosephP.Cannon
versity.
BusinessSchool,Emory
Theauthors
contributed
tothis
University.
equally
was determined
The
research,andthe orderof authorship
randomly.
authorsthankthreeanonymous
JMreviewers
andMichael
Mullen
for
comments
on previous
versionsof thearticle,andChristy
Edwards
and
JohnHobbsfor assistancewithdata collection.
The secondauthor
the supportof a GoizuetaBusinessSchoolSummer
acknowledges
Research
Grant.

Interorganizationaltrustoperatesas a governancemechanism (Bradachand Eccles 1989; Heide 1994) thatmitigates

opportunismin exchange contexts characterizedby uncertainty and dependence(Pfeffer and Salancik 1978). In a distributionchannels context, researchhas shown that a downstream channel member that trusts its supplier exhibits
higher levels of cooperation (Morgan and Hunt 1994) and
exerts more effort on behalf of a principal (Anderson,
Lodish, and Weitz 1987). Trust in a supplier also reduces
conflict and enhances channel membersatisfaction(Anderson and Narus 1990). Finally,a firm thattrustsits supplieris
more committed to and intends to stay in the relationship
(Anderson and Weitz 1989; Morganand Hunt 1994).
Persons and organizations also can develop trust in a
supplierfirm's salesperson.The sales force often plays a key
role in interfacing with customers and implementing marketing strategy.At a basic level, salespeople persuadecustomers to purchasetheir firm's products.However, as firms
actively seek more collaborative relationships with customers, salespeople performan importantfunction in facilitating and developing customer trust (Swan and Nolan
1985). Research has shown that informationprovided by a
trustedpartyis used more and thus providesgreatervalue to
the recipient (Moorman, Zaltman, and Deshpande 1992).
Organizationalbuyers who trust salespeople exhibit more
integrativebargainingstrategies, which lead to benefits for
both parties (Schurrand Ozanne 1985).
Although trustcan be engenderedin a supplierfirm and
its salesperson, the existing marketingresearch focuses on
one or the othertarget,not both.Andersonand Narus (1990)
suggest that trustof an individualdiffers in naturefrom that
of an organization.Understandingsuch differences is particularlyimportantin business marketingsituationsin which
the sales force plays a key role in implementing the supplier's marketing strategies and managing customer relationships.Yet surprisingly,our review of the extant literature
found only two studies (Swan and Nolan 1985; Young and
Wilkinson 1989) that even discuss such issues, neither pro-

Journal of Marketing
Vol. 61 (April 1997), 35-51

Trustin Buyer-SellerRelationships/35

This content downloaded from 192.236.36.29 on Tue, 24 Sep 2013 11:48:26 AM

All use subject to JSTOR Terms and Conditions

viding empirical tests of similarities and differences across

the two targetsof trust.
Given the emphasis placed on trust in commercial
exchange relationships, it is also importantto understand
how trust influences specific aspects of customer behavior.
It is now well established that trust supportsexchange and
helps partnersproject their exchange relationshipsinto the
future. However, the existing marketingresearchpays little
attention to the impact of trust on current purchase decisions. In fact, we found no studies that explicitly examine
how trust affects supplierselection decisions.
The purpose of this study is to provide new insight into
how trust develops and how it influences industrialbuying
behavior. We offer several contributions to the emerging
marketingliteratureon trust.First,drawingon theorydeveloped in social psychology, sociology, economics, and marketing, we isolate five cognitive processes through which
trust can be built. These processes provide the theoretical
frameworkused to identify a set of factors hypothesized to
influence the development of trust. Second, we evaluate
both trust of a supplier firm and its salesperson. Third, we
examine the impactof truston the buying firm'scurrentsupplier selection decision, as well as its futurepurchaseintentions. Finally, we provide a richerunderstandingof trustby
placing both targets of trust in a largernomological net of
constructs.
In the literature review that follows, we discuss the
nature of trust, its definition, and the five cognitive
processes throughwhich trustcan develop. Building on this
review, in the subsequentsection, we develop a theoretical
model that outlines the antecedents and consequences of
trust in an industrialbuying context. In the next several sections, we describe an empirical study designed to test
hypothesized relationships, report study findings, and discuss their implications for marketingtheory and practice.
We conclude by outlining study limitations and an agenda
for furtherresearch.

LiteratureReview
Trust has received a great deal of attention in social psychology (e.g., Deutsch 1960; Lewicki and Bunker 1995;
Lindskold 1978), sociology (e.g., Lewis and Weigert 1985;
Struband Priest 1976), and economics (e.g., Dasgupta 1988;
Williamson 1991), as well as marketing(e.g., Andersonand
Weitz 1989; Dwyer, Schurr,and Oh 1987; Ganesan 1994;
Moorman, Deshpande and Zaltman 1993; Moorman, Zaltman, and Deshpande 1992). Each discipline offers unique
insights into the nature of trust, its definition, and the
processes throughwhich it develops.
Nature of Trust
Much researchexplores the importanceof trust in interpersonal dyads (e.g., Rotter 1967; Schlenker, Helm, and
Tedeschi 1973). Although some researchersdisagree about
whetherorganizationscan be targetsof trust,a large stream
of literature emphasizes that people can develop trust in
public institutions (Lewis and Weigert 1985) or organizations (Morgan and Hunt 1994), as well as individuals.
Therefore, the trust literaturesuggests that in an industrial

buying context, customers can trust the supplier firm, its

salesperson,or both.
The trust literaturealso suggests that, regardlessof the
level of analysis, trustingpartiesmust be vulnerableto some
extent for trustto become operational.In other words, decision outcomes must be uncertainand importantto the trustor
(Deutsch 1962; Moorman, Zaltman, and Deshpande 1992;
Schlenker, Helm, and Tedeschi 1973). In marketing,much
researchon trust has been conducted in the context of distributionchannels (e.g., Anderson and Narus 1990; Anderson and Weitz 1989; Morganand Hunt 1994) in which vulnerabilityis created by the high degree of interdependence
usually found in channel relationships (Gundlach and
Cadotte 1994; Kumar,Scheer, and Steenkamp1995). In typical distribution channel arrangements (i.e., manufacturer-distributoror manufacturer-retailer),switching costs
are relatively high. Therefore, this research stream emphasizes the influence of truston constructscentral to building
long-term relationships with customers, such as commitment (Dwyer, Schurr, and Oh 1987; Morgan and Hunt
1994), long-term orientation (Ganesan 1994), and propensity to stay in a relationship(Andersonand Weitz 1989).
Althoughsome buying firms have reducedtheirsupplier
base to facilitate collaboration or increase quality (Emshwiller 1991), many buyers still maintainmultiple sources of
supply. In an industrialbuying context, then, risk could be
operativeprimarilywhen purchasesituationsinvolve modified rebuys or new tasks, as opposed to straight rebuys
(Robinson, Faris, and Wind 1967). Such situations involve
greaterdecision making on the partof the buying organization, more people in the decision-making process, and
greateruncertainty(Johnstonand Lewin 1996). Vulnerability stems from the buyer's reliance on a supplierfirm and/or
its salesperson.The supplierfirm sets policies and develops
strategiesthatcan affect the buying firm's costs and quality.
The salesperson, as the primary contact with the buying
firm, provides valuable information and consultation to
members of the buying center. To make currentpurchase
decisions and long-term relational commitments, buyers
must determinethe extent to which they can trustsuppliers
and their salespeople.
Defining Trust
Drawing on literaturein social psychology (Larzelereand
Huston 1980) and marketing, we define trust as the perceived credibility and benevolence of a target of trust (cf.
Ganesan 1994; Kumar,Scheer, and Steenkamp 1995). The
first dimension of trust focuses on the objective credibility
of an exchange partner,an expectancy that the partner's
wordor writtenstatementcan be relied on (Lindskold1978).
The second dimension of trust,benevolence, is the extent to
which one partneris genuinely interestedin the other partner's welfare and motivated to seek joint gain. This definition of trust is relevant in an industrialbuying context. A
buying firm facing some degree of risk in a purchasesituation turs to a supplieror salespersonthatthe buyerbelieves
is able to performeffectively and reliably (credible) and is
interestedin the customer's best interests(benevolent).

36 / Journalof Marketing,April1997

This content downloaded from 192.236.36.29 on Tue, 24 Sep 2013 11:48:26 AM

All use subject to JSTOR Terms and Conditions

Developing Trust
The development of trust relies on the formation of a
trustor'sexpectations about the motives and behaviors of a
trustee. Because of the broad natureof trust and its varied
conceptualroots, our review of the trustliteratureuncovered
five distinct processes by which trust can develop in business relationships.
The economics literature (e.g., Dasgupta 1988;
Williamson 1991) suggests that trust primarily involves a
calculative process, as when an individual or organization
calculates the costs and/orrewardsof anotherpartycheating
or staying in the relationship(Lindskold 1978). To the extent
thatthe benefits of cheating do not exceed the costs of being
caught (factoring in the likelihood of being caught), one
partyinfers that it would be contraryto the otherparty'sbest
interest to cheat and therefore the party can be trusted
(Akerlof 1970). For example, Rao and Bergen (1992) find
that buying firms pay premiumprices to suppliersto ensure
high levels of quality. Essentially, buyers raise the costs of
cheating because suppliers caught acting "untrustworthy"
lose a streamof premiumrents from futurepurchases.
The predictionprocess of developing trustrelies on one
party's ability to forecast anotherparty'sbehavior.Because
trust requiresan assessment of the other party's credibility
and benevolence, one partymust have informationaboutthe
other party'spast behavior and promises. Repeated interaction enables the partyto interpretprioroutcomes better,providing a basis for assessing predictability. For example,
through repeatedly making promises and delivering on
them, a salesperson develops the confidence of a buying
firm (Doyle and Roth 1992; Swan and Nolan 1985). Extending this line of reasoning, Lewicki and Bunker(1995) suggest that predictabilityas a source of trustrequiresnot only
repeated interaction, but also courtship (see also Shapiro,
Sheppard, and Cheraskin 1992). Courtship behavior is
directed at relationshipdevelopment-specifically, learning
more about the other party.This suggests that trust grows
when two parties share a variety of experiences, thereby
improvingeach's ability to predictthe other's behavior.
The capability process involves determining another
party's ability to meet its obligations, thereby focusing primarily on the credibilitycomponentof trust.For example, a
salespersonpromises the customer promptdelivery, despite
a supply being on allocation due to shortages.Yet if the customer doubts whether the salesperson has the clout needed
to move its orderup in the queue, the customerwill be reluctant to trustthe salesperson's word.
Trust also emerges through interpretationand assessment of the other party's motives. Using the intentionality
process, the trustorinterpretsthe target's words and behaviors and attempts to determine its intentions in exchange
(Lindskold 1978). People or groups motivated to help or
reward the perceiver will be more trusted than those suspected of harboring exploitative intentions. Inferences of
benevolent intentions also can result when two parties
develop sharedvalues or norms (Macneil 1980) that enable
one party to understandthe other partner'sobjectives and
goals better(i.e., what drives their behavior).

