LIBRARY

I Michigan State
, Universlty

 

 

 

PLACE IN RETURN Box
to remove this checkout from your record.
TO AVOID FINES return on or before date due.

 

DATE DUE DATE DUE MTE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

use mm“

AN INVESTIGATION INTO THE ANTECEDENTS OF TRUST IN COOPERATIVE
BUYER-SUPPLIER RELATIONSHIPS

by

Christian Bechtel

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirement
for the degree of

DOCTOR OF PHILOSOPHY

Department of Marketing and Supply Chain Management

1998

ABSTRACT
AN INVESTIGATION INTO THE ANTECEDENTS OF TRUST IN COOPERATIVE
BUYER-SUPPLIER RELATIONSHIPS
By

Christian Bechtel

Prior research on COOperative buyer-supplier relationships has largely focused on
supplier identification and initial relationship development. The balance of power in
many industries, however, between buyers and supplier is changing as supplier take on
more responsibility. This study will analyze the increasingly important dynamics of
interorganizational trust development in buyer-supplier relationships.

To develop a better understanding of the importance of trust and the antecedents
to trust in buyer-supplier, an empirical study was performed. Three different types of
trust were identified and their antecedents were modeled and tested. Data was collected
from 344 purchasing executives, who were members of the National Association of
Purchasing Management, through a mail survey instrument. A response rate of 23.8%
was achieved.

The study found statistically Significant support for the existence of three different
types of trust which all have high significant relationships with performance. The results
also show a strong positive association of non-coercive power with trust and strong
negative association of coercive power with trust. In addition to non-coercive power
dedicated assets were also found to have a significant impact on trust but it is unclear

whether trust or dedicated assets come first.

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................... vii
LIST OF FIGURES ....................................................................................................... ix
INTRODUCTION ......................................................................................................... 1
CHAPTER 1
OVERVIEW OF RESEARCH ................................................................................... 1
1.1 Problem Statement .............................................................................................. 2
1.2 Background ......................................................................................................... 5
1.2.1 Cooperative versus traditional buyer-supplier relationships .................. 5
1.2.2 Cooperative Buyer-Supplier Relationships ............................................ 6
1.3 Research Questions and Constructs ..................................................................... 9
1.3.1 Research Questions ................................................................................ 9
1.3.2 Types of Trust ........................................................................................ 10
1.3.3 Organizational versus Individual-based Trust ....................................... 11
1.4 Antecedents to Trust ............................................................................................ 12
1.4.1 Power/Dependence ................................................................................. 12
1.4.2 Cooperative Buyer-Supplier Relationship Planning Horizon ................ 12
1.4.3 Asset Specificity .................................................................................... 15
1.4.4 Risk/Environmental Uncertainty ........................................................... 15
1.5 Scope of this Dissertation .................................................................................... 16
1.6 Proposition/Hypotheses ....................................................................................... 18
1.7 Level of Analysis ................................................................................................. 19
1.8 Structure of Dissertation ...................................................................................... 19
CHAPTER 2
REVIEW OF THE LITERATURE ........................................................................... 22
2.1 Introduction ......................................................................................................... 22
2.2 Conceptual Paradigms of Trust ........................................................................... 22
2.2.1 Reliability ............................................................................................... 26
2.2.2 Competence ............................................................................................ 27
2.2.3 Goodwill ................................................................................................ 27
2.2.4 Vulnerability .......................................................................................... 29

iii

2.2.5 Loyalty-Based Trust ............................................................................... 31

2.2.6 Multiple Forms of Trust Exist ............................................................... 31

2.2.7 Combining Affect and Cognition-based Trust with Vulnerability ........ 34

2.2.8 Non-partisan Trust ................................................................................. 36

2.3 Context ................................................................................................................ 37

2.4 Differentiating Trust ............................................................................................ 40

2.4.1 Risk ........................................................................................................ 40

2.4.2 Commitments ......................................................................................... 42

2.4.3 Cooperation ............................................................................................ 43

2.4.4 Confidence ............................................................................................. 43

2.4.5 Predictability .......................................................................................... 44

2.5 Antecedent Conditions for Trust ......................................................................... 44

2.5.1 Power ..................................................................................................... 44

2.5.2 Dependence ............................................................................................ 45

2.5.3 Alliance Planning Horizon ..................................................................... 47

2.5.3.1 Economic (short-term) Planning Horizon ............................... 47

2.5.3.2 Mutually shared/Long Term Philosophy ................................ 52

2.5.4 Asset Specificity-Suppliers Perspective ................................................ 56

2.5.5 Risk ........................................................................................................ 62

2.5.6 Communication ...................................................................................... 64

2.5.6.1 Clarity ..................................................................................... 64

2.5.6.2 Communication Openness ...................................................... 66

2.5.6.3 Disclosure ............................................................................... 66

2.5.7 Contracts ................................................................................................ 66

2.5.8 Formal Trust Building Process .............................................................. 68

2.5.8.1 Does trust grow in stages? ...................................................... 68

2.5.9 Integrity/Moral Philosophy (antecedent and component of trust) ......... 70

2.5.9.1 Benevolence ............................................................................ 70

2.5.9.2 Integrity ................................................................................... 70

2.5.9.3 Honesty ................................................................................... 72

2.6 Summary of Literature on Trust .......................................................................... 72

2.7 Strategic Alliances ............................................................................................... 73

2.8 Theory Development - Supplier Buyer-Seller Relationships .............................. 74

2.8.1 General Systems Theory ........................................................................ 75

2.8.2 The IMP Interaction Model ................................................................... 75

2.8.3 Exchange Bargaining Power .................................................................. 76

2.8.4 Channel Relationships ........................................................................... 77

2.8.5 Markets and Contracts ........................................................................... 79

2.9 Summary of Alliance Literature .......................................................................... 81

2.10 Summary of the Entire Literature Review ........................................................... 82
CHAPTER 3

PRE-RESEARCH AND RESEARCH DESIGN ....................................................... 83

3.1 Proposed Levels of Trust Model ......................................................................... 83

3 .2 Exploratory Research .......................................................................................... 86

3.3 Trust Development Model ................................................................................... 87
3.4 Definition of Factors in Figure 1-2 ...................................................................... 90
3.4.1 Trust ....................................................................................................... 90
3.4.2 Power and Dependence .......................................................................... 93
3.4.3 Relationship time Horizon ..................................................................... 93
3.4.4 Asset Specificity .................................................................................... 94
3.4.5 Risk/Environmental Uncertainty ........................................................... 95
3.4.6 Trust and Performance ........................................................................... 96
3.5 Hypotheses .......................................................................................................... 96
3.6 Details of Measures to be Used ........................................................................... 100
3.6.1 Trust ....................................................................................................... 101
3.6.2 Types of Trust ........................................................................................ 102
3.7 Antecedents to Trust ............................................................................................ 107
3.7.1 Power ..................................................................................................... 107
3.7.2 Dependence ............................................................................................ 108
3.7.3 Relationship Time Horizon .................................................................... 108
3.7.3.1 Length of Relationships .......................................................... 108
3.7.3.2 Learning Races ........................................................................ 109
3.7.3.3 Mutually Shared Paradigm (continued commitment
even in adversity) .................................................................... 109
3.7.4 Asset Specificity .................................................................................... 111
3.7.5 Uncertainty ............................................................................................. 113
3.8 Performance ......................................................................................................... 113
3.9 Proposed Research Design .................................................................................. 115
3.9.1 Data Collection ...................................................................................... 116
3.9.2 Validity Issues ........................................................................................ 118
3.9.3 Reliability Issues .................................................................................... 120
3.10 Statistics ............................................................................................................... 121
3.11 Limitations of this Research ................................................................................ 124
CHAPTER 4
ANALYSIS OF DATA ................................................................................................ 126
4.1 Overview ............................................................................................................. 126
4.1.1 Description of the Sample ...................................................................... 126
4.2 Non-Response Bias ............................................................................................. 128
4.3 Missing Data ........................................................................................................ 129
4.4 Initial Data Analysis ............................................................................................ 130
4.4.1 Univariate Distributions ......................................................................... 130
4.5 Confirmatory Factor Analysis ............................................................................. 133
4.5.1 Trust Measurement Model ..................................................................... 134
4.5.2 Individual Versus Organizational Trust ................................................. 139
4.6 Measurement and Structural Model Testing of Trust Antecedents ..................... 141
4.7 Testing of Structural Models ............................................................................... 141
4.8 Summary and Conclusions .................................................................................. 150

CHAPTER 5

SUMMARY AND DISCLOSURES ........................................................................... 153

5.1 Overview of Research ......................................................................................... 153
5.2 Discussion of Research Findings and the Theoretical Implications

of Findings ........................................................................................................... 155

5.2.1 Calculus-based, Knowledge-based, and Identification-based Trust ...... 156

5.2.2 Individual Versus Organizational Trust ................................................. 161

5.2.3 Coercive and Non-Coercive Power ........................................................ 163

5.2.4 Dependence ............................................................................................ 1 65

5.2.5 Relationship Planning Horizon .............................................................. 167

5.2.6 Human Asset Specificity ........................................................................ 170

5.3.2.1 Asset Specificity Behavior ...................................................... 172

5.2.7 Dedicated Asset Specificity ................................................................... 173

5.2.8 Environmental Uncertainty .................................................................... 175

5.2.9 Performance ........................................................................................... l 77

5.3 Practical Implications .......................................................................................... 179

5.4 Limitations ........................................................................................................... 186

5.5 Directions for Future Research ............................................................................ 187

APPENDIX A ................................................................................................................ 189

APPENDIX B ................................................................................................................ 195

APPENDIX C ................................................................................................................ 197

BIBLIOGRAPHY .......................................................................................................... 199

vi

LIST OF TABLES

Table 2-1 ........................................................................................................................ 25

Table 2-2 ........................................................................................................................ 34

Table 2-3 ........................................................................................................................ 40

Table 4-1 Respondent’s Titles ...................................................................................... 127
Table 4-2 Respondent Companies’ Gross Annual Sales .............................................. 128
Table 4-3 Univariate Distributions of all Variables ...................................................... 130
Table 4-4 Transformed Univariate Distributions of all Variables ................................ 132
Table 4-5 Statistics for Manifest Variables of Trust (Measurement Model) ................ 133
Table 4-6 Trust Measurement Model Fit Statistics ....................................................... 136
Table 4-7 Trust Measurement Model Fit Statistics ....................................................... 139
Table 4-8 Analysis of Individual versus Organizational Trust ..................................... 140
Table 4-9 Statistics for Manifest Variables of Trust Measurement Model ................... 142
Table 4-10 Calculus-Based Trust ................................................................................... 143
Table 4-11 Knowledge-Based ...................................................................................... 145
Table 4-12 Identification-Based Trust .......................................................................... 146
Table 4-13 Additional Lagrangian Pathways Identified ............................................... 148
Table 4-14 Propositions and Hypotheses ...................................................................... 149
Table 4-15 Statistics for Manifest Variables of Trust Measurement Model ................. 151

vii

Table 5-1 Propositions .................................................................................................. 156

Table 5-2 Comparison of Theory with Results of this Study ....................................... 157
Table 5-3 Hypotheses 2-3 ............................................................................................. 164
Table 5-4 Hypotheses 4-5 ............................................................................................. 171
Table 5-5 Hypotheses 6-7 ............................................................................................. 176
Table 5-6 Actions of Trust ............................................................................................ 182

viii

LIST OF FIGURES

Figure 1-1 ....................................................................................................................... 7
Figure 1-2 ....................................................................................................................... 9
Figure 2-1 ....................................................................................................................... 23
Figure 2-2 ....................................................................................................................... 41
Figure 2-3 ....................................................................................................................... 80
Figure 3-1 ....................................................................................................................... 88
Figure 3-2 ....................................................................................................................... 92
Figure 3-3 ....................................................................................................................... 123
Figure 4-1 Trust Measurement Model .......................................................................... 135
Figure 4-2 Lewicki & Bunker’s Model/New Model Supported by this Study ............. 137
Figure 4-3 ....................................................................................................................... 143
Figure 4-4 Knowledge-Based Trust .............................................................................. 144
Figure 4-5 Identification-Based Trust ........................................................................... 147
Figure 4-6 Modified Identification-Based Trust Model ............................................... 149
Figure 5-1 Typical Trust Development ......................................................................... 180
Figure 5-3 Modified Identification-Based Trust Model ............................................... 185

INTRODUCTION

A search of the literature (chapter 2) on cooperative buyer-supplier relationships

across the disciplines of organizational theory, sourcing management, marketing, and

economics reveals a body of knowledge which provides managers with general

guidelines in each of the following areas:

1.

2.

When cooperative buyer-supplier relationships are appropriate.

Why cooperative buyer-supplier relationships are important.

. The benefits that can accrue as a result of cooperative buyer-supplier

relationships.
The risks associated with cooperative buyer-supplier relationships.
Appropriate supplier selection techniques.

While this information has provided value to managers, it does not address what

is arguably the most important and most difficult task in relationship management: how

to develop and maintain trust after the initial “honeymoon” period.

Three research questions that guide this investigation are:

1. What are the different types of trust experienced by purchasing executives in
cooperative buyer-supplier relationships?

2. What are the key antecedents that affect the level of trust experienced by
cooperative buyer-supplier relationships?

3. Are there significant differences between the three different types of trust and
cooperative buyer-supplier relationship performance?

Chapter 1

Overview of the Research

1.1 Problem Statement

An increasing body of research in operations and sourcing management,
economics, corporate strategy, and marketing has noted that firms are moving away from
traditional transaction based relationships to more cooperative buyer-supplier
relationships (Landeros 1988; Anderson and Narus 1991; Ring and Van de Ven 1992;
and Lewis 1996; Kumar 1996). A cooperative buyer-supplier relationship is defined as a
buyer and supplier uniting to pursue agreed upon goals, share benefits, management of
tasks and contribute to the relationship on a continuing basis (Yoshino & Rangan 1995).
This fundamental change in how buyers and suppliers communicate and work together
has resulted in a radical shift in the purchasing management and marketing paradigms
applied in managing interorganizational relationships.

Both the academic and practitioner literature discuss the need for more
cooperative interorganizational relationships in the United States. Firms in Japan have
been shown to significantly outperform European and North American firms in areas of
cost and technology through stronger buyer-supplier communication and
interorganizational coordination (Dyer 1994; Nishiguchi 1994). A common explanation

is that performance differences among countries is due to cultural differences, but

Nishiguchi (1994) has collected empirical evidence stating this is not the case. Although
evidence of a significant performance advantage with cooperative buyer-supplier
relationships exists, in depth investigations must be made into the specific processes
firms must apply to maximize benefits in cooperative buyer-supplier relationships. For
example, studies have shown that firms applying human asset specificity (dedicated
engineer to engineer interaction) to achieve higher levels of trust benefit from better cost,
quality, and delivery performance (Handfield, Bechtel, and Stimpert 1996; Monczka,
Petersen, Handfield and Ragatz 1996). Although these studies found that companies
benefit from human asset specificity and trust in cooperative buyer-supplier relationships,
they fail to specify the processes underlying the development of trust. In-depth studies
which delineate the actions leading to the development of trust are needed to help
managers develop successful relationships.