Finally,trustcan develop througha transferenceprocess.

Struband Priest (1976, p. 399) describe the "extensionpattern"of gaining trust as using a "thirdparty'sdefinition of
This
anotheras a basis for definingthatotheras trustworthy."
trusted
one
from
be
transferred
can
trust
that
"proof
suggests
source"to anotherpersonor groupwith which the trustorhas
little or no direct experience (Milliman and Fugate 1988;
Struband Priest 1976). For example, a new salespersonrepresenting a highly trusted firm would benefit from the
buyer's past experience with the supplier firm. Conversely,
distrustalso can be transferredif a person lacks other information. For example, a person's past experience with automobile or insurance salespeople often results in an initial
state of distrust(Adkins and Swan 1980-81).
Clearly, some factors can invoke multiple trust-building
processes. For example, frequentcontact with a supplier's
salespersoncan invoke the predictionprocess by helping the
buyer more accurately predict the salesperson's behavior.
Or, the buyer could interpretfrequentcontact as an indication of the salesperson's genuine interest in the buying
firm's welfare, thereby invoking the intentionalityprocess.
Therefore, each process represents a different manner in
which subjective probabilityjudgments of a partner'strustworthinesscan be made.

Research Hypotheses
The five trust-buildingprocesses provide insight into how
trustof a selling firm and salespersonare established in the
industrialbuyer-suppliercontext of this research.As the primary suppliercontact, the salespersonrepresentsan important personal source of trust for the buying firm. The supplier firm provides an alternativesource of trustthroughits
policies, actions, and personnel.Because of the personaland
impersonalnatureof these two targets of trust, the manner
throughwhich each fosters trust is likely to differ. In Table
1, we summarizethe five trust-buildingprocesses and link
them to factors hypothesized to develop supplier firm and
salesperson trust.Although each antecedentfactor can trigger multiple trust-buildingprocesses and one process can
triggeranother,we focus on the primaryprocesses likely to
be invokedby each antecedentfactor.
In Figure 1, we offer a conceptual overview of the
empiricalstudy.Hypothesizedantecedentsto trustrepresent
characteristicsof the supplierfirm and its salesperson, and
aspects of the buying firm's relationshipwith each, which
provide a basis for the buying firm to infer trust.The conceptual model also includes key outcomes of trust-the buying firm's currentsupplierselection decision and anticipated
future interaction with the supplier. We detail the logic
underlyingour theoreticalmodel in the following sections.
Developing Trust in a Supplier Firm:
Characteristics of the Supplier
Supplierreputation.We define supplier reputationas the
extentto which firmsandpeople in the industrybelieve a supplier is honestandconcernedaboutits customers.A favorable
reputationis easily transferableacrossfirmsandenhancesthe
credibility of the vendor (Ganesan 1994). If a buying firm
assumes the supplier'sreputationis well deserved,trustwill
Trustin Buyer-SellerRelationships/37

This content downloaded from 192.236.36.29 on Tue, 24 Sep 2013 11:48:26 AM

All use subject to JSTOR Terms and Conditions

Trust-Building Processes,

TABLE 1
Generic Drivers, and Factors that Invoke Each Process
Factors that Invoke the TrustBuilding Process
*supplierfirmreputation
*supplierfirmsize
*supplierfirmwillingnessto customize
*supplierfirmconfidentialinformation
sharing
*lengthof relationshipwithsupplierfirm
*lengthof relationshipwithsalesperson

Trust-Building Process
Calculative:Trustorcalculates the costs
and/or rewardsof a target acting in an
manner
untrustworthy

Generic Driverof the Process

Costs are higherwhen a target makes
largerand/or relationship-specific
investments

Prediction:Trustordevelops confidence
that target'sbehaviorcan be predicted

Trustorlearns more about the target

throughrepeated and broader
experience

*lengthof relationshipwithsupplierfirm
*salesperson likability
*salesperson similarity
*frequentsocial contact with
salesperson
*frequentbusiness contact with
salesperson
*lengthof relationshipwithsalesperson

Trustorassesses the target's

Capability:
abilityto fulfillits promises

Evidence of the target'sabilityto fulfill

its promises

*salesperson expertise
*salesperson power

IntentionalityTrustorevaluates the
target'smotivations

Target'swords and/orbehavior
indicates concern forthe trustor

?supplierfirmwillingnessto customize
*supplierfirmconfidentialinformation
sharing
?salesperson likability
?salesperson similarity
*frequentsocial contactwithsalesperson

Identificationof trustedsources closely

Transference:Trustordrawson "proof
sources,"fromwhich trustis transferred associated withthe target
to the target

?supplierfirmreputation
?supplierfirmsize
?trustof supplierfirm
*trustof salesperson

be grantedon the basis of the supplier'shistory in relationships with otherfirms. In other words, buyersinfer the trustworthiness of a supplier throughthe words and actions of
other people and organizations.Therefore, the process of
transferencecan be used to predict a positive relationship
between supplierreputationand trustof the selling firm.
Buyers also can use the calculative process to estimate
that the costs of a supplieracting in an untrustworthymanner are quite high for firms with good reputations.Because
developinga favorablereputationinvolves significantinvestment and representsa valuableasset (Dasgupta 1988), firms
are reluctantto jeopardize their reputationby acting opportunistically (Telser 1980). Even if opportunitiesexist, any
short-termgains from untrustworthybehavior can be balanced againstthe rewardsof maintaininga good reputation.
Empiricalevidence supports the link between supplier
reputationand buyer trust. In a study of industrialchannel
dyads, Ganesan (1994) finds that a retailer'sfavorableperception of a vendor's reputationleads to increasedcredibility, one of two dimensions of trust.Similarly,Andersonand
Weitz (1989) find that a channel member's trust in a manufactureris positively relatedto the manufacturer'sreputation
for fair dealings with channel members.

trusted.First, throughthe process of transference,a buying

firm could draw on the experience of others to infer trustworthiness from supplier size. Overall size and market
share indicate that many other businesses trustthis supplier
enough to do business with it. This suggests that the supplier consistently delivers on its promises to others or it
would not have been able to maintain its position in the
industry. Second, we might expect that less trustworthy,
more opportunistic suppliers operate as fly-by-night organizations and would be unable to build sales volume or
marketshare (Hill 1990). As a result, buyers could use the
calculative process to determine that larger suppliers incur
more significant costs throughuntrustworthybehaviorthan
do smaller suppliers.
Formallystated,

Supplier size. Supplier size encompasses the firm's

overall size and its marketshare position. Suppliersize provides a signal to the buying firm that the selling firm can be

Supplierwillingness to customize.A suppliercould offer

to make, or actually make, idiosyncraticinvestmentsin the
relationship. Such investments might include specialized

Hi: Buyingfirmtrustin the supplierfirmis positivelyrelated

to the supplier'sreputation.
H2:Buyingfirmtrustin the supplierfirmis positivelyrelated
to the supplier'ssize.
Developing Trust in a Supplier Firm:
Characteristics of the Relationship

38 / Journalof Marketing,April1997

This content downloaded from 192.236.36.29 on Tue, 24 Sep 2013 11:48:26 AM

All use subject to JSTOR Terms and Conditions

Antecedents

and Consequences

FIGURE 1
of Trust of a Supplier Firm and Salesperson

I
r~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Characteristics of the Supplier Firm

*Reputation
*Size
J\

kl-

Characteristics of the Supplier Firm Relationship

*Willingnessto customize
*Confidentialinformationsharing
*Lengthof relationship

l^^~~~~~~~~~~-

Characteristics of the Salesperson

-Expertise
-Power

Characteristics of the Salesperson Relationship

-Likability
-Similarity
?Frequentbusiness contact
?Frequentsocial contact
?Lengthof relationship

2-1-

<^

-I

r
Control Variables
*Deliveryperformance
*Relativeprice/cost
*Product/serviceperformance
*Purchaseexperience with supplier

equipmentor adaptationof productionprocesses to meet the

buyer's needs. Williamson (1985) suggests that idiosyncratic investments change the firm's incentive structure;
firms making idiosyncratic investments are unlikely to
engage in opportunistic behavior because such behavior
threatensthe continuationof the relationship.Because idiosyncratic assets can lose substantialvalue unless the relationship is continued (Anderson and Weitz 1992), buyers
could use the calculativeprocess to estimate thatthe costs of
untrustworthybehavior are higher for suppliers with idiosyncratic investmentsat stake.
Willingness to make idiosyncraticinvestmentsalso provides evidence that a vendorcan be believed, it cares for the
relationship, and it is willing to make sacrifices (Ganesan
1994). The supplier's willingness to place itself at risk signals to the buying firm that the supplier is willing to cooperate (Lindskold 1978; Struband Priest 1976). Such behavior could invoke the process of intentionality-in which the
idiosyncratic investments provide evidence that the supplier's motives are benevolent. Empirically,Ganesan(1994)
demonstrates that buyers trust suppliers they perceive as
having made idiosyncraticinvestmentson their behalf.
Confidentialinformationsharing. Confidentialinformation sharinginvolves the extent to which suppliersshareprivate information with their customers. The calculative
process suggests that buyers will trust suppliers that share
confidential information,because such suppliers risk a sub-

stantial investment. For example, because the buyer could

disclose confidential informationto the supplier'scompetitors, the costs of this informationbeing used improperlycan
be high (cf. Kelly and Kerwin 1993). Buying firms that
believe their suppliers cannot be trustedare more likely to
behave in an untrustworthymannerthemselves, such as by
passing along this sensitive information(John 1984). As a
result,buyerscould calculate that the costs of untrustworthy
behaviorincrease as the selling firm shares more confidential informationwith them.
The extent to which a suppliersharesconfidentialinformation with the buyer also provides a signal of "good faith"
to the buying firm. The supplieruses a "disclosurepattern"
to establish trust by providing tangible evidence that the
supplier is willing to make itself vulnerable (Strub and
Priest 1976). Therefore, the process of intentionality also
could be invoked because sharing confidential information
signals to the buyer that the supplier's motives and intentions are benevolent-that is, the suppliercan be trusted.
Length of relationship. Most researchers agree that
trust develops and builds over time. Two trust-building
processes can explain why trust in the selling firm
increases with the length of the buyer-supplier relationship. First, length of time represents an investment both
parties make in the relationship. To the extent that buyers
perceive such investments on the part of suppliers, they
could calculate that a supplier would incur losses by acting
Trustin Buyer-SellerRelationships/ 39

This content downloaded from 192.236.36.29 on Tue, 24 Sep 2013 11:48:26 AM

All use subject to JSTOR Terms and Conditions

in an opportunistic (i.e., untrustworthy)manner. Second,

the process of prediction also can be invoked as a relationship grows older. When exchange relationships have a history, the outcomes of previous business episodes provide a
framework for subsequent interaction. With increased
experience, firms are more likely to have successfully
weathered critical shake-out periods in their relationship
(Dwyer, Schurr,and Oh 1987; Scanzoni 1979) and gained
a greater understanding of each other's idiosyncrasies
(Williamson 1985). Therefore, longer relationships facilitate the buyer's ability to predict the supplier's future
behavior.Empirically,Anderson and Weitz (1989) find that
a channel member's trust in a manufacturerincreases with
the age of the relationship.
Formally stated,
H3:Buyingfirmtrustin the supplierfirmis positivelyrelated
to a supplier'swillingnessto makeinvestments
specificto
a buyingfirm.
H4:Buyingfirmtrustin the supplierfirmis positivelyrelated
to more confidentialinformationsharingby a supplier
firm.
H5:Buyingfirmtrustin the supplierfirmis positivelyrelated
to the lengthof time a supplierfirmandbuyerfirmhave
beenin contact.
Developing Trust in a Salesperson:
Characteristics of the Salesperson
Expertise. Trust that is based on a partner'sexpertise
focuses on an expectancy held by a person that the partner's word or written statement can be relied on (Lindskold 1978). Through the capability process, salesperson
expertise builds a buying firm's trust by increasing its confidence that a salesperson can deliver on his or her
promises. Empirically, the role salesperson expertise plays
in fostering trust has received some attention in the marketing literature. Busch and Wilson (1976) find that customers view salespeople with higher levels of perceived
expert power as more trustworthy.In a study examining
the use of marketing research, Moorman, Deshpande, and
Zaltman (1993) find that researcherexpertise is an important foundation for trust. Finally, Crosby, Evans, and
Cowles (1990) show that the perceived expertise of an
insurancesalesperson is a significant predictorof customer
trust.
Power. Salesperson power is defined as the buying
firm's belief that the salesperson is capable of providing
buyer outcomes that match what the salesperson says or
promises.Forexample, a salespersonwho promisesto expedite an ordercould be doubted if the buyer believes that the
salesperson lacks control over the organizationalresources
needed to fulfill the promise(Swan and Nolan 1985). Therefore, power can increasetrustin the salespersonby means of
the capability process. Empirically,Moorman, Deshpande,
and Zaltman(1993) find that when researcherpower in the
organizationis higher, trustof the researcheris higher.
Formallystated,
is positively
H6:Buyingfirmtrustin thesupplier'ssalesperson
relatedto the buyer'sperceptions
thatthe salespersonhas
expertise.

is positively
H7:Buyingfirmtrustin thesupplier'ssalesperson
thatthe salespersonhas
relatedto the buyer'sperceptions
powerin the supplierfirm.
Developing Trust in the Salesperson:
Characteristics of the Relationship
Likability. Salesperson likability refers to the buyer's
assessment that people in the buying firm find the salesperson friendly, nice, and pleasantto be around.Priorresearch
in psychology (Rotter 1980) generally finds a positive relationship between a person's likability and the extent to
which the person is trustedby others. Because initial trust
depends on the buyer's confidence in predictingthe accuracy of statementsmade by the salesperson,a positive relationshipbetween trustand liking can occur as a result of the
predictionprocess. Buyers are more confident of their predictions about someone they like (Swan and Nolan 1985).
Salesperson trust also can be engenderedby means of the
intentionalityprocess: Buyers attributefavorablemotives to
those they like (Rotter 1980).
Empiricalstudies in business marketingshow that feelings of trust in the salesperson are positively relatedto liking. For example, Swan and colleagues (1988) find that likability is a distinct dimension of trust in the salesperson.
Hawes, Mast, and Swan (1989) find that buyers attribute
positive trust-earningvalue to manufacturers'representatives who were "likable."Finally, salespeople reportthat to
gain customertrust,customersmust see them as "likable"or
"a friend"(Swan, Trawick,and Silva 1985).
Similarity.Similarityassesses the buyer'sbelief that the
salespersonsharescommon interestsand values with people
in the buying firm. In goal-interdependentsituations, similarity can be a cue for expecting the other partyto facilitate
one's goals (Johnson and Johnson 1972). Therefore, similarity can trigger the intentionality and/or prediction
process. Buyers who perceive salespeople to be similar to
themselves could expect such salespeople to hold common
beliefs about what behaviors,goals, and policies are appropriate.Trustis fosteredbecause the buyerfeels betterable to
assess the salesperson's intentions-that is, buyers attribute
benevolent intentions to "similar"salespeople they believe
share their values. Understandingthe salesperson'smotivations also makes it easier for a buyerto predictthe salesperson's future behavior, thereby building trust through the
process of prediction.
Frequentcontact with the salesperson. When salespeople have frequent contact with customers for business or
social purposes,trustcan be engenderedbecause buyerscan
observe the salesperson'sbehavioracross a varietyof situations. Frequent interaction fosters trust by giving buyers
informationthat helps them predictthe salesperson'sfuture
behavior with confidence. Social interactionwith members
of the buying firm also gives the salesperson a chance to
"court"the buyer.Social settings provide an informalenvironmentconducive to enhanced informationflow, building
closer interpersonal relationships, and fostering better
understandingof mutual needs. Through the intentionality
process, frequent social interaction can engender trust,
because buyers attributebenevolent intentions to salespeo-

401 Journalof Marketing,April1997

This content downloaded from 192.236.36.29 on Tue, 24 Sep 2013 11:48:26 AM

All use subject to JSTOR Terms and Conditions

pie with whom they share common professional and social

bonds.
Empirically,the frequency with which salespeople and
customers communicate for business or social reasons has
been shown to be a key determinantof relationshipmaintenance in the insurance field (Crosby, Evans, and Cowles
1990). Although Crosby,Evans, and Cowles (1990) primarily study business contacts, they find a positive relationship
between contact intensity and salespersontrust.
Lengthof relationship.As stated previously,the process
of predictionand the calculative process could explain why
trust of the selling firm increases with the length of the
buyer-supplier relationship. The same processes are
involved when the target of trust is the salesperson. First,
buyerscan calculate thatthe salesperson'sinvestmentin the
relationshipincreases with time. Because the salesperson's
investmentwould be jeopardizedby untrustworthybehavior,
the salespersoncan be trusted.Second, longer relationships
facilitate the buyer's ability to predict the salesperson's
future behavior. Because people learn through experience
how much to trustothers (Lindskold 1978), the length of a
buyer's relationshipwith a salesperson should enhance trust
because it provides assurance that the salesperson's future
behaviorwill mirrorpast behavior.
Formallystated,
H8: Buyingfirmtrustin the supplier'ssalespersonis positivelyrelatedto perceivedlikability.
H9: Buyingfirmtrustin the supplier'ssalespersonis positivelyrelatedto perceivedsimilarity.
Hlo:Buyingfirm trustin the supplier'ssalespersonis positivelyrelatedto frequentbusinesscontact.
:
HI Buyingfirmtrustin the supplier'ssalespersonis positively relatedto frequentsocial contact.
Hi2: Buying firm trust in the supplier's salesperson is positively relatedto length of time the salesperson has called
on the buying firm.

Relationship Between Trust in the Supplier and

Trust in the Salesperson
In industrialmarkets, a salesperson's behavior is partially
attributableto the supplier firm's culture, reward systems,
and trainingprograms.Buying firms assume that the salesperson'sbehaviorreflects the supplier'svalues and attitudes.
Therefore,when a customerhas limited experience with the
supplierfirm, trustof the firm can be inferredon the basis of
perceptionsof the salesperson'strustworthiness.Essentially,
the customer's trust in the salesperson transfersto the supplier firm.
The transferenceprocess should operate in both directions. According to Strub and Priest (1976), trust is transferred from the better-knownparty to a closely associated
but less well-known group or individual.This suggests that
previousinteractionswith the supplier firm and other members of its sales force (more well-known parties) provide a
basis for inferringthe extent to which a current(new or less
well-known) salesperson can be trusted. Therefore, we
hypothesizea reciprocalcausal relationshipbetween the two
targets of trust, whereby trust in the supplier's salesperson
leads to trust in the supplierfirm and vice versa.

Hi3: Buyingfirmtrustin the salespersonis positivelyrelated

to buying firm trust in the supplierfirm.
H14:Buying firm trustin the supplierfirm is positively related
to buying firm trust in the salesperson.

Consequences

of Trust

Role of trust in supplier selection. Relatively little

researchexamines the role of trustin supplierselection decisions. The only example we found was in a consumer tele-

marketingcontext (Milliman and Fugate 1988). Conversely,

a relatively large body of work in social psychology suggests that source credibilityhas importantlinks to trustand
purchasedecision making. Source credibilityis the extent to
which a communicatoris perceived to be a source of valid
assertions (i.e., credible), and the degree of confidence in the

communicator'sintent to communicatethe assertions he or

she considers most valid (i.e., benevolent). Other research
shows that the extent to which the source is seen as interested in the target'sbest interests, as opposed to their own
(i.e., appearsbenevolent), also enhances persuasion (Choo
1964). Therefore,supplierfirms and salespeople deemed to
be trustworthy(credible and benevolent) will be considered
credible sources. Because credibility enhances persuasion,

trustedsuppliersand salespeople should be more successful

in favorably influencing the supplier selection decision than

should less trustworthysuppliersand salespeople. This prediction is supportedby the fact that highly trusted sources
induce more positive attitudestoward the ideas they advocate than do less trustworthysources (Kelman and Hovland
1953).
Hi5: Buying firm trustin a supplierfirm is positively relatedto
the buying firm's selection of that supplier in a particular
buying decision.
H16:Buying firm trustin a supplier'ssalesperson is positively
relatedto the buying firm's selection of that supplierin a
particularbuying decision.