This dissertation seeks to investigate the interorganizational dynamics of trust
development encountered in established, ongoing buyer-supplier relationships. In
particular, the dissertation targets those buying and selling firms which are already
involved in viable, long-term cooperative relationships with their exchange counterparts.
While trust has been acknowledged to be a key component to effective buyer-supplier
cooperative relationships (Lewis 1990; Handfield et al... 1995), a clear understanding of
what trust is or how it can be fostered is not defined (Hosmer 1995; Barber 1983).

A search of the literature (chapter 2) on cooperative buyer—supplier relationships
across the disciplines of organizational theory, sourcing management, marketing, and
economics reveals a body of knowledge which provides managers with general

guidelines in each of the following areas:

1. When cooperative buyer-supplier relationships are appropriate.
2. Why cooperative buyer-supplier relationships are important.

3. The benefits that can accrue as a result of cooperative buyer-supplier
relationships.

4. The risks associated with cooperative buyer-supplier relationships.

5. Appropriate supplier/customer selection techniques.

While this information has provided value to managers, it does not address what is
arguably the most important and most difficult task in relationship management: how to
develop and maintain trust after the initial “honeymoon” period.

The research reviewed will be shown to focus primarily on the conditions when a
cooperative relationship may be appropriate, as well as the potential benefits and risks
inherent in COOperative buyer-supplier relationships. This research is helpful to managers
in organizations that are beginning the cooperative relationship development process, but
disregards those organizations that have been involved in cooperative relationships for
some period of time (Lewis 1990). This dissertation will test organizations that have
been involved in cooperative buyer-supplier relationships for at least one year, so that
trust development and trust management processes can be studied. Research is needed to
address the key linkages between risks/results and the actual process of relationship
development because the literature largely ignores the fact that relationships are dynamic
and changing in nature. Cooperative buyer-supplier relationships take considerable time
and personal involvement to manage (Hart et a1. 1986; Good 1988; Kamath & Liker
1994; Nishiguchi 1994), yet there is little theoretical or empirical work addressing how

managers handle these processes. Ring and Van de Ven (1994) point out that the manner

in which managers involved in interorganizational relationships handle the interaction
with their supplier can have a positive, neutral, or negative impact on the relationship.
This perspective is supported by the work in marketing on channel relationships (Heide
and John 1992), the work in strategy on cooperative interorganizational relationships
(Ring and Van de Ven 1994), and the work in sourcing management on supplier
development (Kamath & Liker 1994; Handfield et a1. 1995) which will all be discussed in
detail in chapter 11.

This dissertation research is designed to reduce the knowledge gap discussed
above through an investigation of the dynamic nature of ongoing buyer-supplier
cooperative relationships. It identifies the key drivers of trust as well as the different
types of trust that are crucial to buyer-supplier relationship performance. The dissertation
will seek to address how managers can effectively develop different types of trust in their
day to day efforts as they try to maximize performance of their cooperative buyer-

supplier relationships.

1.2 Background

1.2.1 Cooperative verg; traditional buyer-supplier relationships

Traditional buyer-supplier relationships exhibit a number of properties including:

0 Minimal communication beyond contract negotiations.

o Bargaining rather than problem solving between firms (N ishiguchi 1994).
o A constant threat of changing suppliers or multiple sourcing.

o No long-term commitment to the supplier.

0 Fully developed bidding specifications

A key buyer objective is to obtain the lowest possible price with little regard to other
competitive priorities (quality, delivery, etc..) . The short term nature of the relationship
does not allow either buyer or supplier to invest any significant resources into the
relationship which may provide significant long term benefits.

Trust is often not a critical issue in traditional relationships because if a supplier
proves untrustworthy the buyer only loses immediate inconveniences caused by
switching suppliers. Since the buyer has not invested resources into the relationship, the
cost of switching and level of trust is generally low. In certain instances, however, where
a buyer and supplier are at the beginning of what they both hope will become a
cooperative relationship the investment by both parties in trust built to that point can also
be lost. Traditional relationships often use detailed contracts and controls as a
replacement for trust (Gulati 1995). The detailed contracts provide motivation to each
party to perform through the threat of legal retribution. While this threat forces the
short-term compliance of the other party it does so with a loss in trust between parties

(Aguayo 1991; Whitney 1995).

1.2.2 Cooperative Bayer-Sypplier Relationships

Writers in many areas including marketing channels, organizational theory, and
purchasing have provided a variety of normative managerial guidelines instructing
managers on how to create and establish cooperative interorganizational relationships
(Dwyer, Schurr, & Oh 1987; Ring & Van deVen 1992; Nishiguchi 1994). Cooperative
buyer-supplier relationships are characterized by involvement of a buyer and a supplier in

a strategic long-term relationship. It is important in making a contribution to the buyer-

supplier relationship literature to understand that buyer-supplier relationships can range
from traditional contracts to joint equity ventures, as discussed by Yoshino and Rangan
(1995) (see Figure 1-1). While Figure 1-1 does an effective job of giving a detailed

description of the breadth of buyer-supplier relationships that can occur the figure does

not make clear distinctions between the various relationships.

Figure 1-1

[nterfirm Links

 

 

 

 

 

 

 

 

 

 

 

 

1
Contractual Agreements Equity Anangements
I
I l , l 1
I No New Entity . _
Traditional Nontraditional Creation of Entity D;:°:}:"°n
o n r y
Contracts Contracts Minority
‘ — Joint R&D Equity
-- Arm 543118111 _ Investments
Contracts — Joint Product
Development
F h' ' — Emmy M d
— ranc rsmg _ Long-term Swaps . . ergers an
. . Sourcing Agreements Nonst'rb/sstdiary J V Acquisitions
—- Licensmg Subsidiaries
*— Joint Manufacturing 0f MNCS
_. Cross: Joint Marketing
licensmg Fifty-fifty
*- Shared Distribution/ joint ventures
Service
Unequal equity
" Standards setting/ joint ventures

research consortia

Traditional contracts are used for commodity and irregular transactions that
include purchases of non-production or non-essential items. Traditional contracts are
based heavily on price and usually have low asset Specificity. Traditional contracts are
appropriate for spot purchases or non-critical repetitive purchase items where little trust

or exchange of information is needed among parties.

On the other end of the taxonomy, strategic alliance and equity relationships exist.
These alliances and equity relationships differ from contracts in two respects. First,
buying firms typically purchase not only their suppliers’ products or services but also
their suppliers’ systems and capabilities. Second, buying firms provide more than just
financial compensation to their suppliers. Buyers must share information with their
suppliers if alliances are to be successful for both parties (Ellram, 1991; Handfield, 1993;
1994), and they must also provide suppliers with guarantees of future volumes and prices
which may be tied to suppliers’ cost reduction and quality improvement efforts. As a
result, buyers and suppliers may become actively involved in the joint development of
new products, which involves co-location and cross-functional and cross-organizational
teaming (Trent, 1993). Such relationships are thus characterized by mutual
interdependence, close organizational cooperation, increased levels of trust, and a strong
tendency towards information sharing. Firms are currently moving to more cooperative
buyer-supplier relationships (non-equity relationships) to avoid the costs and inflexibility
of equity relationships (Lewis 1990). Equity ventures move relationships into a separate
category (Pisano 1989) and involve either the creation of a new and independent jointly
owned entity or a supplier taking a minority equity position in the relationship. Such
relationships provide financial hostages or ownership (Yoshino & Rangan 1995) which
provide a level of control which can substitute for trust. This dissertation focuses on
non-equity relationships only, since equity is often used under conditions of a lack of

trust.

The explanatory model for this dissertation in Figure 1-2 draws on prior research
in the fields of marketing, purchasing, organizational theory, and industrial economics.

The level of analysis will be most closely aligned with the marketing literature.

Figure l-2

2

1A
Coercive
Power

“13
HlC Pl

Non—coercive

Power Calculus-Based
Trust

Dependence

3A Knowledge-Based .__> Relationshi
. . > Trust (Organizational p
Relationship H33 Performance
A

Planning Horizon “7

Asset Specificity H4 Identification-Based /
Behavior ¥ Trust (Organizational H78

Dedicated
Asset Specificity H53

I
I
I
I
. ' Knowledge-Based Knowledge-Based
Envrronmcntal l Trust (Individual) = Trust (Or anizational
Uncertainty H6 . g
l
I
I
I
I Identification-Based Identification-Based
: Trust (Individual) = Trust (Organizational
r

E

I:

 

 

\.

1.3 Research Questions and Constructs
1.3.1 Research Questions
An investigation that tries to uncover and discuss every factor influencing trust in

cooperative buyer-supplier relationships would require resources and time beyond the

scope of this project. Therefore, after an exhaustive review of the literature on trust and
cooperative buyer-supplier relationships, three research questions were selected.
Three research questions that guide this investigation are:

1. What are the different types of trust experienced by purchasing executives in
cooperative buyer-supplier relationships?

2. What are the key antecedents that affect the level of trust experienced by
cooperative buyer-supplier relationships?
3. Are there significant differences between the three different types of trust and
cooperative buyer-supplier relationship performance?
The answers to these research questions will be addressed by conducting a survey
of a large sample of purchasing executives and statistical analysis of the data from those
executives. A conceptual model of the relationships between trust, performance, and

eight key antecedent conditions will be used to explain how trust is developed in

cooperative buyer-supplier relationships.

1.3.2 Types ofiTrust

The trust literature is beginning to recognize the existence of three types of
interorganizational trust. The three types of trust are calculus, knowledge, and
identification-based trust and are defined in detail below (Sheppard & Tuchinsky 1995,
Lewicki & Bunker 1995).

Calculus-based Trust

Trust based on control or assuring that the other party will do what they say.

Trust is sustained through a clear deterrent (punishment) and likely to occur if

trust is violated. The threat of punishment is likely to be a more significant

motivator than promise of a reward. Calculus-based trust often involves a high
degree of monitoring to assess whether a party is being opportunistic.

10

Knowledge-based Trust

Knowledge-based trust is grounded in the predictability - knowing the other party

sufficiently well so that the other’s behavior is anticipatable. Knowledge-based

trust relies on information rather than deterrence as a motivator.

Identification-based Trust

Identification-based trust is based on identification with the other party’s desires

and intentions. Trust exists because the parties effectively understand and

appreciate the other’s wants. A mutual understanding develops as each side
clearly understands the motivations and problems of the other side.
Knowledge and identification—based trust are similar to the concepts of cognition
(reliability and competence) and affective-based (faith and openness) trust recently tested
by McAllister (1993).

There exists only two known studies analyzing the concepts of calculus,
knowledge and identification-based trust. Lewicki and Bunker (1995) provide a model of
trust development in organizational relationships. Sheppard and Tuchinsky (1995)
suggest that the study of calculus, knowledge, and identification-based trust are needed if

the field of organizational behavior is going to be able to effectively help in the

description and theory development of new organizational forms.

1.3.3 Organizational versus Individgal-based Trust

Does trust exist at the individual or organizational level or both? Work in the
areas of marketing (Anderson and Weitz 1989; Dwyer Schurr & Oh 1987) purchasing
(Nishiguchi 1994; Lewis 1990) have adopted primarily and organizational unit of
analysis while organizational behavior (McAllister 1993; Rotter 1967) are has adopted
primarily an interpersonal unit of analysis. This dissertation will investigate whether

individual and organizational trust are unique constructs.

ll

1.4 Antecedents to Trust
1.4.1 Power/Dependence

Emerson (1962) was the first to establish a relationships between power and
dependence. Emerson stated that power and dependence had a reciprocal relationship.
Because of this reciprocal relationship this dissertation will analyze both dependence and
power. Specifically, this dissertation will analyze dependence of both firms or
interdependence. In the area of power there are five types of power according to French
and Raven (1959) and two of these types of power will be analyzed. Coercive and non-

coercive (referent power) powers will be analyzed.

1. 4. 2 Cooperative Buyer-Supplier Relationship Planning Horizon

Buyer-supplier planning horizons can be characterized as either adopting an
economic based (short term) paradigm or a mutually shared (long term) paradigm.
For many years, a short-term economic paradigm dominated the interorganizational
literature (Williamson 1975; Gulati & Nohria 1992). The economic paradigm was based
on two assumptions. The first assumption was that interorganizational relationships can
be accurately represented by the prisoner’s dilemma. In the prisoner’s dilemma, two
prisoners are given their choice of pleas without knowing what the plea of the other party
is. By analyzing the responses of people put in the prisoner’s dilemma, the optimal
strategy for a prisoner is “tit for tat” or do to the other party what they just did to you
(Gulati & Nohria 1992; Axelrod 1984). Serious concerns have arisen about the accuracy
of the prisoner’s dilemma in modeling cooperative relations (Fisher & Brown 1989;

Gulati & Nohria 1992). The concerns relate to whether the prisoner’s dilemma provides

12

a large enough “payout” to parties which cooperate. A large cooperation payout may
change the optimal strategy from tit for tat to long-term cooperation.

The second major short-term economic concept is transaction cost economics
(TCE) (Williamson 1975). Transaction cost analysis states that there exists transaction
and production costs of doing business between firms. Transaction costs are caused by
market inefficiencies in the form of opportunism, small number games, and bounded
rationality in individuals, while production costs are costs associated with business
transactions containing no inefficiencies. The problem with transaction cost analysis is
that it ignores organizations that work with each other repeatedly (Zaj ac & Olsen 1993;
Ring & Van de Ven 1992; Gulati 1995). TCE treats each transaction independently and
also assumes all parties are opportunistic.

The assumptions of opportunism and independent transactions are appropriate
when discussing markets versus hierarchies (Hirschman 1982), but the emergence of
more cooperative hybrid relationships such as strategic alliances are not accurately
depicted by this approach. In fact, recent studies (Ghoshal & Moran 1996) challenge the
effectiveness of TCE in any capacity. TCE’s assumption of transaction independence
ignores key factors such as trust (Lorenz 1988) which is developed when each transaction
is not seen as independent.

Over the past eight years, alternative paradigms to the economic paradigm have
emerged. One of these paradigms is referred to as mutually shared alliances (Gulati &
Nohria 1992). In mutually shared alliances, the payoff scheme in the prisoner’s dilemma
is challenged. Specifically, Gulati and Nohria (1992) feel that the payoff for the two

parties working together is substantially more than is assumed in the prisoner’s dilemma.

l3

A key omission in the economic model is the assumption of independence between party
transactions (Gulati & Nohria 1992). This omission is significant because experience can
engender trust among parties, and trust can limit the transaction costs associated with
future alliances (Granovetter 1985, 1992; Marsden 1981).