Role of Trust in Future Purchase Intentions

The proposedlink between trust in the selling firm and the
buyer's future purchase intentions also reflects that buyers
can rely on the integrity of suppliers they trust: The supplier
will not knowingly act contrary to the buyer's best interests.
According to Ganesan (1994), trust is a necessary ingredient

for long-termorientationbecause it shifts the focus to future

conditions. Empiricalevidence supportsthe notion thattrust
of the selling firm is central to a buyer's intention to continue the exchange relationship. Morgan and Hunt (1994)
demonstrate a negative relationship between trust and
propensity to leave, which is defined as the perceived likelihood that a partner will terminate the relationship in the near
future. Anderson and Weitz (1989) find evidence that trust is
key to maintaining continuity in conventional channel relationships. Therefore, we propose the following:
H17:Buying firm trustin a supplierfirm is positively relatedto
the buying firm's anticipationof future interactionwith
the supplier.

We do not hypothesize a direct relationship between

trust of the salesperson and anticipated future interaction
Trustin Buyer-SellerRelationships/41

This content downloaded from 192.236.36.29 on Tue, 24 Sep 2013 11:48:26 AM

All use subject to JSTOR Terms and Conditions

with the supplier firm. Because salespeople frequently

change employers and territories,long-term intentions are
driven primarily by expectations about the supplier firm.
This is not to suggest that the salespersonis not importantin
the buyer's future intentions but rather that the effects of
salesperson trust operate indirectly through supplier firm
trust.
Control Variables
Although we focus on the influence of supplier and salesperson trust on current supplier selection decisions and
anticipated future interactions with the supplier, previous
research suggests that several other variables affect industrial buying decisions. Controlling for such variables provides a strongertest of the theory developed in our model.
Supplierperformance.Considerableempiricaland anecdotal evidence suggests that the primarycriteriafor current
supplier selection decisions and future purchase intentions
involve a supplier'sperformance.Our review of the organizational buying literaturerevealed three aspects of supplier
performancethatconsistently emergedas centralto the buying firm's evaluation of a supplier's product offering: (1)
delivery performance,(2) relative price/cost, and (3) product/service performance (cf. Lehmann and O'Shaugnessy
1974; Wilson 1994). Performanceon these dimensions can
change from transactionto transactionas suppliersrespond
to competitive conditions and supply capabilities. Therefore, the performance measures focus on the supplier's
product offering relative to competitors' for the focal purchase. We control for the effects of supplierperformanceon
the buying firm's current purchase choice and anticipated
future interaction.
Purchase experience with vendor. Many business purchases representrebuy and modified rebuy situations. Furthermore,many firms have reducedtheir supplierbases and
tend to do business with a limited set of suppliers. Because
a buying firm might be partialto a supplierwith which it has
more purchasing experience, we control for the effects of
priorexperience on currentpurchasechoice and anticipated
futureinteraction.
Purchase choice. A decision to select or not select the
focal supplierto supply currentneeds might influence a buying firm's long-term intentions with respect to that supplier.
Therefore, we control for the effects of current purchase
choice on anticipatedfuture interaction.

Method
Sample and Data Collection
Data on a recent purchasewere collected from a sample of
firms involved in industrial manufacturing, specifically
firms in Standard Industrial Classification codes 33-37.
Although the sample firms purchaseda variety of products,
limiting the sample enabled us to develop more grounded
measures.The sample frame comprised678 membersof the
National Association of PurchasingManagement.A threewave mailing that was based on Dillman's (1978) recommendationswas employed: (1) the entire sample was mailed

a questionnaireand letter requestingtheir participation;(2)

one week later a reminder post card was mailed; and (3)
three weeks after the initial mailing, nonrespondentswere
sent anotherletter and questionnaire.A total of 21 mailings
were returnedbecause of bad addresses or respondentsno
longer involved in purchasing decisions. This process
yielded 210 completed questionnaires (a 31% response
rate).
Three items at the end of the instrument assessed
respondents'(1) involvement in the purchasedecision, (2)
influence over the purchasedecision, and (3) confidence in
their ability to answer the surveyquestions. One respondent
indicated a low level of confidence and was removed from
the sample. Remainingresponseswere uniformlyhigh, with
mean ratings (on a seven-point scale) of 6.4, 6.1, and 6.4,
respectively.
As previously noted, purchase decisions must involve
some level of risk for trustto be operative.In industrialbuying, modified rebuy and new task situations involve more
risk and uncertaintyfor the buying firm. Therefore,respondents were asked to identify "a specific purchasedecision in
which you have recently been involved ... where more than
one supplier was seriously considered."The questionnaire
then directed respondentsto write the names or initials of
two suppliers their company considered for the purchase.
After answeringa series of measuresregardingthe purchase
situation and supply market, half of the respondentswere
asked to reply to remaining questions with respect to the
"firstsupplier"they identified,and the otherhalf reportedon
the "second supplier."The objective was to ensure variance
on one of the dependentvariables,currentpurchasechoice.
The process was successful in that respect, with 42% of
respondentsreportingon a supplier they did not select for
the focal purchase. Later in the questionnaire,respondents
were directed to indicate the name or initials of the supplier's salesperson with whom their firm had the most contact. Respondents then replied to a series of measures
involving this salesperson. Respondents were asked to
answer all survey questions on behalf of their firm.
Characteristics of the sample and purchase decision.
Respondents were primarilymale (76%) and averaged 15
years' purchasingexperience. The suppliers' primarybusiness was most frequently listed as manufacturing(62%) or
distribution (30%). Few suppliers were strangers to the
buying firms; buying firms averaged 11 years' contact with
the focal supplierand 5.5 years with its sales representative.
These values were not significantly different (p < .05)
across "selected" and "unselected" suppliers. Across all
suppliers, buying firms reported that 65% had been used
"manytimes" in the past, 23% had been used "a few times,"
and 12% had never been used before. In 24% of the sample, purchaseswere made from more thanone supplier.Use
of a competitive bidding process was reported in 94% of
purchasesituations.
Nonresponse bias. We assessed potential nonresponse
bias by comparingearly versus late respondents.More than
half (55%) of the completed questionnaireswere returned
in response to the final mailing. Early and late respondents
were compared on several dimensions: characteristics of

42 / Journalof Marketing,April1997

This content downloaded from 192.236.36.29 on Tue, 24 Sep 2013 11:48:26 AM

All use subject to JSTOR Terms and Conditions

the respondents (gender and purchasing experience), purchase situation (number of people from the buying firm
involved in the decision), and supplier relationship (age of
the relationship with the supplier and whether the focal
supplier was selected for this purchase).That there were no
statistically significant differences across early and late
respondentssuggests that nonresponse bias is not a serious
concern.
Measure Development and Validation
We developed measures following standard psychometric
scale development procedures (Anderson and Gerbing
1988; Bagozzi and Phillips 1982). We generatedmulti-item
scales on the basis of previousmeasures,a review of the relevant literature,and interviewswith marketingand purchasing personnel. Statistical proceduresused to validate measures included assessment of item and scale reliability,unidimensionality, and convergent and discriminant validity.
We assessed psychometric propertiesof the final measures
by means of confirmatoryfactor analysis proceduresusing
LISRELVIII (Joreskogand Sorbom 1993), as well as traditional methods (i.e., exploratoryfactor analysis, coefficient
alpha, and adjusteditem-to-totalcorrelations).
Given the large number of constructs and measures
employed in this study, the literatureon confirmatoryfactor
analysis suggests assessing scales by examining smaller
confirmatoryfactor models (cf. Bentler and Chou 1987).
Therefore, we based measure analyses on groupings of
related sets of measures: (1) measures of supplier trust, its
antecedents and outcomes, and control variables and (2)
measures of salesperson trust and its antecedents. Before
discussing measure analysis results, we provide a more
detailed discussion of the trustmeasures.
Trustmeasures.Although analysis of these data provide
evidence of discriminationbetween the two targets of trust
(the salespersonand the supplierfirm), the results were less
compelling with respect to discriminatingbetween the two
dimensions of trust for each target.Assessment of the supplier firm and salesperson trust measures raised questions
about discriminant validity between the two underlying
dimensions of trust, credibility and benevolence. Although
other researchers find evidence of discriminant validity,
these two dimensions of trustare highly correlated(see also
Ganesan 1994; Kumar, Scheer, and Steenkamp 1995).
Although credibility and benevolence could be conceptually distinct, in business relationshipssuch as those studied
here, they may be so intertwined that in practice they are
operationally inseparable. Therefore, we treat trust of the
salesperson and trust of the selling firm as unidimensional
constructs. In both cases, final measures of trust include
items that tap the credibility and benevolence aspects of
trust, along with some global measures of trust. Results of
these operationalizationsand furtherassessment of all measures are reported subsequently. (In Appendices A and B,
we reportall measures used for this study and key summary
statistics.)
Trust of the supplier firm, its antecedents and outcomes,

and controls. Measures of supplier firm trust were developed to accommodatethe industrialbuyer-suppliercontext

of this research.Items were generatedon the basis of interviews with marketingand purchasingpersonnel.The measure of supplierfirm trustexhibits high reliability(a = .94).
We also developed new scales to measurethe selling firm's
size, reputation,and confidential informationsharing, and
the buying firm's anticipatedfutureinteractionwith the selling firm. These measures exhibit good reliability, with the
lowest values of coefficient alpha being .78 for confidential
informationsharingand supplierreputation.
Age of the supplier relationshipwas measuredusing a
single item that asked how long the supplierfirm had been
in contact with the buying firm. Purchasechoice was measuredwith a yes/no responseto whetherthe buying firm had
chosen this supplierfor the focal purchase.
We developed the measureof a supplier'swillingness to
customize its productsand services for the buying firm as a
formative scale (Bollen and Lennox 1991) that identified
five ways a supplier could make changes specific to the
needs of the buying firm. As with all formativescales used
in this study,the items were averagedto form a single index
before being incorporated into the confirmatory factor
analysis model. For purposesof identifying the factor models, reliabilityfor all formativescales was assumedto be .85.
We evaluatedthe measuresfurtherthroughconfirmatory
factor analysis procedures.Although the chi-squarestatistic
was statistically significant (X2(21) = 444; p < .01), this is
not unusual with large sample sizes. The goodness of fit
index (GFI), comparativefit index (CFI; Bentler 1990), and
incrementalfit index (IFI; Bollen 1989) were .89, .93, and
.93, respectively.Together,the results suggest thatthese data
provide a reasonablefit with the hypothesizedmeasurement
model.
We used three methodsfor assessing discriminantvalidity. First, we conducted exploratory factor analyses using
both orthogonaland oblique rotations to ensure high loadings on hypothesized factors and low cross-loadings. Next,
we constructeda 95% confidence interval aroundthe estimates of correlationsbetween the latent constructs.Finally,
a series of nested confirmatoryfactor model comparisons
assessed whether differences existed when correlations
between the latent constructswere constrainedto 1.0. With
one exception, all three tests providedevidence of discriminant validity between each pair of constructsin the set. The
exploratory factor analysis indicated high cross-loadings
between the measuresof selling firm trustand supplierreputation. Because of the lack of strong evidence of discriminant validity, the reputationconstruct was removed from
furtheranalysis.
We used two methods for assessing convergentvalidity.
The LISREL estimates of paths from individual items to
latent factorswere all statisticallysignificant (p < .01), with
parameterestimates 10 to 20 times as large as the standard
errors. Moreover,results of the exploratoryfactor analysis
showed that each item loaded highly on its hypothesized
factor, with no high cross-loadings.
Trustof the salesperson and its antecedents.We adapted
measures of salesperson trust to the industrialbuying context of this research. The seven-item scale of salesperson
trustexhibits high reliability(a = .90).

Trustin Buyer-SellerRelationships/ 43

This content downloaded from 192.236.36.29 on Tue, 24 Sep 2013 11:48:26 AM

All use subject to JSTOR Terms and Conditions

Seven constructsare hypothesized to influence the buying firm's trust of a supplier's salesperson. Because little
empirical research has examined these factors in an industrial buying context, we developed new scales for these
antecedents. Items were generated using interviews with
marketingand purchasing personnel. Because most of the
antecedent factors are relatively narrowly defined constructs, we used three-item scales to capture the relevant
facets of each construct.These reflective measuresexhibited
good reliability,with coefficient alpha values ranging from
.79 for salesperson expertise to .90 for salesperson power,
likability,and similarity.
We developed the measureof social interactionas a formative scale, listing seven types of social interactiontypically encounteredin business marketing.Respondentswere
asked about the frequencyof each form of social interaction
between membersof their firm and the focal salesperson.A
single item asked respondentshow long the focal salesperson had called on their company.We performeda log transformationto attenuateskewness in this measure.
The chi-square statistic for the confirmatory factor
analysis model was statistically significant (2(221) = 549; p
< .01), but the GFI = .84, CFI = .93, and IFI = .93 were
within acceptable standards.We assessed convergent and
discriminant validity using the methods described previously, with good evidence that measureswere valid on these
dimensions.
Finally, additional tests of discriminant validity were
conducted to ensure that highly correlateditems across the
two sets of measures were unique. For example, we might
expect concern with respectto discriminantvaliditybetween
trustof a salespersonand trustof a supplierfirm.The test of
nested models provided evidence that these two latent constructs were discriminant (X2diff(l) = 73; p < .01). In sum-

mary, across several analytical procedures, measures

demonstratedthe sound psychometricpropertiesrequiredto
test researchhypotheses.

Results
We estimated the system of equations using two methods.
The dichotomous dependent variable (supplier
selected/not selected) necessitated the use of logistic
regression. We estimated the three remaining equations
simultaneously using three-stage least squares regression.
This method not only is efficient but also allows for correlated errors across dependent variables, which can be
expected with the two measures of trust. Because of the
expected high correlation among the predictors, we performed tests of multicollinearity (i.e., Belsley, Kuh, and
Welsh 1980) before we estimated the models. However,
these tests provided no indication that collinearity was a
concern.
The three-stage least squares regression model, which
tests the antecedentsto trustand the impactof each targetof
trust on anticipated future interaction, reports a systemweighted R-squaredof .65. This finding shows thatthe comprehensivemodel explains a large partof the variancein the
endogenous variables.The logistic regressionmodel shows
a good fit, because the likelihood ratiogoodness-of-fit test is

nonsignificant(201 d.f. = 192.49; p = .65). The model correctly classifies 86% of the observations(58% of the suppliers were selected in the sample). However, these results
overestimatethe true "hit rate"because the same data were
used in model estimationand validation.We reportresultsof
the two models in Tables 2 and 3 and describe them in the
following sections.
Antecedents

of Trust

Three of the five hypotheses tested regardingantecedentsto

suppliertrustare supportedby these data.As expected, supplier size (b = .09; p < .01) and the supplier'swillingness to
customize (b = .21; p < .01) have a positive impact on buying firm trust. However, the supplier's willingness to share
confidential information (b = .06) and length of the relationship (b = .00) are unrelatedto buying firm trust of the
supplierfirm. Finally, trustof the supplier'ssalespersonhas
a positive effect on trust of the selling firm (b = .77; p <
.01).
Two personal characteristics of the salesperson were
expected to influence buyer trust.As predicted,salesperson
expertise (b = .17; p < .01) has a positive effect on the buying firm's trust of the salesperson. However, the extent to
which the buying firm perceives the salespersonto be powerful is not relatedto salespersontrust(b = -.03).
Three characteristics of the relationship between the
buying firm and the supplier's salesperson had statistically
significant effects on trustof the salesperson.The extent to
which the salesperson is perceived to be likable (b = .20; p
< .01) and similar to membersof the buying firm (b = .09; p
< .05) has a positive impacton the buying firm's trustof the
salesperson.Also as expected, frequencyof business contact
with the salesperson positively influences the buying firm's
trust of the salesperson (b = 12; p < .05). However, frequency of social interactionwith the salesperson(b = -.04)
and length of time the salespersoncalled on the buying firm
(b = -.06) were unrelatedto the buying firm's trust of the
salesperson. Finally, the buying firm's trust of the supplier
firm has a positive effect on its trustof the salesperson(b =
.52;p < 01).
Consequences of Trust
We report results of the logistic regression used to test
hypotheses regardingpurchasechoice in Table 3. The nonsignificant parameter estimates for trust of the supplier
firm (b = .11) and trust of the salesperson (b = -.13) show
that neither targetof trust influences purchasechoice when
controlling for past purchase experience, supplier delivery
performance, relative price/cost, and product/service performance. This is not to say that trustis unimportant.Comparing the means of selling firm trust and salesperson trust
between selected and unselected suppliers, we find that
selected suppliers and their salespeople were more trusted
than those not chosen (t = 6.13, p < .01; t = 4.70, p < .01,
respectively). However, though trust levels differ between
selected and unselected suppliers, trust does not explain
any additional variance in purchase choice after controlling for previous experience with the supplier and supplier
performance.

44 / Journalof Marketing,April1997

This content downloaded from 192.236.36.29 on Tue, 24 Sep 2013 11:48:26 AM

All use subject to JSTOR Terms and Conditions

The results reportedin the last column of Table 2 indicate that after controlling for the .three dimensions of performance (none of which was statistically significant), past
purchaseexperience (b = .29; p < .01), and currentpurchase
choice (b = .20; p < .01), trust of the selling firm plays an
importantrole in anticipatedfutureinteractionwith the supplier (b = .31; p < .01). We discuss the implicationsof these
findings in the following section.

Discussion
Our results confirm that buying firms develop trust in both
the supplierfirm and the supplier's salesperson. Of particular interest, we find that current supplier selection is not
influenced by trustof the supplierfirm or its salesperson. In
the industrialbuying context of this study, the key criteria
for supplier selection are delivery performanceand relative
price/cost. However, trust of the supplier firm and trust of
the salesperson (operating indirectly through supplier firm
trust)do increase the likelihood that buyers anticipatedoing
business with the supplierfirm in the future.Takentogether,
these findings indicate the extent to which trust influences
long-termrelationships(i.e., trust'srole in relationshipmarketing as opposed to transactionalmarketing).We discuss
our results, and their implications for marketingtheory and
practice, subsequently.
Trust Development Processes
Although our resultssuggest thattrustof the salespersoncan
be transferred to the supplier firm and vice versa, the
processes by which trustdevelops appearto differ when the
targetis an organization-a collection of people responsible

TABLE 3
Results of Maximum Likelihood Logistic
Regression on Purchase Choice
Purchase Choice

Independent Variables

.11
-.13

Trustof supplierfirm
Trustof salesperson

.47a
.55a

Delivery performance
Relative price/cost

.11
Product/serviceperformance
.02
Purchase experience withsupplier
of choosing
Positiveparameter
estimatesindicategreaterlikelihood
supplier.
estimatesreported.
Standardized
parameter
Likelihood
ratiowith201 d.f.= 192.49;p = .65.
of chi-square:
p < .01.
aProbability
for large segmentsof customers,as opposed to an individual
salesperson responsible for the needs of individual customers in his or her territory.Although not exclusive to
either target,our theory and results suggest that the calculative process of building trustcould be applied more readily
to supplierfirms thansalespeople. In additionto transferring
trustfrom salespeople, buyersappearedto form trustin selling firms on the basis of characteristicsof the firm (e.g.,
size) and actual behaviors (e.g., investing in customer-specific assets). Together, such characteristicsand behaviors
provide the buyer with an indication that acting in an
untrustworthy(or opportunistic)way would be costly for the
supplier.For example, a supplierthat fails to deliver a critical producton time, beyond paying a penalty, could lose a
valuable customerand jeopardize its staturein the industry.

TABLE 2
Results of Three-Stage Least Squares Estimation
Dependent Variables
Trust of
Supplier Firm

Independent Variables
Supplierfirmsize
Supplierfirmwillingnessto customize
Supplierfirmconfidentialinformationsharing
Lengthof relationshipwithsupplierfirm1
Trust of salesperson

Salesperson expertise
Salesperson power
Salesperson likability
Salesperson similarity
Frequentbusiness contact withsalesperson
Frequentsocial contact with salesperson
Lengthof relationshipwithsalesperson1
Trustof supplierfirm
Deliveryperformance
Relativeprice/cost
Product/serviceperformance
Purchase experience withsupplier
Purchase choice
= .65.
SystemWeightedR-squared
estimatesreported.
Standardized
parameter
Natural
logtakenof thisvariable.

Trust of
Anticipated
Salesperson Future Interaction

.09a
.21a
.06
.00
.77a

.17a
-.03
.20a
.09b
.12b
-.04
-.06
.52a

.31a
.12
.07
.01
.29a
.20a

ap < .01.
bp <

.05 (allhypothesistests one-tailed).