In the new mutually shared paradigm, a key underlying factor is the expected time
frame of the relationship. Parties see the relationship as long term (Nohria & Gulati
1992). In fact, Lorenz (1988) points out that trust is not necessarily a by-product of other
means but can be created intentionally. Lorenz (1988: 209) states that “The key in
partnerships is to sacrifice short-term gains for the long-term benefits of mutual
cooperation.” Empirical work by Heide and John (1992) found significant support for a
long term orientation impacting the ability of firms to work together.

While the short term economic paradigm ignores the role of trust in cooperative
buyer-supplier relationships, the long term paradigm posits that trust in a mutually shared
alliance can be a solution to transaction costs described by Williamson (1975). Lorenz
(1988) points out that transaction costs are directly linked to the possibility that actors
will behave opportunistically (Gulati 1995). The presence of trust can avoid the expense
of drafting comprehensive agreements for contingencies. Lorenz (1988) uses the
metaphor of trust as a lubricant and transaction costs as the economic friction. Trust can
also act as a response to opportunism and increase social bonds over time (Luhmann
1988). This proposition has been empirically tested by Campbell (1993) and partially
confirmed.

While many researchers are addressing the presence of a mutually shared

philosophy, no existing empirical evidence testing this philosophy exists.

l4

1.4.3 Asset Specificity

In this dissertation, three antecedents are posited to have an effect on the level of
trust present in COOperative buyer-supplier relationships: asset specificity, relationship
philosophy, and trustworthiness. Each construct will now briefly be described.

The argument of whether trust or asset specificity comes first is complex. There
is evidence that supports the notion of trust as a precursor to asset specificity (Lorenz
1988; Ring and Van de Ven 1994; Gulati 1995; Kamath & Liker 1994) and vice versa
(Handfield et al. 1995). Nishiguchi (1994: l 70) states that “ a strategy that points to asset
specific contractual relations cannot by itself guarantee that it will work. There must be
inter-firm relational mechanisms that enable them to function.” Nishiguchi implies that
some degree of trust or relational mechanisms must take place for effective asset
specificity to take place. This could explain the results obtained by authors such as J oshi
(1993), Campbell (1992) and Zaheer & Venkatraman (1995) which directly contradict
Nishiguchi’s results. Joshi interestingly found that relations with high dependence (asset
specificity with high switching costs) were more likely to see opportunism. This

dissertation seeks to establish a temporal order between trust and asset specificity.

1. 4. 4 Risk/Environmental Uncertainty

The relationship between risk (uncertainty) and trust draws discussion and
speculation but little empirical testing. Two theories on the relationship between risk and
trust have emerged. The first can be summarized by the statement “trust is a substitute
for risk” (Lorenz 1988; Lewis 1990; Williamson 1985). This theory is similar to Gulati’s

ideas on contracts as a mitigating factor to risk. Then as parties become familiar with

15

each other trust becomes more prevalent and contracts less prevalent. Therefore, we
would expect to see high levels of risk associated with low levels of trust.

A second theory looks at trust and risk developing together. Trust can only
develop with risk because an agreement without risk is a rational decision and does not
require trust (Luhmann 1988; Ring and Van de Ven 1992). In relationships then we
would expect to see high levels of both risk and trust because buyers would develop high
trust suppliers in parts or commodities where risk is high. This dissertation will only
survey cooperative buyer-supplier relationships where it is hypothesized there will be
high levels of risk and trust.

Similar to risk is the concept of uncertainty developed by Williamson (1975).
Williamson states that uncertainty is a contributing factor to information impactedness
(inefficient allocation of information) which creates situations of opportunism.
Environmental uncertainty involves areas including change in pricing, product features,
technology, and customer demand (Williamson 1975; Zaheer and Venkatraman 1994).

Risk can be broken down into two types of trust. First, is perceived risk which is
the risk that a party perceive in a situation. Second is actual risk which is the risk taken
in a situation. Actual risk can differ from perceived risk depending on the propensity of a
party to risk. Parties with a higher propensity to risk may take more risk than is
warranted in a Situation. This dissertation will investigate actual risk and its impact on
trust.

1.5 Scope of this Dissertation
This dissertation is focused on investigating existing and on-going cooperative

buyer-supplier strategic relationships. Prior research has largely focused on supplier

l6

identification and initial relationship development. This study will extend this body of
research by analyzing the dynamics of interorganizational trust development by providing
industry professionals with specific actions they can incorporate into current cooperative
relationship practice.

Prior studies have adopted a channel relationships view of trust (Heide and John
1988; Anderson and Weitz 1992), a transaction cost perspective (N ishiguchi 1994), or
explained trust as a cognitive artifact (Mohr and Spekman 1994), but no investigation has
analyzed the antecedents to trust. This study sets out to integrate:

o The four key antecedents to development of trust.

0 The three different types of trust that exist in interorganizational relationships.

0 The impact of the three different types of trust on cooperative buyer-supplier
relationship performance.

The explanatory model guiding the research is presented in Figure 1-2. This model
argues that three antecedent conditions influence trust development in cooperative buyer-
supplier relationships: buyer-supplier trustworthiness, buyer-supplier planning horizons,
and asset specificity. Several interorganizational studies (Mohr and Spekman 1994;
Gulati 1995) and buyer-supplier studies (Heide and John 1988) have tested the existence
of trust in buyer-supplier relationships or impact of trust on performance, but no studies
examine how buyers or suppliers can develop trust. This study will resolve a gap in the
literature between the positive impact on cooperative buyer-supplier relationship

performance and how a firm can develop trust.

1.6 Proposition/Hypotheses

Research on trust has left major unresolved theoretical and empirical issues
regarding the model in Figure 1-2. Based on the literature review in chapter two the
following proposition has been generated addressing the three different types of trust that
will be tested.

Pl. Three types of trust exist which are calculus-based, knowledge-based,
and identification-based trust which have a hierarchical relationship.

P2. Organization trust and individual trust are not unique.
Testing of this proposition and hypotheses below will help to bring theory and empirical
research closer together on the study of trust. The following specific hypotheses have
been generated:

HlA. Use of coercive power will be positively associated with calculus-based
trust.

HlB. Use of non-coercive power will be positively associated with knowledge-
based trust.

H1 C. Use of non-coercive power will be positively associated with
Identification-based trust.

H2. Dependence between buyer and supplier will be positively associated
with the level of identification-based trust in a supplier.

H3A. A long-term buyer-supplier relationship planning horizon will be
positively associated with the level knowledge-based trust in a supplier.

H3B. A long-term buyer-supplier relationship alliance planning horizon will
be positively associated with the level of identification-based trust in a

supplier.

H4. Asset specificity behavior focusing on information sharing and synergy
will be positively associated with identification-based trust in a supplier.

H5A. The overall level of asset specificity will be positively associated with
the level of knowledge-based trust in a supplier.

18

HSB. The overall level of asset specificity will be positively associated with
the level of identification-based trust in a supplier.

H6. Environmental uncertainty will be positively associated with the level of
identification-based trust.

H7A. Cooperative buyer-supplier relationship performance will be positively
associated with the level of knowledge-based trust in a supplier.

H7B. Cooperative buyer-supplier relationship performance will be positively

associated with the level of identification-based trust in a supplier.
1.7 Level of Analysis

The unit of analysis employed in the study of trust is the interpersonal and
organizational buyer-supplier relationship in analyzing trust and organizational in
analyzing the antecedents and performance. The interpersonal level and organizational
levels are appropriate given recent results from Zaheer et al. 1996 which indicates that

both levels are present.

1.8 Structure of Dissertation

The following chapters of the dissertation reviews the relevant literature, develops
hypotheses and models, and discusses the research design and methods. Chapter II will
review the literature on trust and cooperative buyer-supplier relationships and identifies
three antecedents to trust development. Chapter 3 will develop hypotheses, models and
measures, and explain data collection and data analysis methods to be used. The
dichotomy between a reliability-based trust and a faith goodwill-based trust which is not
new (Pennings & Woiceshyn 1987; Johnson-George & Swap 1982; Rempel et al. 1985).

Cognition-based trust recognizes that actors make a cognitive assessment of how

much trust and under what circumstance trust will be given (Lewis & Weigert 1985).

I9

The two key areas of cognitive based trust are competence, (the ability and resources to
carry out a task) and reliability of supplier’s performance over a period of time.
Competence is essentially the perception of the ability of party B to meet B’s
commitments set forth/promised to party A. The issues of trust and competence have
been addressed by several authors (Gabarro 1978; Butler 1991; Luhmann 1979; Ghoshal
& Bartlett 1995). Gabarro breaks competence-based trust into three key areas. First,
“specific competence” which is trust in the other’s specific function or area. Second,
“interpersonal competence” is the ability of a person to work with people or people skills.
Finally, “business sense” addresses a person’s experience, wisdom, and common sense.

Reliability can be broken down into requisite elements. Reliability is dependent
on prior contact with a party or experience. Several authors (Gabarro 1978; Lorenz 1988;
Fairholm 1994) mention time as a key element to reliability-based trust. A second related
major element is repeated interaction among the firms (Gabarro 1978; Good 1988; Gulati
1995; Ring & Van de Ven 1992). Repeated interaction and time (Lorenz 1988) then
leads to levels of confidence, consistency (F airholm 1988 ) and finally trust (Zucker
1986; Lewis 1990). Reliability can then lead to predictability which is confidence in
future actions (Fairholm 1988). Reliability must be based on integrity or honesty (Fisher
and Brown 1989) to be effective. Reliability based coercion or stress eventually creates a
sub-optimal relationship or total breakdown in the relationship.

Identification-based trust is proposed to consist of two elements. First, openness
with the other party (Gabarro 1978; Farris et al... 1973; Hart et al... 1986), and concern.
Openness is characterized by the perception on the part of the other party that the they

can share problems or information with confidence in the actions of the other party.

20

Second, affect based trust is characterized by concern for the other party (Mishra 1995).
The concept of concern is new but the similar concept of benevolence has been used
previously in describing trust and antecedents to trust (Mayer et al. 1995, Barber 1983,
Rempel & Holmes 1986, Butler 1991). Hosmer (1995: 392) points out that trust
“generally is accompanied by an assumption of an acknowledged or accepted duty to

protect the rights and interests of others.”

21

Chapter 2

Review of the Literature

2.1 Introduction

Chapter two will review the literature on trust and strategic supplier alliance
development. Figure 2-1 is a map of how the literature in trust and strategic supplier
alliance development was analyzed. Sections 2.1 through 2.6 analyze the trust literature.
Section 2.2 fits the trust literature into eight different conceptual paradigms that have
emerged over the years. Beyond conceptual paradigms the issue of context is addressed
in section 2.3. Section 2.3 analyzes the various circumstances under which trust occurs.
Section 2.4 clarifies the differences between trust and similar concepts such as confidence
or cooperation. Section 2.5 analyzes the trust literature theory on antecedents to trust

development and the final section summarizes the trust literature review.

2.2 Conceptual Paradigms of Trust

Inter-organizational trust is not a well-defined concept and defining it is
problematic (Zucker 1986:58). Other researchers note that there are many definitions of
trust, and the increasing interest and importance of trust warrants more consistent use of
the term (Hosmer 1995; Mayer et al. 1995; Lewicki & Bunker 1995). A problem with

creating a consistent use of the term is that trust is often discussed in a flippant manner.

22

Figure 2-1
2.1 Introduction

2.2 Conceptual Paradigms of Trust
2.2.1 Reliability
2.2.2 Competence
2.2.3 Goodwill
2.2.4 Vulnerability
2.2.5 Loyalty-based trust
2.2.6 Multiple forms of trust exist
2.2.7 Combining Affect and Cognition-based trust with Vulnerability
2.2.8 Non-partisan—based trust

2.3 Context

2.4 Differentiating Trust from Similar Constructs
2.4.1 Trust vs. Risk
2.4.2 Trust vs. Commitments
2.4.3 Trust vs. Cooperation
2.4.4 Trust vs. Confidence
2.4.5 Trust vs. Predictability

2.5. Antecedents to Trust
2.5.1 Power
2.5.2 Dependence
2.5.3 Alliance planning horizon
2.5.4 Asset Specificity
2.5.5 Risk
2.5.6 Communication
2.5.7 Contracts
2.5.8 Formal trust building process
2.5.9 Integrity/moral philosophy

2.6 Summary of Trust Literature
2.7 Strategic Alliances

2.8 Theory Development
2.8.1 General Systems Theory
2.8.2 The IMP Interaction Model
2.8.3 Transaction Cost Economics
2.8.4 Exchange/Bargaining Power
2.8.5 Channel Relationships
2.8.6 Markets and Contracts

2.9 Summary of Strategic Alliance Literature Review

2.10 Summary of Literature Review

23

Barber notes that:

“In both serious social thought and everyday discourse, it is assumed that the meaning of
trust and of its many apparent synonyms is so well known that it can be left undefined or to
contextual implications.” Barber (l983:7) Hosmer (op cit 380)

This observation is corroborated by the evolution of trust in the fields of industrial
economics, organizational behavior, marketing, and organizational theory. A comparison
of the various definitions of trust across fields shows that trust can be grouped into eight
conceptual paradigms (shown in Table 2-1). In paradigm one, authors posit that trust is a
cognitive predictability or reliability of another party. The second paradigm addresses
the competence of a party as a component of trust. In the third paradigm, a recognition of
trust as an altruistic faith or goodwill felt toward another party is proposed. The fourth
paradigm relates the concept of vulnerability to trust. Paradigm number five specifies
that loyalty-based trust exists when a partner consistently goes beyond the call of duty.
The sixth conceptual paradigm realizes that multiple components of trust exist, which are
defined by cognitive (reliability or task) trust and aflective (altruistic) faith trust. The
seventh body of theory adopts some of the most recent definitions of trust which combine
vulnerability and the notion of affect and cognition-based trust. The final conceptual
paradigm combines these different perspectives into a concept we define as non-partisan
trust. Non-partisan trust assumes that the actors do not focus solely on the
trustworthiness of the other party, but consider their relative vulnerability, as well as the
cognition and affect-based trust demonstrated by themselves and the other party. In the
following section, we identify each of these representative paradigms that are drawn from
the marketing, organization theory, sourcing, and organization behavior literature

streams .