Trustin Buyer-SellerRelationships/ 45

This content downloaded from 192.236.36.29 on Tue, 24 Sep 2013 11:48:26 AM

All use subject to JSTOR Terms and Conditions

If buying organizationsassume that suppliers develop policies designed to maximize shareholderwealth, predictions
of trustworthybehavior emerge primarilyfrom calculating
the effects of particularpolicies on firm performance.
Inferringthe costs and benefits that accrue to individual
salespeople is likely to be more difficult. Although supplier
firms develop policies that are based on treating groups of
customers in a similar manner,salespeople have a narrower
"span of surveillance" that lets them adapt their messages
and selling style to individual customer needs (Cespedes
1995). This gives salespeople the chance to show buyers
that they are expert and likable and share values consistent
with the customer's employees. In other words, unlike supplier firms, salespeople can demonstrate more easily predictable behaviors and benevolent intentions through customized interpersonalinteractionwith the customer.Therefore, it appears that the processes of prediction and intentionality are more likely to be invoked in developing trust
of a salesperson.
The implications of these different routes to trust are
important.Lewicki and Bunker(1995) proposethatthereare
different levels of trustingrelationships.Relationshipsbuilt
on calculative trust are at the lowest and most fragile level.
The highest levels of trusting relationships are based on
internalizingthe other's desires and intentions-where trust
emerges throughthe process of intentionality.Trustbuilt on
closely identifying with the partner is more flexible to
changing conditions and a more difficult bond to breakthan
is calculative trust (Lewicki and Bunker 1995). This suggests that the interpersonaltrustengenderedby salespeople
and transferredto the supplierfirm plays a key role in developing and maintainingenduringbuyer-seller relationships.
Consequences of Trust
Our findings that trust of the salesperson and supplier firm
were not related to current supplier choice are contraryto
our predictionsand those of priorresearch(cf. Millimanand
Fugate 1988). Although trustis higher for selected suppliers
than for those not selected, our results suggest that aspects
of the marketingmix-price and reliabledelivery-actually
make the sale. These findings could reflect the fact thatprofessional buyers are trained to focus on objective evidence
that demonstrates the superiority of the product offering,
ratherthan on subjective assessments of trust.For example,
"buyerscurrentlyare taughtto carefully evaluate long-term
total costs associated with dealing with a particularsupplier"(Hutt and Speh 1995, p. 43). The fact that 94% of the
purchasesevaluated in our study involved competitive bidding indicates a formalization of the supplier selection
process, which could mitigate the effects of trust-particularly because more objective criteria (price and delivery
capability) were significantly related to choice and more
subjective criteria (product/serviceperformanceand trust)
were not.
Therefore, it appears that trust operates as an "order
qualifier,"not an "orderwinner."Orderqualifiersare "those
criteria that a company must meet for a customer to even
consider it as a possible supplier,"whereasorderwinnersare
"those criteria that win the order" (Hill 1994, p. 33). In
industrial markets, some suggest that product quality now

could be an order qualifier for many purchase decisions

(Dertouzos, Lester,and Solow 1989). To even be includedin
a buying firm's considerationset, suppliersmust providethe
requisite level of product quality. Our results suggest that
like quality, trust could be requiredjust to enter into a customer's consideration set. This contention is supportedby
the fact thatmean scores for trustof the salespersonand supplier firm (whether chosen or not) were 5.54 and 5.30,
respectively,on a seven-point scale. These findings suggest
thatfirms focusing solely on providingthe lowest-cost product on a timely basis could find they are not even considered
by the buying firm if they have not established a trusting
relationship.
Our results also indicate that trust of the supplier firm
and trust of the salesperson (operating indirectly through
trust of the supplier) are related to anticipatedfuture interaction. These findings have several implicationsfor supplier
firms. First,the importanceof trustin futureintentionsindicates thattrustis a criterionbuyersuse to evaluatesuppliers.
Second, developing trusting relationships represents an
investmentwith a long-termpayoff. Finally,because trustin
the salespersonoperatesthroughthe supplierfirm, the salesperson's primaryrole in implementing a relationshipmarketing strategycould be to institutionalizetrust in the supplier firm.
Theoretical Implications
These findings suggest several importantavenues for the
role of trustin theories of industrialbuying and buyer-seller
relationships. First, the framework provided by the five
trust-buildingprocesses gives new theoretical insight into
how trust is generated.Although our researchdoes not test
the existence of these processes, it provides a theoretical
foundationfor understandinghow trustdevelops.
A second contributionof this researchemerges from the
inclusion of two targetsof trust in the same study.The theoretical development and results suggest that the processes
by which trustis engendereddiffer between salespeople and
their employers. Although only trust of the supplier firm
directly influences future purchase intentions, trust of the
salesperson is an importantfactor in building trust of the
supplierfirm.
A third contributionto theory emerges from the relatively large number of relationships considered in this
research.Previous empirical researchhas not involved testing the effects of supplier size, confidential information
sharing, perceived salesperson power, or the reciprocal
effects of salesperson and supplier firm trust. Furthermore,
previous researchhas not involved examining trust'srole in
purchasechoice. Only by understandingthe broadernomological network surroundingtrust will we be able to understand better how to build and maintain trusting business
relationships.
Finally, the study extends other research in the area of
trust by exploring its role in a new context. Some of our
results confirm previous empirical research-though others
(notably the lack of a link between length of the relationship
and trust) are contraryto priorfindings. Previous empirical
researchprimarilyhas been conducted in channels of distribution (Anderson and Weitz 1989; Ganesan 1994; Kumar,

46 / Journalof Marketing,April1997

This content downloaded from 192.236.36.29 on Tue, 24 Sep 2013 11:48:26 AM

All use subject to JSTOR Terms and Conditions

Scheer, and Steenkamp 1995; Morgan and Hunt 1994) or

service settings (Crosby, Evans, and Cowles 1990; Moorman, Deshpande, and Zaltman 1993; Moorman, Zaltman,
and Deshpande 1992)-contexts that differ from the industrial buying setting of the current study. Compared with
channel relationships, industrial markets tend to involve
only moderate levels of dependence, and the role of the
salesperson is less critical. Conversely, in service settings
dependencecould be low and the service provideror salesperson is often key (Crosby, Evans, and Cowles 1990;
Moorman,Deshpande,and Zaltman 1993; Moorman,Zaltman, and Deshpande 1992). Togetherthe results of this and
previous research suggest that the selling situation could
moderatethe developmentand influence of trust.
Managerial Implications
One of the key findings of this researchis thattrustof a supplier firm is positively related to the likelihood that buyers
plan to do business with suppliersin the future.This association emerges despite controls for supplier performance,
past experience with the supplier, and the currentsupplier
selection decision. Apparently,buying firms value trusted
suppliers and consider trust an importantprerequisitefor
building long-term relationships. A major determinantof
futurebusiness opportunityis the extent to which customers
trust the selling firm, so suppliers should engage in trustenhancingactivities.
The results, and the processes that build trust, provide
insights into the directionsuch efforts might take. Forexample, our findings suggest that buyers rely on supplier firm
characteristicssuch as size to assess a supplier firm's trustworthiness.Buyers could bestow trustthrougha calculative
process by determiningthatlarge suppliershave a lot to lose
by acting in an untrustworthyway. Therefore,it is important
for suppliersto managecustomerperceptionsof firm sizeperhapsthroughimage-enhancingtechniquessuch as advertising and publicity.Buyers also could use size as a basis for
transferringtrustto unknownor untriedsuppliers,relying on
the experience of others. This suggests that marketers
should emphasize customer satisfaction, perhaps including
testimonials in their marketingcommunications.
The findings also suggest thatbuyerswho perceivedthat
suppliers are willing to make idiosyncratic investments
judged their partnersto be trustworthy.Although customerspecific adaptationsimply costs for the seller, our findings
suggest that these are investmentsthat tie the firms together
in strong buyer-supplierrelationships.Because such investments contributeto forging strong buyer trust in the selling
firm, they can be expected to pay off in the long run.
In addition, the results indicate that a buyer's trust in a
supplier firm is based, in part,on encounters with the supplier's salesperson.Therefore,the company should teach its
salespeople how to develop trust(cf. Doyle and Roth 1992).
Managers could use our results, and the trust-building
processes, to guide their training efforts. For example,
through the process of intentionality, buyers attribute
benevolent motives to those they like or perceive as similar
to themselves. Similarity and likability also can bolster the
buyer's confidence in predicting the supplier's future
behavior. Salespeople might be taught to exhibit likability

by being friendly and making efficient use of the customer's time. To project similarity, salespeople probably
should establish common ground with the buyer. These
antecedents to trust can be taught by means of formal lectures or role-playing exercises. When these skills have been
mastered,managersshould place salespeople with accounts
where they will be well liked and perceived as similar to
their customers.
Salespeople also must masterthe technical skills necessary to convey expertise with respect to their products.Buyers trust salespeople they perceive to be expert, perhaps
because they think expert salespeople can deliver on their
promises. Moreover, salespeople should contact customers
often, because study results show thatfrequentcontact plays
a centralrole in developing trust.Finally,salespeople should
be rewardedfor such trust-buildingbehaviors,because they
strengthenthe link between the buying firm and the supplier.
Limitations and Further Research
The choice of researchdesign forced certain trade-offsthat
could limit the findings. First, though the five trust-building
processes provide a theoretical foundation for research
hypotheses, the current study does not directly measure
these processes. Further research employing alternative
methods could address this limitation usefully. Second,
though our conceptual model suggests several important
antecedents and outcomes of trust, other variables could
have been omitted. For example, buyers could bestow trust
on the supplier firm on the basis of its capabilities-for
example, through assessment of the supplier's past performance. Our decision to examine purchasechoice meant that
some buyers would not have had previous experience with
the focal supplier,a fact that precludedmeasuringpast performance in the current study. Additional research could
involve examining a broader spectrum of antecedents and
outcomes of trust.
A third research design issue-the decision to ask
respondents to identify a recent purchase for which more
than one supplier was considered seriously-undoubtedly
resultedin a truncationin the range of trustexamined. This
approachwas necessary because a certain amountof ambiguity must exist in the decision context for trustto be operative. The fact that buyers reportedthe use of a competitive
bidding process in 94% of purchasesituationssuggests that
buyerscould have identifiedtwo suppliersthatwere finalists
in a bidding competition. This, combined with the fact that
65% of the suppliershad been used many times before, suggests that even suppliers not selected for the current purchase decision were highly trusted. However, the current
trend toward closer collaborationbetween buyers and suppliers suggests that a supplier firm must be trusted to be
included in the buyer's consideration set. Therefore, the
study design realisticallyportraysthe situationfacing industrial buyers and suppliers.
It is importantto note thatour findings may not generalize to dissimilarcultures.For example, in the United States,
concerns about conflict of interest and unethical behavior
arise when personal relationships develop (Schuster and
Copeland 1996). Researchersin the future should examine
the role national culture plays in buyer-seller relationships
Trustin Buyer-SellerRelationships/ 47