24

Table 2-1

 

Body of Theory

Definition

Authors

 

1. Reliability

Time and experience are critical
elements in evaluating trust

Rossiter and Pearch 1975
Deutsch 1958

Rotter 1967

Gambetta 1988
Fairholm 1994

Lorenz 1988

Zucker 1995

Lewis 1990

Gulati 1995

Good 1988

 

2. Competence

Experience and wisdom displayed
by partner

Ghoshal and Bartlett 1994
Luhmann 1988
Butler 1991

 

3A. Goodwill (openness)

Confidence you can share
information or problems with the
other party

Pennings and Woiceshyn 1987
Granovetter 1985

Johnson Georges & Swap 1982
Ring and Van de Ven 1994
Farris et al.. 1973

Hart et al.. 1986

 

3B. Goodwill (Benevolence)

Accepted duty to protect the
rights of your partner

Mayer et al... 1995
Barber 1983

Rempel & Holmes 1986
Butler 1991

Hosmer 1995

 

4. Vulnerability

Being unprotected or exposed
while including an element of
uncertainty or risk

Deutsch 1958

Akerlof 1970

Barney & Hansen 1994
Klein, Crawford & Alchian
1978

land 1972

Holrnstrom 1979

Sabel 1993

Lorenz 1988

Gambetta 1988

 

5. Loyalty

A partner is not just reliable but
performs well in extraordinary
situations

Rempel & Holmes 1986
Larson 1990
Friedland 1990

 

 

6. Multiple Forms of Trust

There are more than one type of
trust

McAllister 1993

Mishra 1996

Gabarro 1979

Bromiley & Cummings 1996
Rempel & Holmes 1986
Ghoshal & Bartlett 1995

 

7. Combining Trust with
Vulnerability

8. The Future of Trust (Non-
Partisan proactive-Based
Trust)

Cognition and Affect -based trust
are combined with vulnerability
Trust is the primary attention to
your own trustworthiness and
secondary attention to your

 

 

 

partner’s trustworthiness

Mayer et al ..... 1995
Hosmer I995

Sabel 1993

Mishra 1995

 

 

25

 

2. 2. 1 Reliability

Reliability can be broken down into requisite elements. Reliability is dependent
on prior contact with a party or experience. Several authors (Gabarro 1978; Lorenz 1988;
F airholm 1994) mention time as a key element to reliable trust. A second related major
element is repeated interaction among the firms (Gabarro 1978; Good 1988; Gulati 1995;
Ring & Van de Ven 1992). Repeated interaction and time (Lorenz 1988) leads to levels
of confidence, consistency (Fairholm 1988 ), and finally trust (Zucker 1995; Lewis 1990).
Reliability then leads to predictability which is confidence in future actions (Fairholm
1988). While reliability is important, what motivates reliability is often more important.
Reliability must be based on integrity or honesty (Fisher and Brown 1989) to be effective.
Reliability based coercion or stress eventually creates a suboptimal relationship or total
breakdown.

A series of definitions define trust in terms of a firm or person’s reliability or
expectation of performance. Deutsch (195 8) created one of the first definitions of trust
which accepted the extreme position that for trust to be present, expected loss must be
greater than expected gain. Authors afier Deutsch extended trust as an expectation to
include situations where expected gain are greater than loss (Rotter 1967; Garnbetta
1988). Reliability can often be confused with predictability. Reliability primarily
addresses a party’s past behavior while predictability actually takes past behavior and
other information to address probabilities of future performance (Rossiter & Pearch
197 5). Reliability and predictability are closely related terms and definitions addressing

either term fall into this body of theory. Firms or people who meet a threshold level of

26

predictability can by definition be trusted. This paradigm is best described by the
following definition:

Trust is a range of observable behaviors and a cognitive state that encompasses
predictability (Rossiter and Pearch 1975).

2. 2.2 Competence

Competence is one’s perception of the ability of a party to meet commitments.
The issues of trust and competence have been addressed by several authors (Gabarro
1978; Butler 1991; Luhmann 1980; Ghoshal & Bartlett 1995). Gabarro breaks
competence based trust into three key areas. First, specific competence which is trust in
the other’s specific function or area. Second, interpersonal competence is the ability of a
person to work with people or people skills. Finally, business sense which addresses a
person’s experience, wisdom, and common sense. Ghoshal and Bartlett (1995) analyze
specific competence in technology or functional areas, which is similar to Gabarro’s
second type of competence. Other authors address competence and trust in more general
terms linked closely to a person or firm’s specific business experience or wisdom
(Luhmann 1980; Lewis 1990). This dissertation is primarily concerned with specific
functional and technological competence and interpersonal competence. Both of these

competencies must be in place for maximum effectiveness.

2. 2. 3 Goodwill

This was first defined by McAllister (1993), and is identified with a heavy
dependence on openness between people and emotional investment in the relationship.
Affect based trust could almost be confused with interpersonal or personal trust because

personal issues creep into the relationship in terms of problem solving, listening, and

27

sharing. A key distinction between cognitive and affect-based trust is that while cognitive
based trust may or may not exist at the interpersonal level, affect-based trust almost
always exists only at the interpersonal level. The importance of interpersonal relations is
recognized to be an important element of trust (McAllister 1993; Granovetter 198 5).

The move towards an affect-based trust construct was driven by Granovetter
(1985) who recognized that economic relations occur within social relations. In fact,
organizations can be seen as fertile soil for the growth of interpersonal relationships
grounded in faith and dependability (Pennings & Woiceshyn 1987; Johnson Georges &
Swap 1982).

While interpersonal relations are important, it is critical to break these
relationships down into their key elements. Authors describe the shift to affect-based trust
as ranging from an economic focused reliance on contracts to a psychological reliance on
relationship building (Ring & Van de Ven 1994). Affect-based trust is proposed to
consist of two elements. First, openness with the other party (Gabarro 1978; Farris et. al.
1973; Hart et. al. 1986). Openness centers on both parties feeling that the they can share
problems or information with confidence in the actions of the other party. Openness is
closely related to the interdependence between the parties which is addressed in more
detail later in this dissertation. Second, affect based trust considers the concept of
benevolence (Mayer et. al. 1995; Barber 1983; Rempel & Holmes 1986; Butler 1991).
Hosmer (1995: 392) points out that trust “generally is accompanied by an assumption of
an acknowledged or accepted duty to protect the rights and interests of others.”

One of the few empirical tests of affect-based trust is by Handfield et al. (1995),

which found significant relationships between asset specificity and affect-based trust and

28

also between supplier responsiveness and affect-based trust. This study lends empirical
support to the concept of affect-based trust and its role in buyer-supplier relations.
Beyond reliability or predictability, trust can also be defined in terms of a faith in
the goodwill of others. This faith or goodwill recognizes the importance of interpersonal
relations as an important element of trust (McAllister 1993; Granovetter 1985). This
new type of goodwill trust evolved from discussion and research on benevolence
(Solomon 1960; Strickland 1958), integrity (Lieberman 1981; Ring & Van de Ven 1992),

and honesty (Larzelere & Huston 1980) as antecedents to trust. For instance:

Trust is a faith in the moral integrity or goodwill of others, which is produced
through interpersonal interactions that lead to social-psychological bonds of
mutual norms, sentiments and friendships (Homans 1962) in dealing with
uncertainty (Ring & Van de Ven 1994).

Faith enables people to go beyond the available evidence and feel secure that a
partner will continue to be responsive and caring. Feelings of faith begin with
past experiences that show how much our partner cares” (Rempel and Holmes
1986)

2. 2. 4 Vulnerability

A key breakthrough in the use of the term trust is the relationship between
vulnerability and trust. Vulnerability is a key issue, because trust without some kind of
vulnerability simply cannot exist. If a party chooses a course of action that involves no
vulnerability then the firm has simply made a rational decision. One of the first
definitions to include vulnerability was provided by Deutsch (195 8), who stated that trust
involved choosing a course of action even if the probability of failure was greater than
50%.

There are three types of vulnerability which include adverse selection, moral

hazard, and hold up vulnerability. These three types of vulnerability range from inability

29

to accurately evaluate the quality of assets a partner says they will bring (adverse
selection) (1979 op cite Barney & Hansen 1994:176 Akerlof 1970), to the ability to
evaluate assets committed when a relationship exists (moral hazard) (op cite Barney &
Hansen 1994:176 Homstrom), to asymmetric investment by one party ( op cite Barney &
Hansen 1994:176 Klein, Crawford, and Alchian, 1978). land (1972) believed that trust
goes beyond expectation outcomes under uncertainty to expectation outcomes under
vulnerability. Zand’s definition is broader, in that he identifies two dimensions of trust
vulnerability. Vulnerability projects a feeling of being unprotected or exposed While
including an element of uncertainty or risk. If there is no uncertainty or risk, then the
party is freely giving the other party something. If there is no exposure by both sides,

then the firms are simply making a rational decision based on probabilities.

Sabel (1993:1133) Trust is the mutual confidence that no party to an exchange
will exploit another’s vulnerabilities.

land (1972) Trust is a risk relationship which increases the trustor’s
vulnerability.

Gambetta (1988: 217) Trust is a particular level of the subjective probability with
which an agent assesses that another agent or group will perform an action, both
before he can monitor such an action and in a context in which it affects his won

action. For trust to be present there must be the possibility for disappointment or
betrayal.

The discussion on vulnerability uncovers a key distinction that must be made
between trust and trusting behavior. Trust can exist without action, but trusting behavior
is the action taken, based on trust in another party. Lorenz discusses how vulnerability is

a key component of trusting action:

Lorenz (1988: 197) trusting behavior consists of action that (1) increases one’s
vulnerability to another whose behavior is not under one’s control, and (2) takes
place in a situation where the penalty suffered if the trust is abused would lead
one to regret the action.

30

A paradox which exists in was uncovered by Rempel and Holmes (1986), who wrote that
to be able to trust you must be willing to take the risk of trusting another party. To be a

party to trust, you must take this risk.

2. 2.5 Loyalty-Based T ru_st

Loyalty based trust occurs when after a period of reliable performance, goodwill
or faith is developed in the other party (Rempel and Holmes 1986). This good will
motivates a party to take altruistic actions in the relationship that they would not take in
other relationships. “Loyalty” based trust often occurs when a firm is not just reliable but

performs well in extraordinary situations.

“the first aspect of trust is predictability the second aspect of trust is
dependability. The feeling that your partner is a dependable person is based on
the emerging sense that he or she can be relied on when it counts. Paradoxically,
we can only be certain someone genuinely cares when a situation makes it possible

for that person not to care ..... dependability grows out of a special set of
circumstances that involve risk and personal vulnerability.“ (Rempel and Holmes
1986:30)

Loyalty based trust also occurs when individuals in both firms create strong interpersonal
bonds beyondjust their business relationship.

Trust is “confidence the other party could be relied on and confidence that extra
effort would be consistently made.” (Larson 1988:117)

“Trust is most typically promoted when a party to an interaction shows genuine
responsiveness to the needs of its partner” (Friedland 1990:317)

2. 2. 6 Multiple Forms of Trust Exist

Another group of authors realized that both altruistic based trust and reliability

based trust exist (McAllister, 1993). For instance:

There are two types of trust which are a business risk view which hedges against
moral hazards and advance selection and involves the predictability of one’s
expectations (op cite Ring & Van de Ven 1994: 93 Luhmann 79, Zucker 86) and
Confidence in another’s good will (op cite Ring & Van de Ven 1994: 93 Dore
1983, Ring & Van de Ven 1992).

31

Holmes and Rempel et al.. identify two different types of trust that are similar to
cognitive and affect-based trust, but also state there are three dimensions which each type
of trust. Holmes (1981, 1991) and Rempel, Hohnes, and Zanna (1985) look at
interpersonal and task trust. Both interpersonal trust and task trust is made up of three
dimensions: reputational, reciprocal, and accommodation trust. Reciprocal and
accommodation trust are emergent forms of trust. Other authors talk about calculative
and altruistic trust. Calculative-based trust is fostered by mutual hostages and built on
reputation ( Op cite Zaheer and Venkatraman 1995, Kreps 1990, Williamson 1993).
Williamson (1993) believes in only one type of trust which is altruistic trust.

Gabarro (1978) goes beyond these two types of trust and also identifies character,
judgment and competence as dimensions of trust. While character is similar to affect-
based trust and competence is similar to cognition-based trust, judgment is a new
dimension. However, judgment serves as a context for the other two types of trust rather
than a unique construct itself. Ghoshal and Bartlett (1995) describe the three keys to
developing trust from a neutral situation, in terms of:

1. A higher level of perceived fairness and equity in the company decision
making process

2. The broader level of involvement in core activities

3. An increase in the overall level of personal competence at all levels of the
organization.

Again, perceived fairness is similar to affect-based trust and competence is a component
of cognition-based trust, but involvement is a new construct. Ghoshal and Bartlett are the
only authors to discuss involvement, defined as the involvement of individuals affected

by the outcome of the decision-making process. While it is difficult to argue that this is

32

not an important point, participation may not qualify as an actual dimension of trust. For
the purpose of this dissertation involvement will not be used as a type of trust but rather
will be referred to as involvement or empowerment.

Rempel and Holmes (1986) describe three key components of trust as shown

below.

“Predictability, dependability and faith: each offers a different basis which
feelings of confidence in a partner can be built.” (Rempel and Holmes 1986)

Rempel and Holmes describe relationships as evolving through a continuum where
predictability is achieved first, then dependability and then faith. Finally, Mishra (1995)
discusses four dimensions of trust which include openness, competence, concern and
reliability. Competence addresses whether a manager will make the right decision.
Openness addresses whether a manager will give a straightforward account of events.
Concern discusses whether a manager acts in ways which benefit employees and
reliability involves a verification that actions and words are consistent. Mishra (1995)
believes that trust plays a key role in crises by allowing decentralization of decisions,
undistorted communication and collaboration. Lewicki and Bunker (1995) and Sheppard
and Tuchinsky (1995) describe three types of trust:
Calculus-based Trust Trust based on control or assuring that the other party
will do what they say. Trust is sustained through a clear deterrent (punishment)
and likely to occur if trust is violated. The threat of punishment is likely to be a
more significant motivator than promise of a reward. Calculus-based trust often
involves a high degree of monitoring to assess whether a party is being
opportunistic. Calculus-based trust is fragile and therefore any violation of trust
has potential to significantly alter or even end a relationship.
Knowledge-based Trust Knowledge-based trust is grounded in the
predictability - knowing the other party sufficiently well so that the other’s

behavior is anticipatable. Knowledge-based trust relies on information rather than
deterrence as a motivator. Knowledge-based trust develops over time as the

33

parties develop confidence in the trustworthiness of the other party. In
knowledge-based trust regular communication and courtship are key processes.

Identification-based Trust Identification-based trust is based on identification
with the other party’s desires and intentions. Trust exists because the parties
effectively understand and appreciate the other’s wants. A mutual understanding
develops as each side clearly understands the motivations and problems of the
other side. Each party comes to understand what they must do to sustain the
other’s trust. Identification-based trust is also characterized as having: collective
identity, colocation of personnel, creating joint products or goals, and commonly
shared values.
If we take a look of the bodies of literature to this point we can split definitions into who
they impact the primary party (me) versus the other party (them) in Table 2-2. Recent
definitions attempt to incorporate the affect and cognition-based elements together into
one definition. Table 2-2 indicates that recent theory in trust is beginning to emphasize
bilateral trust not just trust of the “the other party.” There is an emerging realization that
trustworthiness of yourself as well as the trustworthiness of the partner have significant

impact on the partnership.

 

 

 

 

 

Table 2-2
Emphasis Primarily on Emphasis almost
Yourself and Partner entirely on Partner
Area 4: Willingness to be Vulnerable Area 1: Reliability of the other party
Area 7: Cognition & Affect trust with Vuln. Area 2: Competence of the other party
‘ Area 8: Non-Partisan Trust Area 3: Goodwill of the other party
Area 5: Loyalty of the other party
Area 6: Multiple forms of trust

 

 

 

 

2. 2. 7 Combining Afi’ect and Cognition-based trust with Vulnerability
Two of the most recent definitions on trust come from articles attempting to
connect theoretical and ethical views of trust (Hosmer 1995) and the creation of a model

Of organizational trust (Mayer et al. 1995).

34

“The willingness of one party to be vulnerable to the actions of another party
based on the expectation that the other party will perform a particular action.”
(Mayer et al. pp. 712: 1995)

“Trust is the reliance by one person, group, or firm upon a voluntary accepted
duty on the part of another person, group, or firm to recognize and protect the
rights and interests of all others engaged in a joint endeavor or economic
exchange. “ (Hosmer pp. 393: 1995)

Both definitions are similar in that they emphasize a reliance or willingness on the
part of one party to be vulnerable to another to fulfill their commitment. The definitions
differ slightly because Mayer et al. use the term vulnerability which differentiates trust
primarily from a trustor’s point of view. Hosmer takes a different course where his
definition emphasizes a reliance of one party or another. This “reliance” is similar to
vulnerability because the reliance implies that the actor will incur costs if the other party
does not meet their obligations.

From these definitions several other consistencies between the definitions can be
observed. In order for trust to exist between organizations or individuals, antecedent
conditions exist which can have a positive, negative or no impact on trust. First, trust
exists across organizational levels from interpersonal trust to interorganizational trust.
Second, both definitions contain a cognitive and an affective element. Both address
cognitive trust, in stating that there is an expectation or predictability which addresses the
importance of reliability in the relationships. Both definitions also address the need of a
party to perform an action or protect the interest of the other party. Hosmer’s definition is
much clearer on this point. Third, trust is seen as an optimistic expectation that the other
party will perform. Trust does not assume the worst in people but instead takes an
optimistic view. Fourth, vulnerability is a key issue in discussing trust. Without

vulnerability, two parties are not risking anything and simply acting rationally. For

35

instance, consider the situation where two people have resources that they cannot use and
are going to throw away, yet they decide to combine these resources into a potentially
useful tool. Since neither party is vulnerable in this situation, there is no trust. If we take
the same situation and the parties are using very valuable resources, then trust will be
present. Finally, trust is not easy to identify or evaluate. It often takes parties several
iterations of contact before significant trust can be built. Trust is not something that is

immediately apparent in relations.

2. 2.8 NorL-partisan Trust

As Table 2-2 shows, theories of trust to this point have been constrained to a
focus primarily on the behaviors of the “other” party. All definitions to this point are
focused on the actions either cognitively or affectively of the “other” party. One question
which emerges from this discussion is whether trust is a function of thoughts about the
other party or a balance of actions by both parties? This dissertation will take a view of
trust similar to Sabel (1993). Sabel’s view is more objective and includes the perceptions
of both the primary party (me) and the other party (them). Finally, Sabel (1993) makes a
significant contribution by stating that a lack of opportunism by both parties in a
relationship where both parties are vulnerable constitutes a state of trust. There is a
tendency in relationships for each party to fixate on the behaviors of the other party while
overlooking their own behaviors. Parties eventually begin to require their partners to
embody ideals that they themselves do not embody. This lack of congruence among word

and action quickly leads to degradation of the relationship.

36

2.3 Context

Trust is a concept that exists in a wide variety of literature bases. A recent review
of the various definitions of trust by Hosmer (1995) found that to effectively analyze the
various definitions of trust it was necessary to look at the different contexts in which
authors defined trust. The following are the four major contexts which organizational
theorists have employed in defining and analyzing trust.

Individual Expectations: Trust reflects expectations of the individual in the areas

of competence, persistence and moral correctness (Barber 1983). This is the

context that this dissertation will adopt.

Interpersonal Relations: Trust is seen as a condition for cooperation and makes a
significant contribution to interpersonal relations.

Economic Transactions: Views trust as transactions among groups, companies,
or individuals primarily through distrust, taking a pessimistic view of people.

Social Structures: Trust is based on formal and informal rules in societies which
help to create cooperation among diverse elements in society.

Context is important to any discussion on trust because trust is not a clearly
defined term. Most authors explain how interpersonal relations eventually lead to agreed
upon expectancies and norms between firms. Larson (1988), for example, describes
interpersonal relations and expectancies as leading to social structures of norms in firms.
Similar to Larson, Gulati describes the concept of trust emerging from a series of prior
contacts or familiarity between firms, so that trust is built on certain norms and equity.
Macaulay (1963) takes a slightly different road than Gulati by stating that interpersonal
relations don’t just create familiarity but create pressure on interfirrn expectancies and

norms. While Gulati, Macauley, and Larson state that interpersonal trust precedes social

37

structures, Zucker (1986) argues that trust can be an interfirm phenomena, (not just
between individuals) with no causality present.

While there are relations among the individual expectations, interpersonal
relations, and social systems contexts, the economics context is not linked to the others.
In the economics context, trust is of secondary importance or is used as a lubricant for the

primary area of importance (transactions).

Economics sees trust as a public good a social lubricant which makes possible
production and exchange (Akerlof 1970; Arrow 1974; Op cit Dasgupta 1988)

Since the economics context assumes opportunism in other parties’, trust must be
developed through alternative mechanisms (hostages, contracts, and negative
consequences). Another way to develop trust is through the use of bilateral hostages
which provide incentives for parties to remain in the relationship (Borys and Jemison
1989; Kamath & Liker 1994). Economists look at trust from the other side (distrust) and

see how alternative control mechanisms or costly sanctions can deter distrustful behavior.

Trust can substitute for hierarchical contracts in many exchanges and serve as an
alternative control mechanism (Bardach & Eccles, 1989)

Gulati (1995) brings up the idea of deterrence based trust also discussed by
Shapiro et al., 1992. This perspective emphasizes utilitarian considerations when the
partner realizes that untrustworthy behavior will lead to costly sanctions. A universal
definition of trust does not currently exist so the context in which trust is used provides
vital information about a relationship. The list of contexts stated above is helpful in
understanding the various conceptualizations of trust which are discussed in this section.
While definitions could have been broken down by context, definitions were instead

broken down by a combination of contexts and content. Several content areas were added

38

to the context areas to provide a richer understanding of the key issues surrounding trust.
While the main focus of this dissertation is not a study of the different contexts of trust,
context must play a role in any study looking at trust. This dissertation will concern itself
primarily with the interpersonal and individual expectancies and how they play a role in
forming social structures in organizations. The definitions and explanation on context
show that research to date has not taken a partisan look at trust. The emphasis has been
on what the “other” party has done and their trustworthiness. Contexts in research to date
has also been constrained to economic transactions that affect individual expectations.
Research has represented trust as an expectation on the part of one party, with trust
playing a small role in economic transactions because the threat of opportunism by the
other party is so high. This dissertation will argue that this pessimistic view of people has
stifled the use of trust (Ghoshal & Moran 1995) and that a new viewpoint based on
interpersonal relations is needed.

Bromily and Cummings make an interesting distinction between trust as a
dimension and as a component or belief shown in Table 2-3 below. The authors use the
matrix below to test nine different combinations of trust dimensions and beliefs.

The Y axis identifies areas very similar to honesty, reliability, and opportunism
which are closely tied to the different concepts and paradigms of trust discussed prior in
section 2.2. The X axis shows the three contexts of trust: a feeling, cognition or
intention. This matrix then is able to simultaneously classify three different types and
three different contexts of trust. Empirical testing of these dimensions and contexts

showed significant differences among the three dimensions.

39

Table 2-3

Affective Cognition Intended
Behavior

 

Keep
Commitments

 

Negotiate
Honestly
Avoid taking
excessive
advantage

 

 

 

 

 

 

2.4 Differentiating Trust
In addition to describing what trust is, it is also important to identify what trust is

1.1.91.

2.4.1 Risk

The relationship between risk and trust is a subject of debate. Trust can be a
solution to risk (Luhmann 1988) or risk can combine with trust to determine firm
governance structure (Ring and Van de Ven 1992). Mayer et al (1995) provide the most
accurate relationship between risk and trust as depicted in Figure 2-2 below. Risk need
not be taken to trust another party, but risk must be taken if trusting action is to occur
(Mayer et al. 1995). In discussing risk, it is important to draw a distinction between risk,
trust, risk taking behavior, and trusting behavior. Risk is similar to trust in that it can
exist without any associated action. For example, the risk involved in developing a new
product can exist whether the firm actually decides to make the product or not. Trusting
behavior, on the other hand, must involve trust, risk, and vulnerability. If vulnerability to

a specific party is absent and only risk is present, then the party is involved in risk taking

40

Figure 2-2

 

 

Risk Taking
Behavior

 

 

 

/

 

RISK
(Environmental factors
outside relationship)

 

 

 

 

l Trusting
Vulnerability Behavior

1

 

 

 

 

TRUST

(Factors specific to relationship)

 

 

 

 

 

 

action as shown in Figure 2-2. An example is the decision to leave for work without an
umbrella hoping that it will not rain. While the person has clearly taken a risk there is no
other party that the person is depending on and he/she bears all the risk. If that same
person watches the news and decides to not carry an umbrella because the weather person
said the weather will be sunny, then the person trusts the weatherperson because they are

acting on their advice, and fiirther, they are performing a trusting behavior in the action of

41

not carrying an umbrella. Mayer et al. (1995) make a key distinction between trust and
risk. Trust takes into account the specific vulnerabilities involving the partner which a
firm is possibly going to enter into a trusting relationship with. Risk involves the
vulnerabilities that exist in the environment outside of the knowledge known about the
other party. Using the same example, not carrying an umbrella is a risk because there is
environmental vulnerability. If the person asks a coworker to bring an extra umbrella to
work (believing it may rain on the way home) then both trust and risk are involved.

There is still the risk due to the environment (whether it rains or not), but in this situation

trust exists (the person relies on the other party to bring an extra umbrella).

2. 4.2 Commitments

Commitment according to Dasgupta (1988) is similar to asset specificity.
Dasgupta states that commitments (exchange of contracts, people of honor, a promise) are
resources put forth by a party and are compared to expected ‘gain’ in order to decide
whether commitments should be entered into. Commitment is the physical, verbal, or
written manifestation of trusting behavior. Commitment and trust are often associated
with each other, but either can occur without the other. A party can make a commitment
to another without any trust in the other party. Similarly, trust can exist without
commitment. For instance, an individual can trust the president of the United States
without making any kind of commitment to him. However, trusting behavior must
involve a vulnerability or exposure which involves some type of commitment. The

relationship between trust and commitment is discussed in-depth later in this dissertation.

42

2. 4.3 Cooperation

Trust is easily confused with other words such as cooperation. While trust often
leads to cooperation, cooperation can exist without trust (Mayer et al. 1995). A party may
even cooperate with another party because they actually do not trust the other party and
want to avoid possible detrimental consequences. The relationship between trust and
cooperation has also been described as symbiotic with increased trust and cooperation
feeding off of each other (Ghoshal & Bartlett 1994).

Cooperation with trust by definition involves risk and vulnerability. Cooperation
without risk(Johnson-George and Swap 1982) or vulnerability (land 1972) is simply a
rational decision. Because confidence is closely related to cooperation, it is a concept

often confused with trust.

2. 4. 4 Confidence

The distinction between trust and confidence is not clear in the literature, but
Luhmann (1988) attempts to make a distinction. The author states that both confidence
and trust center on future expectations (Fairholm 1993) by a party, but in trust there is at
least one previous encounter on which feelings are based. A person who chooses among
alternatives is simply behaving rationally; however, if one of those alternatives is chosen
with the actor being aware of the risk of this choice relative to other choices, trust is
present (op cit Mayer et al. p. 713: 1995; Luhmann 1988; 102). With confidence, there
may be no previous experience with the other party to draw on. Without prior knowledge
of the other party, the party is acting on probabilities and making a rational decision.

Trust is a risk associated wherein the other party is known.

43

2. 4. 5 Predictability

Finally, trust can be compared to predictability. Clearly trust goes beyond
predictability (Deutsch 195 8). A person can consistently be dishonest (which is
predictable) but not worthy of trust. It is possible to have no trust in a person who is very
predictable. Predictability is actually a component of a specific kind of trust called

cognitive trust that will be discussed in detail in later pages.

2.5 Antecedent Conditions for Trust
2. 5. 1 Power

The area of power has similarities to the area of trust if you compare the types of
power from French and Raven (1959) referent, legitimate, and coercive power with the
three types of trust discussed prior. If we compare these three types of power (French &
Raven 195 9) with calculus-based, knowledge-based and identification-based trust

(Lewicki & Bunker 1995) there appears to be some similarities as shown below.

Coercive Power Calculus-Based Trust
Expectation that the other party Trust sustained through deterrence

will punish in situation of nonconformance punishment

Expert Power Knowledge-based Trust

Power due to Knowledge of perception Knowing the other party so that their
of knowledge in a given area behavior is anticipatable

Referent Power Identification-based Trust

Basis of referent power is in Identification with the other party’s
identification of one party with the desires and intentions

other, a feeling of oneness
The definitions above reveal a link between the areas of power and trust. Perhaps the
judicious use of power or even restraint from power use as discussed by Gaski and Nevin

(1985) can lead to the various types of trust discussed. This dissertation will analyze

44

coercive versus non-coercive power. Coercive power is the expectation of punishment
from another party unless there is compliance. Non-coercive power involves referent,

expert, legitimate, and reward power.

 

Power
(1959) French Jr. & There are different types of power including referent,
Raven legitimate and coercive.
Gaski (1984) There is not a strong relationship between power and

dependence. Coercive versus non-coercive power usage
has not been effectively tested yet

Gaski & Nevin (1985) Exercise of power has an effect on satisfaction and
conflict beyond the mere presence of power

Venkatesh et al.. (1995) Found significance between use of a particular influence
strategy and type of power in a relationship.