This content downloaded from 192.236.36.29 on Tue, 24 Sep 2013 11:48:26 AM

All use subject to JSTOR Terms and Conditions

Conclusion

and buyer-salesperson encounters. As firms expand their

businesses
businesses
beyond national borders, studies that involve
Supplierfirms must make significantinvestmentsto develop
how
antecedents and outcomes of trust differ
examining
and maintain
maintaincustomertrust.For suppliers,
suppliers,the value of such
across cultures should prove useful to marketingmanagers.
eefforts i m
most apparentwhen high levels of buyertrustlead
is
Finally,we need to know more abouthow trustdevelopsurchasin
su er. Our
to more favorablepurchasingoutcomes for the supplier.
over time and whetherthe processesthatfostertrustat the outt though the process of buildingcustomer
results suggest that
set of a relationshipare the same as those that maintainit in
tieconsuing
and complex, its outtrust is epensie
laterstages. Furtherresearchusing a longitudinaldesign could
i expensive,
s of time-consuming,
c
oin
stong buyer-seller bonds and
address these questions and provide for strongerinferences
e
o . could
t
oui be critically
enhanced
loyalty
importanttoo supplier
aboutthe directionsof causalitypositedhere. By understandfirms
ing the dynamicsof trust,we can obtaina greaterunderstanding of how businessrelationshipschange,grow,and decline.
APPENDIX A
Measures and Key Summary Statistics for Supplier Firm Trust and Its Antecedents,
Outcomes, and Control Variables
Scale Name [Response Cue1] (Scale Mean and Standard Deviation) and IndividualItems
Supplier Firm Reputation (5.78; 1.11)
This supplierhas a reputationfor being honest.
This supplieris knownto be concerned about customers.
This supplierhas a bad reputationin the market.(R)
Supplier Firm Size (3.98; 1.71)
This supplieris a very large company.
This supplieris the industry'sbiggest supplierof this product.
This supplieris a small playerin the market.(R)
Supplier FirmWillingness to Customize for Buyer [not at all-very much](4.48; 1.47)
Just for us this supplieris willingto customize its products.
Just for us this supplieris willingto change its productionprocess.
Just for us this supplieris willingto change its inventoryprocedures.
Just for us this supplieris willingto change its deliveryprocedures.
Just for us this supplieris willingto invest in tools and equipment.
Supplier Firm Confidential InformationSharing (3.90; 1.66)
This suppliershares proprietaryinformationwithour firm.
This supplierwillshare confidentialinformationto help us.
Length of Relationship With Supplier Firm [_ years & _ months](10.67; 9.92)
Abouthow long has yourcompany had contact withthis supplier?
Trust of Supplier Firm (5.30; 1.20)
This supplierkeeps promises it makes to our firm.
This supplieris not always honest withus. (R)
We believe the informationthat this vendor providesus.
This supplieris genuinelyconcerned that our business succeeds.
When makingimportantdecisions, this supplierconsiders our welfareas well as its own.
We trustthis vendor keeps our best interests in mind.
This supplieris trustworthy.
We find it necessary to be cautious withthis supplier.(R)
Purchase Choice [yes or no] (.58; .49)
Forthis purchase, did yourfirmbuy fromthis supplier?
Anticipated Future Interaction [verylittlechance-definitelywilluse] (5.66; 1.85)
How likelyis it that yourfirmwillmake a purchase fromthis supplierduringthe next 3 years?
How likelyis it that yourfirmwillmake a purchase fromthis supplierduringthe next year?
Delivery Performance [Howdid this suppliercompare withothers on each of these criteria?
(much worse than others-equal to others-much betterthan others)](4.68; 1.24)
deliveryspeed
deliveryreliability
productavailability
Relative Price/Cost [Howdid this suppliercompare withothers on each of these criteria?
(much worse than others-equal to others-much betterthan others)](4.56; 1.33)
price
total cost
terms of sale
481 Journalof Marketing,April1997

This content downloaded from 192.236.36.29 on Tue, 24 Sep 2013 11:48:26 AM

All use subject to JSTOR Terms and Conditions

Scale
Reliability
a = .78

a = .83

n/a2

a = .78

n/a
a = .94

n/a
a = .95

n/a2

n/a2

Product/Service

Performance [How did this supplier compare with others on each of these criteria?

(much worse than others-equal to others-much betterthan others)] (4.79; .99)

product/servicefeatures
product/servicequality
product/servicereliability
technical support
after sale service and support

n/a2

Purchase Experience with Supplier [Has your firm made purchases from this supplier in the past?

(yes, many times-yes, a few times-no)] (1.53; .69)

n/a

Fit Statistics:chi square with211 d.f. = 444.13 (p < .01); GFI= .88; CFI= .93; IFI= .93.

'Unlessotherwisenoted,responsecues forallscales arestronglyagree-strongly

disagree.
forassessingreliability.
2Asa formative
scale,coefficient
alphais notappropriate
items.
(R)Reverse-scored
APPENDIX B
Measures and Key Summary Statistics for Salesperson

Trust and Its Antecedents

Scale Name [Response Cue1] (Scale Mean and Standard Deviation) and IndividualItems

Scale
Reliability

Salesperson Expertise (5.54; 1.08)

This salesperson is very knowledgeable.
This salesperson knows his/her productline very well.
This salesperson is not an expert. (R)

a = .79

Salesperson Power (5.05; 1.22)

This salesperson has the clout to get his/herway withthe supplier.
This salesperson is one of this supplier'smost importantsalespeople.
This salesperson has power in his/herfirm.

a = .90

Salesperson Likability(5.54; 1.05)

This salesperson is friendly.
This salesperson is always nice to us.
This salesperson is someone we liketo have around.

a = .90

Salesperson Similarity (4.47; 1.29)

This salesperson shares similarinterests withpeople in our firm.
This salesperson has values similarto people in our firm.
This salesperson is very similarto people in our firm.

a = .90

Frequent Contact with Salesperson (4.39; 1.45)

This salesperson frequentlyvisits our place of business.
This salesperson takes a lot of time learningour needs.
This salesperson spends considerabletime gettingto knowour people.

a = .85

Extent of Social Interaction with Salesperson

[How often do you and other people from your firm

interactwiththis salesperson in these ways? (never-often)](2.47; 1.01)

Meet away fromthe workplace.
Talkabout family,sports or other personal interests.
Attendentertainmentevents (sports, theater etc.)
Meet over breakfast,lunch or dinner.
Get together primarilyto have fun.
Talkabout common interests besides work.
Get together withother familymembers.

Length of Relationship With Salesperson [_ years &_ months](5.56; 4.68)

Abouthow long has this salesperson called on yourcompany?

n/a2

n/a
a = .90

Trust of Salesperson (5.54; 1.18)

This salesperson has been frankin dealing withus.
This salesperson does not make false claims.
We do not thinkthis salesperson is completelyopen in dealing with us. (R)
This salesperson is only concerned about himself/herself.(R)
This salesperson does not seem to be concerned withour needs. (R)
The people at my firmdo not trustthis salesperson. (R)
This salesperson is not trustworthy.(R)
Fit Statistics:chi square with221 d.f. = 548.72 (p < .01); GFI= .84; CFI= .93; IFI= .93.

disagree.
1Unlessotherwisenoted,responsecues forallscales arestronglyagree-strongly
forassessingreliability.
2Asa formative
scale,coefficientalphais notappropriate
items.
(R)Reverse-scored
Trustin Buyer-SellerRelationships/ 49

This content downloaded from 192.236.36.29 on Tue, 24 Sep 2013 11:48:26 AM

All use subject to JSTOR Terms and Conditions

REFERENCES
Adkins, Robert T. and John E. Swan (1980-81), "Improvingthe
Public Acceptance of Salespeople Through Professionalism,"
Journal of Personal Selling & Sales Management,2 (Fall/Winter) 32-38.
Akerlof, George A. (1970), "The Market for 'Lemons': Quality
Under Uncertainty and the Market Mechanism," Quarterly
Journal of Economics, 84 (August), 488-500.
Anderson, Erin, LeonardM. Lodish, and BartonA. Weitz (1987),
"ResourceAllocation Behavior ConventionalChannels,"Journal of MarketingResearch, 23 (February),254-62.
and BartonWeitz (1989), "Determinantsof Continuityin
ConventionalIndustrialChannel Dyads,"MarketingScience, 8
(4), 310-23.
and
(1992), "The Use of Pledges to Build and Sustain Commitment in DistributionChannels,"Journal of Marketing Research, 29 (February),18-34.
Anderson, James C. and David W. Gerbing (1988), "Structural
Equation Modeling in Practice:A Review and Recommended
Two-Step Approach," Psychological Bulletin, 103 (May),
411-23.
and JamesA. Narus (1990), "A Model of DistributorFirm
and ManufacturerFirmWorkingPartnerships,"
Journalof Marketing, 54 (January),42-58.
Bagozzi, Richard P. and Lynn W. Phillips (1982), "Representing
and Testing OrganizationalTheories: A Holistic Construal,"
AdministrativeScience Quarterly,27, 459-89.
Belsley, D. A., E. Kuh, and R. E. Welsh (1980), RegressionDiagnostics: IdentifyingInfluentialData and Sources of Collinearity. New York:John Wiley & Sons, Inc.
Bentler, Peter M. (1990), "ComparativeFit Indexes in Structural
Models,"Psychological Bulletin, 107 (2), 238-46.
and Chih-PingChou (1987), "PracticalIssues in Structural
Modeling," Sociological Methods and Research, 16 (August),
78-117.
Bollen, Kenneth (1989), Structural Equations with Latent Variables. New York:John Wiley & Sons, Inc.
and Richard Lennox (1991), "ConventionalWisdom on
Measurement:A StructuralEquationPerspective,"Psychological Bulletin, 110 (2), 305-14.
Bradach,JeffreyL. and RobertG. Eccles (1989), "Price,Authority,
and Trust,"Annual Review of Sociology, 15, 97-118.
Busch, Paul and David T. Wilson (1976), "AnExperimentalAnalysis of a Salesman's Expertand ReferentBases of Social Power
in the Buyer-Seller Dyad,"Journal of MarketingResearch, 13
(February),3-11.
Cespedes, Frank V. (1995), Concurrent Marketing: Integrating
Product, Sales and Service. Boston: HarvardBusiness School
Press.
Choo, T. (1964), "CommunicatorCredibilityand Communication
Discrepancy as Determinantsof Opinion Change,"Journal of
Social Psychology, 64, 1-20.
Crosby, Lawrence A., Kenneth R. Evans, and Deborah Cowles
(1990), "RelationshipQuality in Services Selling: An Interpersonal Influence Perspective,"Journal of Marketing,54 (July),
68-81.
Dasgupta, Partha(1988), "Trustas a Commodity,"in Trust:Making and Breaking CooperativeRelations, Diego Gambetta,ed.
New York:Basil Blackwell, Inc.
Dertouzos, Michael L., Richard K. Lester, and Robert M. Solow
(1989), Made in America: Regaining the Productive Edge.
Cambridge,MA: The MIT Press.
Deutsch, Morton (1960), "The Effect of MotivationalOrientation
Upon Trust and Suspicion," Human Relations, 13 (May),
123-39.
(1962), "Cooperationand Trust:Some TheoreticalNotes,"
in Nebraska Symposiumon Motivation,MarshallR. Jones, ed.
Lincoln, NE: Universityof NebraskaPress.