Heide (1994) The more unilateral power is in a buyer-supplier
relationship the higher the use of explicit contracts.

Lusch & Brown (1996) The higher the level of dependence of the supplier or
buyer on the other the higher the higher the use of
explicit contracting.

 

 

 

 

2. 5. 2 Dependence

Dependence has been observed two ways. First, the dependence of a party
(usually supplier) on another party (usually buyer). Second, the power one party has over
another usually due to a high percentage of a supplier’s output going to one buyer.
Several authors including Lascelles and Dale (1989), Dwyer Schurr and Oh (1987) and
Krause (1995) have addressed the issue of dependence from a volume perspective. They

hypothesize that the more a buyer buys from a supplier the more likely the buyer will be

45

Dependence

 

Treleven (1987)

Mohr & Spekman
(1994)

Emerson (1962)
Cadotte & Stern (1979)
Lascelles and Dale
(1989)

Frazier, Spekman and
O’Neal (1988)

Dwyer, Schurr and Oh
(1987)

Noorweir, John and
Nevin (1990)

Williamson (1975)

Heide (1994)

Etgar & Valency (1983)

Heide (1994)

Lusch & Brown (1996)

 

 

In situations with fewer suppliers buyers have fewer
opportunities to exploit suppliers

Interdependence is correlated with relationship performance

Power and dependence have a reciprocal relationship

The power dependence relationship determines the amount
of interdependence between parties

The volume of business with a supplier impacts the ability
of a buyer to impact a supplier

Coercive use of power can damage a relationship

Power is a function of dependence of parties on one another
Voluntary restraint from the use of power improves the
relational exchange norms of a relationship

Power assymetries will always be exploited

The more dependent a supplier is the higher the use of
explicit contracts is.

The more dependence that is present the more vulnerable
the weaker member is to the other.

The higher the degree of interdependence the more
commitment exhibited by both parties.

The more dependence a buyer has on a supplier the more
likely the buyer is to have a long-term orientation.

 

able to influence the supplier. Williamson (1975) first noted in transaction cost analysis

that in market advantages (based on power or dependence) lead to opportunistic behavior.

Empirical testing, however, has not supported these ideas.

46

 

Dependence of a party on another means that one party will have power over
another. Treleven (1987) notes that in markets with limited numbers of suppliers there is
less leverage for buyers in negotiating with suppliers. Resource dependence theory also
notes that when power between parties is in relative balance (high uncertainty)
organizations will attempt to create negotiated environments.

Beyond Williamson a growing number of authors assert that power can exist
without opportunism (Frazier et al... 1988; Noordeweir et al... 1990). Buyer’s do not
automatcially exploit power opportunities but may limit the use of power to strengthen a
relationship (N oordeweir 1990). Similarly, a buyer may refrain from using power to
avoid the negative consequences associated with exploiting a dependent supplier (Frazier
et al.. 1988). Beyond dependence is the concept of interdependence between two firms.
This interdependence is likely the product of a relationship between power and
dependence (Codotte and Stern 1979) which was first hypothesized by Emerson in 1962.
Emerson stated that power and dependence have a reciprocal relationship. In one of the
few empirical analyses of interdependence Mohr and Spekman tested the relationship
between interdependence and relationship performance however, a relationship was not
proven.

2.5.3 Alliance Planning Horizon

2.5.3.1 Economic (short-term) Planning Horizon For many years, an economic
paradigm dominated the interorganizational relationship (IOR) literature Williamson
1975; Gulati & Nohria 1992). The economic paradigm relied on two concepts: 1) The
prisoner’s dilemma and 2) Transaction cost analysis. In the prisoner’s dilemma, two

prisoners are given their choice of pleas without knowing what the plea of the other party

47

is. By analyzing the responses of people put in the prisoner’s dilemma, the optimal
strategy for a prisoner is “tit for tat”, (or do to the other party what they just did to you)
(Gulati & Nohria 1992; Axelrod 1984). Serious concerns have arisen about the accuracy
of the prisoner’s dilemma in modeling cooperative relations (Fisher & Brown 1989;
Gulati & Nohria 1992). These concerns focus on whether the prisoner’s dilemma
provides a large enough payout to parties which cooperate. A large cooperation payout
may change the optimal strategy from “tit for tat” to cooperation.

The second key economic concept relevant to the IOR literature is transaction cost
economics (TCE), put forth by Williamson (1975). Transaction cost economics states that
there are transaction and production costs of doing business between firms. Transaction
costs are caused market inefficiencies which are in the form of opportunism, small
number games, and bounded rationality in individuals. A problem associated with
transaction cost analysis is that it ignores organizations that work with each other
repeatedly (Zajac & Olsen 1993; Ring & Van de Ven 1992; Gulati 1995) and treats each
transaction independently. TCE also assumes all parties are opportunistic and in this
respect TCE is a “flat” paradigm. The assumptions of opportunism and independent
transactions may be appropriate in a discussion of markets versus hierarchies
(Hirschmann 1982), but the emergence of hybrid relationships such as strategic alliances
do not mesh with the TCE framework. The assumption of transaction independence
ignores key factors such as trust (Lorenz 1988) which is developed when each transaction
is not seen as independent. Economic theory is now taking into consideration the area of
game theory in looking at interorganizational relationships (Dasgupta 1988; Dixit and

Nalebuff 1991). Game theory adopts a less stagnant approach towards relationships than

48

TCE because each party begins to look at possible moves by the other party before taking
action. Game theory, however, still does not recognize the full payoff firms can attain by
cooperating, assuming parties are opportunistic.

Not all areas of economics ignore trust. Trust can occur incidentally in a
relationship. Several authors state (Axelrod 1984; Gambetta 1988; Bateson 1988) that
trust seems to evolve as firms cooperate over a period of time, but that initial cooperation
occurs at random. While incidental trust is possible, this dissertation recognizes and is
more concerned with trust that is proactively developed because proactive trust is under
the control of partnership participants and, therefore, can be controlled.

In the research performed by Hamel, D02, and Prahalad (1989) and Hamel and
Prahalad (1994), the emphasis in alliances is on one party winning and losing. As Hamel,
D02, and Prahalad (1989: 138) state “too much collegiality should set off warning bells
to senior managers. Gulati and Nohria (1992) point out that the thinking of Hamel, D02,
and Prahalad is an extended version of the prisoner’s dilemma where the firm that learns
the most wins. Hamel, Doz, & Prahalad argue that alliances should be seen as learning
races. CEOS or division presidents should expect occasional complaints from their
counterparts about reluctance of lower level employees to share information.” The major
emphasis in this paradigm is controlling the flow of information so that the “other side”
does not get unintended information. The problem with this view is that one must adopt
by definition security measures against opportunistic behavior like other economic
theories such as TCE. Hamel et al. treat each relationship independently, assume
opportunistic partners and try to maximize the benefits to a party often at the expense of

any further business with the other company. Their emphasis on preventative measures

49

contrasts with the notions of trust and commitment that are necessary for benefits to be
achieved, as noted in Lewis (1990), Gulati and Nohria (1992) and Barber (1983).

The assumption that parties are opportunistic underlies the short term economic
perspective. The solution to opportunism involves taking bilateral hostages, who
provide manufactured vulnerability of both parties in the relationship. A hostage induced
vulnerability must be distinguished from a synergistic driven vulnerability. Hostage
vulnerability usually has tangential value to the alliance. The key to effectiveness in using
hostages is the value of the asset to the donating party. From strategic vulnerability
perspectives assets are donated because they are valuable to the alliance. The asset may
also be valuable to the donator but the primary reason the donation is made is because of
the value of the asset to the relationship. Beyond hostages, Tesler (1980) argues that self
enforcing agreements are the answer to opportunism in situations of asset specificity and
argues that the threat to terminate the agreement is enough to deter opportunism. Tesler
(1980) provides three assumptions underlying his argument including:

1. The sequence of transactions are open ended

2. One off gains from cheating must always result in less than expected benefits
from cooperating

3. Players share knowledge in the ranking of respective gains to be had from
violating or upholding the agreement.

Lorenz (1988) points out two key flaws in this argument including:

1. If information is shared, trust is still involved because you must trust that the
information being provided is accurate.

2. Unexpected events do occur but are ignored in this model, and trust will be
essential in forming adaptations to contingencies in a jointly optimal way.

50

Economic/Short Term Perspective

 

Dasgupta (1988)

Hirschman (1982)

Lorenz (1988)

Gulati (1995)

Ring & Van de Ven (1992)

Zajac & Olsen (1993)

Axelrod (1984)

Fisher & Brown (1989)

Gulati & Nohria (1992)

Williamson (1975)

 

 

Looks at the impact of credibility, commitment, and
reputation on trust in a game theory environment.

An ideal market rests on large numbers of price
taking buyers and sellers with perfect information.
Trust and alliances have no place.

Trust is an economic lubricant which can reduce the
cost of friction caused by transaction costs.

Transaction cost analysis is flawed because it looks
at each relationship independently ignoring repeat
relationships.

Transaction cost analysis is flawed because it looks
at each relationship independently ignoring repeat
relationships.

Transaction cost analysis is flawed because it looks
at each relationship independently ignoring repeat
relationships.

Cooperation and trust can be attained through a tit
for tat strategy in the prisoner’s dilemma.

The prisoner’s dilemma is a poor model of
interorganizational cooperation.

The payoff of alliances is not represented well in
prisoner’s dilemma. Actually working together has
highest payoff not one cheating and other trusting.

Transaction cost analysis can provide a framework
for explaining why firms engage in markets or
hierarchies.

 

Schmitz et al.. (1995), Hart et al.. (1986) and Gabarro (1978) also found that reactions in

unexpected events are critical to trust failure or development. The economics point of

51

 

view highlights vulnerability of alliance partners (Gulati 1995) and assumes that all
partners are opportunistic.

2.5.3.2 Mutuallv Shared/Long Term Philosoplm Over the past eight years
alternative paradigms to the economic paradigm have emerged. One of these paradigms
is referred to as mutually Shared alliances (Gulati & Nohria 1992). In mutually shared
alliances, the payoff scheme in the prisoner’s dilemma is challenged. Specifically, Gulati
and Nohria (1992) feel that the payoff for the two parties working together is substantially
more than what is reflected in the prisoners dilemma. In mutually shared alliances greater
emphasis is placed on ongoing managerial practices. Trust, generosity, feedback and
repair are the guiding principles in mutually shared alliances. A key omission in the
economic model is the assumption of independence between party transactions (Gulati &
Nohria 1992). This omission is significant because experience can engender trust among
partners, and trust can limit the transaction costs associated with future alliances
(Granovetter, 1985, 1992; Marsden 1981). The economic theory, specifically TCE is
unraveled by Fisher and Brown (1989) who argue that in relationships the game theory
approach as depicted in the prisoner’s dilemma (Axelrod 1984) is not representative of
relationships (strategic alliances ) as a whole. The fatal flaw of transaction cost
economics is that it treats each alliance as independent ignoring the possibility of repeat

alliances (Ring & Van de Ven 1992; Zajac & Olsen 1993). Gulati (op cit. 1995: 86)

argues that:

“This is because 1. in an area such as understanding , a party is better off
understanding the other regardless if the other party is malicious because
information is gained. In the prisoner’s dilemma, if I am cooperative and the
other party stingy, I clearly am worse off. Second, in real life it is difficult to
measure the impact of a decision on another party. Can we measure an action by
another party as clearly equal to ours in cooperativeness?”

52

In the prisoner’s dilemma, payoffs are clear. Also, a strong bias enters into perceptions
of equity in actions by both sides. Often these relations lead to downward spiraling
substantive (results based) actions.

In the new mutually shared paradigm, a key underlying factor is the expected time
frame of the relationship. Partners see the relationship as long term (Gulati & Nohria
1992). In fact, Lorenz (1988) points out that trust is not necessarily a by product of other
means but can be created intentionally. Lorenz (1988: 209) states that “The key in
partnerships is to sacrifice short-term gains for the long-term benefits of mutual
cooperation.” The new hypothesized prisoner’s dilemma payoff that considers a mutually
shared paradigm, more accurately indicates the benefits that can accrue from synergies
between firms when they work together. Firms today must pay more attention to
mutually shared alliances incorporating a long term outlook, trust, generosity, feedback
and repair (Gulati & Nohria 1992).

A major component of the new relationship centered paradigm is congruence.
Congruence ensures that actions match words (Fisher and Brown 1989; Sinatar 1988,
Anderson & Weitz 1989; Schmitz et al. 1995). In fact, actions and not reputation are the
foundations of trust in a relationship (Fairholm 1994), and congruence between actions
and words are identified as an antecedent to trust (Sinatar 1988). Congruence is also a
problem in relationships when firms want the benefits of long term relationships yet their
behaviors are the same ones they had in their short term relationships (Fisher and Brown
1988)

While short term economic paradigms embodied in TCE ignore the role of trust,

an evolving long term paradigm identifies trust in mutually shared alliance as a mediating

53

Mutually Shared Long Term Paradigm

 

 

Campbell(1992)

Heide & John (1990)

Gulati & Nohria (1992)

Granovetter (1985,1992)
Marsden (1981)
Luhmann (1988)

Fisher and Brown
(1989)

Lorenz (1988)
Fisher & Brown (1988)

Anderson & Weitz
(1989)

Sinatar (1988)

Fairholm (1994)

Schmitz, Frankel,
Frayer (1995)

Fisher & Brown (1989)

Ganesan (1996)

Lewis (1990)

 

Trust mitigates opportunism and increases social bonds over
time.

Firms with a long term orientation were found to be better at
long term relationships.

Trust, generosity and feedback, and repair should be the
guiding principles of an alliance.

Trust can limit transaction costs with future alliances.
Trust can limit transaction costs with future alliances.
Trust can increase social bonds over time.

The optimal behavior in relationships is to be unconditionally
constructive.

Trust can be proactively developed.
Statements and actions must be congruent to build trust.

Pledges must be followed by enactment and reciprocity by
the other party for strong relations to develop.

Statements and actions must be congruent to build trust.

Actions and not reputation build trust in a relationship.
Congruence of corporate culture and actions impacts trust.

Partners continuously evaluate the consistency of promises
and statements of credibility and behavior over the life of the
alliance.

Ofien firms behave differently in short term relations or will
behave as though their goals from relations are short term
which leads to incongruent behavior to satisfy short term
needs.

Benevolence, dependence & trust impact long-term
orientation.

Certain firms have a more cooperative philosophy which is
an advantage yet avoid coercion. Know that it is in your best
interest to help a key partner and avoid coercion.