Dillman, Don A. (1978), Mail and TelephoneSurveys: The Total

Design Method. New York:John Wiley & Sons, Inc.
Doyle, Stephen X. and George Thomas Roth (1992), "Selling and
Sales Managementin Action: The Use of Insight Coaching to
ImproveRelationshipSelling,"Journal of PersonalSelling and
Sales Management, 12 (Winter)59-64.
Dwyer, F. Robert, Paul H. Schurr,and Sejo Oh (1987), "Developing Buyer-Seller Relationships," Journal of Marketing, 51
(April), 11-27.
Emshwiller,John R. (1991), "SuppliersStruggle to ImproveQuality as Big Firms Slash TheirVendorRolls," WallStreetJournal,
(August 16), B1, B2.
Ganesan, Shankar(1994), "Determinantsof Long-TermOrientation in Buyer-Seller Relationships,"Journal of Marketing,58
(April), 1-19.
Gundlach, Gregory T. and Ernest R. Cadotte (1994), "Exchange
Interdependenceand InterfirmInteraction:Researchin a Simulated Channel Setting," Journal of Marketing Research, 31
(November), 516-32.
Hawes, Jon M., KennethE. Mast, and John E. Swan (1989), "Trust
Earning Perceptions of Sellers and Buyers," Journal of Personal Selling & Sales Management,9 (Spring), 1-8.
Governancein MarketHeide, Jan B. (1994), "Interorganizational
ing Channels,"Journal of Marketing,58 (January),71-85.
Hill, Charles W. (1990), "Cooperation, Opportunism, and the
Invisible Hand: Implications for Transaction Cost Theory,"
Academyof ManagementReview, 15, 500-513.
Hill, Terry (1994), Manufacturing Strategy. Burr Ridge, IL:
RichardD. Irwin, Inc.
Hutt,Michael D. and Thomas W. Speh (1995), Business Marketing
Management,5th ed. FortWorth,TX: The Dryden Press.
John, George (1984), "An Empirical Investigation of Some
Antecedentsof Opportunismin a MarketingChannel,"Journal
of MarketingResearch, 21 (August), 278-89.
Johnson, D. W. and S. Johnson (1972), "The Effects of Attitude
Intensity, Expectation of Goal Facilitation, and Actual Goal
Facilitationon InterpersonalAttraction,"Journalof Experimental Social Psychology, 8 (3), 197-206.
Johnston,Wesley J. and Jeffrey E. Lewin (1996), "Organizational
Buying Behavior:Towardan IntegrativeFramework,"Journal
of Business Research, 35, 1-15.
Joreskog, Karl G. and Dag Sorbom (1993), LISREL8, A Guide to
the Programand Applications. Chicago: SPSS Inc.
Kelly, Kevin and Kathleen Kerwin (1993), "There'sAnother Side
to the Lopez Saga,"Business Week,(August 23), 26.
and Eric Schine (1992), "How Did Sears Blow This Gasket?"Business Week,(June 29), 38.
Kelman, H. C. and C. I. Hovland (1953), "'Reinstatement'of the
Communicatorin Delayed Measurementof Opinion Change,"
Journal of Abnormaland Social Psychology, 48, 327-35.
Kumar, Nirmalya, Lisa K. Scheer, and Jan-Benedict E. M.
Steenkamp(1995), "The Effects of Perceived Interdependence
on Dealer Attitudes," Journal of Marketing Research, 32
(August), 348-56.
Larzelere,RobertE. and Ted L. Huston (1980), "The Dyadic Trust
Scale: Toward Understanding Interpersonal Trust in Close
Relationships,"Journal of Marriage and the Family, 42 (3),
595-604.
Lehmann,Donald R. and John O'Shaugnessy (1974), "Difference
in AttributeImportancefor Different IndustrialProducts,"Journal of Marketing,38 (April), 36-42.
Lewicki, Roy J. and BarbaraBenedict Bunker (1995), "Trustin
Relationships:A Model of Development and Decline," in Conflict, Cooperation, and Justice, BarbaraBenedict Bunker and
JeffreyZ. Rubin, eds. San Francisco:Jossey-Bass.
Lewis, David J. and Andrew Weigert (1985), "Trustas a Social
Reality,"Social Forces, 63 (4), 967-85.

50 / Journalof Marketing,April1997

This content downloaded from 192.236.36.29 on Tue, 24 Sep 2013 11:48:26 AM

All use subject to JSTOR Terms and Conditions

Lindskold,Svenn (1978), "TrustDevelopment,the GRIT Proposal

and the Effects of ConciliatoryActs on Conflict and Cooperation," Psychological Bulletin, 85 (4), 772-93.
Macneil, lan R. (1980), The New Social Contract.New Haven,CT:
Yale University Press.
Milliman, Ronald E. and Douglas Fugate (1988), "Using Trust
Transference as a Persuasion Technique:An Empirical Field
Investigation,"Journal of Personal Selling and Sales Management, 8 (August), 1-7.
Moorman, Christine, Rohit Deshpande, and Gerald Zaltman
(1993), "FactorsAffecting Trust in MarketResearch Relationships,"Journal of Marketing,57 (January),81-101.
, GeraldZaltman,and Rohit Deshpande(1992), "Relationships Between Providers and Users of Market Research: The
Dynamics of Trust Within and Between Organizations,"Journal of MarketingResearch, 29 (August), 314-28.
Morgan, Robert M. and Shelby D. Hunt (1994), "The Commitment-Trust Theory of Relationship Marketing,"Journal of
Marketing,58 (July), 20-38.
Noordewier,Thomas G., George John, and John R. Nevin (1990),
"PerformanceOutcomes of PurchasingArrangementsin Industrial Buyer-Vendor Relationships,"Journal of Marketing, 54
(October), 80-93.
Pfeffer, Jeffrey and Gerald R. Salancik (1978), The External Control of Organizations:A Resource DependenceApproach.New
York:Harperand Row Publishers,Inc.
Rao, Akshay and Mark Bergen (1992), "Price Premiumas a Consequence of Lack of Buyers' Information,"Journal of Consumer Research, 19 (December) 412-23.
Robinson, Patrick J., Charles W. Faris, and YoramWind (1967),
Industrial Buying and Creative Marketing.Boston: Allyn and
Bacon.
Rotter,Julian(1967), "A New Scale for the Measurementof InterpersonalTrust,"Journal of Personality,35 (4), 651-65.
(1980), "InterpersonalTrust, Trustworthiness,and Gullibility,"American Psychologist, 35 (1), 1-7.
Scanzoni, John (1979), "Social Exchange and BehavioralInterdependence," in Social Exchange in Developing Relationships,
R.L. Burgess and T. L. Huston,eds. New York:Academic Press.
Schiller, Zachary (1992), "Goodyear Is Gunning Its Marketing
Engine,"Business Week,(March 16), 42.

Schlenker,BarryR., Robert Helm, and James T. Tedeschi (1973),

"The Effects of Personalityand SituationalVariablesof Behavioral Trust,"Journal of Personalityand Social Psychology, 25,
419-27.
Schurr, Paul H. and Julie L. Ozanne (1985), "Influences on
Exchange Processes: Buyers' Preconceptions of a Seller's
Trustworthinessand BargainingToughness,"Journal of Consumer Research, 11 (4), 939-53.
Schuster,Camille and Michael Copeland(1996), Global Business:
Planningfor Sales and Negotiations.FortWorth,TX: The Dryden Press.
Shapiro, D., B. H. Sheppard,and L. Cheraskin(1992), "Business
on a Handshake,"NegotiationJournal, 8 (4), 365-77.
Strub,Peter J. and T. B. Priest (1976), "TwoPatternsof Establishing Trust: The MarijuanaUser," Sociological Focus, 9 (4),
399-411.
Swan, John E. and JohannahJones Nolan (1985), "GainingCustomerTrust:A ConceptualGuide for the Salesperson,"Journal
of Personal Selling & Sales Management,5 (2), 39-48.
, I. FrederickTrawick, Jr., David R. Rink, and Jenny J.
Roberts (1988), "MeasuringDimensions of PurchaserTrustof
IndustrialSalespeople," Journal of Personal Selling & Sales
Management,8 (1), 1-9.
, and David W. Silva (1985), "How Industrial
Salespeople Gain CustomerTrust,"IndustrialMarketingManagement, 14, 203-11.
Telser, Lester G. (1980), "A Theory of Self-EnforcingContracts,"
Journal of Business, 53 (January),27-44.
Williamson, Oliver E. (1985), The Economic Institutionsof Capitalism. New York:The Free Press.
(1991), "Calculativeness,Trust and Economic Organization,"Journal of Law and Economics, 26 (April), 453-86.
Wilson, ElizabethJ. (1994), "The Relative Importanceof Supplier
Selection Criteria:A Review and Update,"InternationalJournal of Purchasing and Materials Management,30 (Summer),
35-41.
Young, Louise C. and Ian F. Wilkinson(1989), "The Role of Trust
and Co-operation in Marketing Channels: A Preliminary
Study,"EuropeanJournal of Marketing,23 (2), 109-22.

Trustin Buyer-SellerRelationships/ 51

This content downloaded from 192.236.36.29 on Tue, 24 Sep 2013 11:48:26 AM

All use subject to JSTOR Terms and Conditions