 

54

 

factor in reducing transaction costs as described by Williamson (1975). Lorenz (1988)
points out that transaction costs are directly linked to the possibility that actors will
behave opportunistically (Gulati 1995). The presence of trust can avoid the expense of
drafting comprehensive agreements for contingencies. Lorenz uses the metaphor of trust
as a lubricant and transaction costs as the economic friction. Trust can also act as a
response to opportunism and increase social bonds over time (Luhmann 1988). This
proposition was empirically tested by Campbell (1993). Finally, in the area of problem
solving the new relationship paradigm emphasizes an unconditionally cooperative and
proactive stance (Lewis 1990; Fisher and Brown 1988). For example, Lewis (1990)
proposes that firms should be as open to persuasion, while Fisher and Brown (1988)
advocate an unconditional constructiveness in relations (i.e. a party should give trust
conditionally but be unconditionally trustworthy.) Gulati and Nohria (1992) argue that in
a mutually assured paradigm it is best for partners to demonstrate continued commitment
to the relationship even if the other party is not pulling their weight. Trust, generosity, and
repair are the guiding principles as opposed to retaliation and restoration.

Similar to the mutually shared paradigm Anderson and Weitz in 1992 and Krause
in 1995 tested a construct called supplier commitment. Both Anderson and Weitz and
Krause define commitment as the willingness to make short-term sacrifices for long-term
benefits. Krause (1995) found supplier commitment to be an important antecedent to
supplier development efforts. Anderson and Weitz looked at the perception of
commitment of distributors from the manufacturer’s perspective. Anderson and Weitz

depict commitment as a never ending cycle beginning with the buyer’s perception of

55

commitment from the distributor. Based on the perception the buyer may make short

term sacrifices to realize long-term benefits.

2. 5. 4 Asset Specificity- Srmpliers Perspective
Nishiguchi (1994) discussed four different types of asset specificity. These four
types are:
Dedicated asset specificity represents discrete and/or additional investment in
generalized (as opposed to specific) production capacity in the expectation of

making a significant sale of a product to a particular customer.

Human asset specificity arises in a learning-by-doing fashion through long-
standing customer specific operations.

Site asset specificity refers to the successive stages that are immobile and are
located in close proximity to one another so as to economize on inventory and
transportation expenses. The fact that set-up and/or relocation costs are great
usually necessitates a bilateral exchange relationship between the parties over the
useful life of the assets.

Physical asset specificity refers to specific equipment such as molds or dies
specific to a relationship.

Several researchers have explored the types of asset specific investments made by firms.
In contrasting Japanese and unsuccessful partnership agreements, Nishiguchi (1994)
found that successful buyer-supplier relationships had greater levels of site asset
specificity and three times as much human asset specificity as their less successful
counterparts in the UK and US.

Another study addressing this issues was recently published by Dyer (1994), who
found that successful Japanese buyer-supplier relations invested substantially more in
site, dedicated, and human asset specificity. These results concur with Nishiguchi (1994),

but Dyer went one step further and suggested that site and dedicated asset specificity

56

actually drive human asset specificity. Dyer states that the most successful automotive
firms during the years 1982-1991 had suppliers who invested site asset specificity that
provided proximity to plants. This proximity helped facilitate human asset specificity in
the form of greater information and personnel exchange, leading eventually to superior
firm performance, and more responsive suppliers. The big advantages of asset specificity
according to Dyer are low transportation and inventory costs and improved product
development coordination through people swapping, close plant location, and investment
in customized assets (Dyer 1994: 174). Dyer makes the sweeping conclusion, however,
that the more interdependence between two firms the more dedicated assets should be
applied. He posits that dedicated assets or asset specificity is dependent on
interdependence. Dyer’s argument states that US. firms fall below the optimum level of
dedicated asset specificity involving human, site, and physical asset specificity. Kamath
and Liker (1994), however, describe a situation where suppliers earn more responsibility
by volunteering more asset specificity.

In sorting out the issue of trust and asset specificity, Nishiguchi (1994:170) states
that“ a strategy that points to asset specific contractual relations cannot by itself
guarantee that it will work. There must be interfirm relational mechanisms that enable
them to function.” Nishiguchi implies that some degree of trust or relational mechanisms
must take place for effective asset specificity to take place. This does not mean that trust
prior to the asset specificity decision is not also crucial.

Qualitative research by Nishiguchi points out that a firm with a problem solving
orientation performs better than firms with a bargaining orientation. Problem solving

involves a more long term perspective by the contractor and devotion to subcontractor

57

improvement programs and time with the subcontractor. Nishiguchi found that British
and US. supplier relations tend to be short-term and more likely to end if the automotive
industry incurs problems. Ring and Van de Ven (1992) identify a different process of
trust development and asset specificity than either of the above primarily automotive
examples. Ring and Van de Ven (1992) state that most firms use contracts initially in
relations and after time asset specificity and performance become primarily important.

Asset specificity can also serve as a signal to partner firms. Gulati and Nohria
(1992) discuss the concept of unilateral commitments, in which a firm makes a significant
commitment to the alliance to signal their support for the long time relationship. This
relationship is born out in work by Cooper, DeJong, Forsythe, & Ross, (1990) who found
a relationship between unilateral commitments and length of relationship.

While asset specificity can be seen as an opportunity for synergy among firms
(Dyer 1994; Nishiguchi 1994; Kamath & Liker 1994), it can also be seen as inhibiting
because inflexible investments by parties have been made. Borys and Jemison (1989)
raise the issue of how collaborators can use asset specificity or what they call “bilateral
hostages” in a cooperative or hybrid relationship. Williamson(1983) calls hostages
“credible commitments.” Kamath and Liker (1994), instead of using the term hostages,
describe asset specificity as “mutual entanglements which bind important suppliers and
their customers.” A key question then arises in the use of asset specificity is whether asset
specificity builds trust or creates inflexibility which narrows the strategic options of both
firms and makes both firms more sensitive or self conscious of the other’s trust breaking

maneuvers .

58

While there is significant support for the use of asset specificity, there is growing
evidence that asset specificity can be harmful to interorganizational relationships. Asset
specificity can create a situation where the ability to change directions is now limited.
This could explain the mixed results obtained by authors such as Joshi (1993), Campbell
(1992) and Zaheer & Venkatraman (1995). Joshi interestingly found that relations with
high dependence (asset specificity with high switching costs) were more likely to see
opportunism. This result was supported by Handfield et al. (1995) who found that power
asymmetries may overshadow asset specificity in buyer supplier alliances. Further
evidence of the complicated role played by asset specificity is provided by Campbell who
found that high asset specificity is the rule, not the exception, and that asset specificity
cannot uniquely explain integration. One possible solution to the problem above is
pointed out by authors who challenge the argument that asset specificity leads to trust. In
fact, it may be that trust must be in place for asset specific investment to occur (Lorenz
1988; Ring and Van de Ven 1994; Gulati 1995; Kamath & Liker 1994). Handfield et al.
(1995) found that human asset specificity had no impact on trust or supplier
responsiveness, which conflicts with studies by Nishiguchi and Dyer. However, it should
be noted that Handfield et al.. used a random sample of North American firms while both
Nishiguchi and Dyer found differences among Japanese firms versus US. and UK. firms.
While culture may be a moderating factor in these relations (Newman 1988, Prescutti
1992), Nishiguchi (1994) points out that Japan for many years had an exploitive
subcontracting paradigm which did not change without significant Japanese government

intervention on behalf of subcontractors in the 1950’s. Therefore, the trusting

59

relationships that the Japanese have developed are not culturally based as Dore (1983)
argues, but rather related to regulation.

The argument of whether trust or asset specificity comes first is complex. As
Kamath and Liker (1994) and Nishiguchi (1994) point out, asset specificity is a process
between firms that grows over time and likewise, trust is also a process that grows over
time (Dasgupta 1988; Hirschmann 1982; Gambetta 1988). A possible explanation for the
mixed results regarding asset specificity and trust is that a moderating influence may be
present. As discussed in the section above on the economic short-term philosophy toward
interorganizational relationships, asset specificity can be seen as an instrument to mitigate
opportunism. This type of application would have no relation to trust. In other instances
asset specificity is used because a certain amount of trust is already present and firms see
an opportunity to mutually benefit as Japanese firms studied by Nishiguchi (1994) and
Dyer (1995) have. In a specific example, human asset specificity can be applied for a
variety of reasons. Visits by the author to a big three automotive assembly plant reveal
that suppliers provide on-site support to the plants, but the support is required only by
poorly performing suppliers. This use of human asset specificity is a far cry from the use
of an on-site technical representative used by steel and electronics manufacturers whose
emphasis is on creating new technical solutions to problems and not ensuring base quality
and contract adherence issues.

In summary, the relationship between trust and asset specificity is complex. The
philosophy of the firms involved is a moderating factor as well as the presence or absence
of trust before implementation of asset specificity. A simple causal relationship between

asset specificity and trust is not an accurate reflection of current industry practice. It

60

Asset Specificity

 

 

Author Constructs/Relationships
Kamath & Liker Suppliers are on a continuum of four levels where depth of
(1994) involvement and responsibility are increased at each level.

Nishiguchi (1994)

Dyer (1994)

Ring & Van de Ven
(1994)

Handfield et al.
(1995)

Lorenz (1988)

Zaheer &

Venkatraman (1995)

Joshi(1993)

Campbell (1992)

Hostages

Borys & Jemison
(1989)

Williamson (1983)

 

Successful partnerships depend on the right balance of supplier’s
technological capabilities, a customer’s willingness to share
information, and both partner’s strategic requirements.

A strategy precedes asset specificity. Relational mechanisms must
accompany asset specificity.

Interdependence of firms (strategy) drives dedicated asset
specificity. Asset specificity reduces cycle time, increases quality,
and increases profitability.

Contracts facilitate trust initially, but later asset specificity and
performance become important.

Dedicated and site asset specificity lead to better supplier
responsiveness and human asset specificity.

Trust is expedient and must be used because comprehensive
contracting is impossible. Trust is particularly critical in asset
specific investment. This is an empirical investigation. Trust is
costly, but lack of trust is more costly.

Mixed results in analyzing the relationship between trust and asset
specificity. More business to a carrier to develop dependence led
to more not less trust.

High levels of dependence (difficulty of switching partners) lead
to more opportunism in a relationship while high levels of
relational norms (shared expectations) lead to less opportunism.

Asset specificity in interorganizational relationships is the rule not

the exception and alone cannot explain high levels of interfirm
integration.

“Bilateral hostages” can be an important part of hybrid
relationships.

Firms make “credible commitments in relationships.”

 

61

 

 

Kamath & Asset specificity provides “mutual entanglements” of between
Liker(1994) parties.

Gulati & Nohria Asset specificity can be used to signal commitment between
(1 992) parties.

Cooper et al. (1990) Unilateral commitments can be made to signal support.

Gulati & Nohria Unilateral commitments can be made to signal support.
(1992)

 

 

 

appears that asset specificity has the possibility of becoming a liability in relationships
where the more powerful partner demonstrates a short-term philosophy. If that same

party demonstrates a long-term philosophy, asset specificity may be a positive influence.

2. 5. 5 Risk

Butler (1991: 647) and Hosmer (Op cit 1995: 380) believed that it would be more
useful to study the conditions or determinants of trust than attempt further definition of
what he called a “global attitude” of the concept. Authors on trust address a variety of
antecedent conditions to trust in both individual and interorganizational relationships.
One area closely related to trust is the area of risk. Authors take one of three positions in
looking at risk in organizations. First, risk leads to trust (Lorenz 1988) or is an element
of trust (Fairholm 1994). If a firm performs well in a high risk situation, trust is
influenced. Second, trust leads to risk (Luhmann 1988; Lewis 1990) or risk taking
behavior (Mayer et al. 1995). A proven performer can be trusted over time in
increasingly risky situations. Third, trust and risk occur simultaneously in a complex

relationship where morality must be absent (Fisher and Brown 1989) and determine

62

 

governance structure (Ring & Van de Ven 1992;). Varying levels of trust and risk may be

present at different stages of a relationship with risk and trust affecting each other.

 

Risk

Luhmann (1988) Trust leads to risk taking behavior; trust and confidence
have a complex relationship.

Lorenz (1988) Trust is a solution to risk.

Lewis (1990) Key to trust is being able to rely on your partner in‘tough»
situations (more risk).

F airholm (1994) Willingness to risk close relations with others is an element
of trust. ' , ' -

Mayer et al.. (1995) The fundamental difference between trust and trusting»

behaviors is between a willingness to risk and actually
assuming risk.

Gulati (1995) Contracts lower risk.

Williamson (1985) Contracts or vertical integration lower opportunism or risk.

Ring & Van de Ven Risk increases with time. Risk and trust determine

(1992) governance structure (contract, alliance, vertical
integration).

Williamson (1985) Use of hostages can reduce risk in high risk situations.

Ring& Van de Ven Use of hostages can reduce risk in high risk situations.

(1992)

Fisher and Brown Determining how much you should trust another party

(1989) depends on a risk analysis of the situation and not the
morality.

Kog‘ut (1986) ' Mutual hostage positions in joint ventures provide

 

 

 

alignment of incentives.

 

63

From the literature, we can break the risk antecedent literature into two
philosophies which involve trust and distrust. We discuss the trust literature above where
trust can mitigate risk. First, trust and risk occur simultaneously as a relationship prospers
and higher trust and risk combined with satisfactory performance by both parties lead to
synergies. In cases where distrust is present, trust is achieved through costly means such
as contracts or asset (hostage) exchange. The hostage stream uses asset specificity and
control mechanisms such as contracts to mitigate opportunism. Some of the believed
benefits to using control mechanisms are better alignment of incentives of both parties
(Kogut 1986) and clear consequences to breach of duties (Gulati 1995). Finally, risk is
discussed in situations where distrust exists and contracts or mutual exchange asset

specificity can be used to mitigate short term opportunism (Gulati 1995; Williamson

1985; Ring & Van de Ven).

2. 5.6 Communication

2.5.6.1 Clarity Firms must not only exploit opportunities to learn in alliances but
must also monitor information flows (Yoshino & Rangan 1995; Lewis 1990; Hamel Doz
& Prahalad 1989). Firms must clarify of what information will be shared and not shared
(Lewis 1990; Hamel Doz & Prahalad 1989). High levels of trust require clarity on
disclosure, specifying levels of access as opposed to deception (Fisher and Brown 1989,
Lewis 1990). Through clarity of information, trust will be developed. When “touchy”
issues are raised constructively, a party can be confident that the response will also be

constructive (Lewis 1990).

64

Communication

 

 

"Zahe'er & Venkatraman
(1995)

Perceived Openness &
* Fairness

Lewis (1990)
Gabarro (1979)
Spekman & Mohr
(1994)

Gabarro (1979)

Disclosure

Altman & Taylor(1979) -

Stephens (1993)

Clarification

3 Lewis (1990)

Hamel, Doz & Prahalad
(1989)

Yoshino & Rangan
3(1995)
Perceptions ' ‘

Gulati'&Nohria'(1‘992)

 

~ Trust impacts the‘joiritaction‘ or firms (planning and»

forecasting together).

Good communication increases trust which Increases
predictability and reduces surprises.

, Informal- meetings' are the best‘communicatidn tool- for -

expectations, not formal meetings.

Communication openness was positively correlated with

' partnership performance.

Proactive communication by both sides at the-beginning
of the relationship is key.

Subordinate openness or candor are key as and symbolic
gestures at the beginning of the relationship'arekey.‘

Trust is a critical antecedent to disclosure and-reciprocity

trust.

Trust and disclosure are positive significant antecedent
conditions for relational acceptance (partially supported).

It is critical to clarify technologies, information, and other I

rescurces that will’participate in the allianceand-which

will not.

It is critical to clarify technologies, information, and other
resources that will participate in the alliance and which
will not.

You must monitor going out into an alliance.

‘ The perceptions of both parties are important and the

perceptions of yourJartner need to be managed.

 

65

 

2.5.6.2 Communication Openness Open communication and understanding of
the other side’s interests is an important component of a healthy relationship (Fisher &
Brown 1989). Openness has been consistently found to be an antecedent to strong
relationship performance (Gabarro 1978; Mohr & Spekman 1994).

2.5.6.3 Disclosure Altman and Taylor (1973) contend that trust is a critical
antecedent to disclosure and to reciprocal trust. Stephens (1993) performed one of the
first empirical studies partially supporting trust and disclosure as positive significant
antecedent conditions for relational acceptance. Empirical testing of this proposition has
only identified the impact of communication on trust (Mohr & Spekman 1994; Gabarro
1978), but Altman and Taylor(1973) and Stephens (1993) suggest that trust leads to
increased disclosure. Overall, firms in an alliance must use communication, clarity,
openness, and disclosure to actively manage the perceptions of other parties (Gulati &

Nohria 1992).

2. 5. 7 Contracts

The use of contracting in alliances tends to decrease as familiarization among the
firms increases (Gulati 1995). The term contract can be used to cover a multitude of
costs associated with alliances including: negotiating, monitoring, and enforcing contracts
(Yoshino and Rangan 1995). Contracts must also cover a multitude of contingencies or
accommodate disagreements between partners in terms of the contributions that partners
make (Williamson 1975; Arrow 1971). A variety of different types of contracts exist
including contingent claims contracts, short-term contracts, long-term contracts, spot

contracts, sales contracts, blanket contracts, and systems contracts. Support exists that

66

contracting does not build trust (Handfield et al. 1995; Van de Ven & Walker 1984);
however, there is empirical support that contracts can lower risk (Gulati 1995). There is
no argument in the contracting literature stating that contracting is not a solution to risk in
relationships (Gulati 1995). The literature also shows that contracts do not positively
impact trust (Handfield et a1 1995) and may negatively impact trust development (Van de
Ven & Walker 1984; Shapiro 1987). What is missing in the contract literature is a
discussion of the different types of contracts that can exist. Possibly the lack of
distinction in the contract literature between the different types of contracts has led to the
unspectacular results of researchers. The size of the contract can also significantly impact
trust. On the Global Procurement and Supply Chain Benchmarking Initiative several
companies operate their supplier alliances without contracts while others choose to still
have a one or two page signed document. A clear benefit of alliances is the significant
reduction in work, size, and scope of contracts or in extreme cases elimination of
contracts altogether.

Contracts

 

Handfield et al... (1995) Contracting does not impact affect based trust.

Gulati (1995) Familiarization reduces need for extensive contracting
Van de Ven & Walker Excessive contracting and formalization leads to

(1984) conflict, distrust, and inhibition of affect based trust.
Shapiro (1987) Impersonal institutions used to controlling trust increase

the opportunities for abuse by agents.
Gulati (1995) Contracts lower risk

Rouseau (1992 Four different types of contracts exist

 

 

 

 

67

The most recent research on contracts is beginning to explore the existence of four
different types of soft or indirect of contracts. Psychological contracts are viewed from
within the individual’s perspective. Normative contracts are group oriented and viewed
from within. Implied contracts are individual oriented and viewed from outside (society)
and social contracts are group oriented and viewed from outside. Lusch and Brown
(1996) and Ring & Van De Ven (1994) make the distinction between discrete exchange
which involves explicit contracts and relational exchange which involves normative

contracts.

2. 5.8 Formal Tryst Building Process

2.5.8.1 Does trust grow in stages? Trust does not appear quickly but takes time
to grow (Fairholm 1994). In the same manner, trust grows as it is used more and more,
(Gambetta 1988; Hirschman 1982; Dasgupta 1988) but without increasing responsibility
or a constant use and expansion of the trust in a relationship, trust will dissipate (Hirsch
1978; Hirschman 1984a; Bateson 1986; Dasgupta 1988).

Trust cannot be demanded or coerced but must be earned and developed over time
(Fairholm 1994; Gabarro 1979) and is built as firms interact repeatedly (Good 1988).
Lorenz (1988) found that clients determine the trustworthiness of a subcontractor not
only through reputation but also through his/her own experience (cognition). While time
plays a role in building trust, it is important that suppliers move through stages with
levels of trust and responsibility changing at each stage (Good 1988; Kamath & Liker
1994). Stages are carefully managed, so that the buyer can simultaneously reduce

exposure to risk while developing the supplier.

68

The importance of exceeding and not just meeting the minimum customer

requirements is also important for firms moving to trusted or preferred status (Kamath &

Formal Trust Building Process

 

Author Construct/Relationship

Gambetta (198 8) Trust grows with use.

Hirschmann (1984a) Trust grows with use.

Dasgupta (1988) Trust grows with use.

Fairholm (1994) Trust can’t he demanded but must be earned over time.

Gabarro (1978) Trust can’t be demanded but must be earned over time.

Bateson (1986 Trust is depleted if not used.

Hirsch (1977) Trust is depleted if not used.

Kamath & Liker (1994) Suppliers move through stages of trust which involve
specific behaviors and investments by buyer and -
supplier. Suppliers that move up exceed expectations.

Nishiguchi (1994) Suppliers are tiered and trust plays a crucial role.

Hart et al ..... (1986) Trust develops through a process similar to Maslow’s
hierarchy as seen through the inferior partner’s eyes.

Good (1988) Trust is built through stages.

 

 

 

Liker 1994). Kamath and Liker (1994) and Nishiguchi (1994) describe a continuum

between the supplier and buyer. The involvement of the buyer as an intermediary

 

between the customer and supplier dissipates as the relationship progresses through the
four stages. However, the buyer must be careful to control the amount of responsibility
also called to the supplier as they move through the stages. Up to this point, we have

reviewed literature on formal trust development primarily through the buyer’s eyes.

69

Just as important is the perception of development through the eyes of the
supplier. Hart et a1 (1986) argue that suppliers use a framework similar to Maslow’s
hierarchy as a framework of seeing how trust develops with a buyer. As suppliers gain
more commitment from the buyer, trust moves from simply meeting physical needs of

survival to higher order psychological and actualization needs.

2. 5. 9 Integrity/Moral Philosophy (antecedent and component of trust)

Integrity, honesty, benevolence and other areas discussed in this section have been
discussed as components as well as antecedents of trust. The integrity based perspective
of trust identifies honesty, integrity, and benevolence as key components of trust.
Fairholm (1994) covers all three areas in positing that trust is equivalent to integrity
which embodies honesty, authenticity, and truthfulness.

2.5.9.1 Benevolence Benevolence is a key antecedent to trust (Rosen & Jardee
1981; Mayer et al. 1994). Benevolence is the degree to which the trustee wants to benefit
to the trustor (Mayer et a1 1995). Benevolence can be applied to resources where a party
helps another to attain more from their resources (Rogers 1964). Benevolence can also be
a reliance on another party to act in both of the parties’ best interest, especially if it is
difficult to monitor the actions of the other party (Lewis 1990). Finally, benevolence can
be seen as the opposite of opportunism (Hovland 1953). Firms or individuals foregoing
the opportunity to put themselves ahead of the organization or partnership are perceived
as being benevolent.

2.5.9.2 Integrig Integrity is a key antecedent to trust (Butler & Cantrell 1984;

Butler 1984; Mayer et a1 1995; Gabarro 1978). Mayer et al. (1995) define integrity as the

70

MORAL PHILOSOPHY

 

‘ Honesty
Gabarro (l 97 8) Honesty is critical to a healthy relationship.
Lewis (1990) Trust grows from honest conduct (1990).
Hovland (1953) Willingness to lie impacts trust.
Integrity

Butler & Cantrell (1984) Integrity is an antecedent to trust.

Butler (1991) Consistency, integrity, and fairness are antecedents to
trust.

Gabarro (1978) Character (including integrity) is a basis of trust.

Lieberman (1981) Integrity is an antecedent factor to trust.

Mayer et al. (1995) Integrity is the perception of the trustor that the trustee
adheres to a set of principles that the trustor finds
acceptable.

Benevolence

Rosen & Jardee (1981) Benevolence (willingness to put organizational goals
ahead of individual goals) is antecedent to trust.

Rogers (1964) One participant intends that another attain more
appreciation of and use of his or her personal resources.

Mayer et al. (1995) Benevolence is the extent to which the trustee is believed
to want to do good to the trustor.

 

 

 

 

perception of the trustor that the trustee adheres to a set of principles that the trustor finds
acceptable. McFall (1987) made the critical distinction between personal and moral

integrity. Personal integrity means a trustee adheres to a set of their own principles while
moral integrity implies that the trustor believes that the principles that the trustee follows

are acceptable.

71

2.5.9.3 Honesty Honesty is really the manifestation of congruent actions.
Honesty can be seen primarily as a term to further describe the presence or absence of
integrity. Honesty is the perception of that the other party is telling the truth. Honesty is
critical to the proper functioning of relationships (Lewis 1990; Gabarro 1978). This

perception is heavily dependent on the consistency or congruence of actions by the other

party.

2.6 Summary of Literature on Trust

This section reviewed the major conceptual and empirical findings and models
relating to effective working relationships. The literature has focused primarily on
interpersonal relationships between individuals although these exists a sizable body of
literature on interorganizational trust. The primary focus of this work is on understanding
the different types of trust and antecedents to trust, including communication, formal trust
development, organizational philosophy, trustworthiness, moral philosophy, and asset
specificity. The literature review also traces the development of the term “trust” through
the past four decades to its present context. While the literature provides useful insights
into important determinants of trust in relationships, key antecedents and consequences of
these factors have not been examined (especially in an interorganizational buyer-supplier
context).

In addition, several new conceptual antecedents including relationship philosophy
and formal trust building process have been identified but are not yet empirically tested.
While there is ample conceptual support for these antecedents (N ishiguchi 1994; Kamath

and Liker 1994; Gulati & Nohria 1992), they have failed to undergo theory testing.

72

Extant research has recognized the importance of trust in interaction between
parties in close relationships but has typically focused on one party’s perspective. Future
research needs to consider both buyer and supplier perspectives when analyzing trust.
While there has been investigation into the importance of trust and different types of trust,

firrther clarification of the antecedents of trust need to be made.

2.7 Strategic Alliances

A second literature base relevant to the study of trust in buyer-supplier alliances is
the research on buyer-supplier alliances. Writers in many areas including industrial
marketing , organizational theory, and purchasing have provided a variety of normative
managerial guidelines instructing managers on how to create and establish cooperative
interorganizational relationships. The primary focus of this dissertation is the role of trust
in strategic buyer-seller alliances. These alliances are characterized by the involvement
of a buyer and a supplier in a strategic long term relationship. It is important in making a
contribution to the strategic alliance literature to consider a framework or continuum that
compares and contrasts various types of interorganizational relationships. At one end of
the continuum in Figure 1-1 are traditional contractual relationships, which are irregular
transactions that include purchases of non-production or nonLessential items. Traditional
contracts are used in buying items that are heavily price competitive. Transaction cost
theory predicts a high degree of opportunism to be present in traditional contracts
(Williamson, 1975; 1979; 1985). Because of the infrequent nature of such purchases, the
amount of information shared between parties is minimal, and the level of trust that the

buyer places in the supplier is similarly very low.

73

Beyond contracts on the continuum in Figure 1-1 are partnerships, strategic
alliances, and equity relationships. These alliances and equity relationships differ from
contracts in two respects. First, buying firms typically purchase not only their suppliers’
products or services, but also their suppliers’ systems and capabilities. Second, buying
firms provide more than just financial compensation to their suppliers. Buyers must share
information with their suppliers if alliances are to be successful for both parties (Ellram,
1991; Handfield, 1993; 1994), and they must also provide suppliers with guarantees of
future volumes and prices which may be tied to suppliers’ cost reduction and quality
improvement efforts. As a result, buyers and suppliers may become actively involved in
the joint development of new products, involving co-location and cross-functional and
cross-organizational teaming (Monczka and Trent, 1993). Such relationships are thus
characterized by mutual interdependence, close organizational cooperation, increased
levels of trust, and a strong tendency towards information sharing. Equity alliances move
relationships into a separate category (Pisano 1989). Equity alliances involve either the
creation of a new and independent jointly owned entity or a partner taking a minority
equity positions in the other partner. Equity alliances provide financial hostages or
ownership, which provide a level of control that substitute for trust. This dissertation is
not concerned with equity relationships. Firms are currently moving to strategic alliances
(non-equity relationships) to avoid the costs of equity relationships and avoid the
inflexibility of equity relationships (Lewis 1990).

2.8 Theory Development - Strategic Buyer-Seller Relationships
Several theories have emerged to explain the phenomena of strategic buyer-

supplier relationships.

74

2. 8. 1 General fistems Theory

General systems theory suggests that strategic , long term collaborative buyer-
supplier relationships will deteriorate if not maintained because of internal and external
environmental change. Over time, collaborative relationships will become more stressful
unless either side takes action. The amount of stress in the system will depend heavily on
the nature of the competitive environment or the entropy (van Gigch 1978) in the
environment. While most of the systems theory literature looks at the uncertainty and
variety in a system, a different model employs the levels as a factor explaining the ability
of a system to handle different levels of entropy. Trust can provide a degree of certainty
by both parties as to what the other party’s actions will be when confronted with

uncertainty.

2. 8.2 The IMP Interaction Model

Based on research in five European countries including France, Italy, Sweden,
West Germany, and Great Britain a purchasing interaction model was created
(Hakaansson 1982). The project emphasizes the importance of relational factors in buyer-
supplier relationships as they become longer and more complex. Four different sets of
variables are presented. The first set of variables describe the parties involved. The
second set looks at the specific elements involved with the interaction process. The third
set analyzes the environment in which the relationship takes place, and the final set looks
at the atmosphere affecting and affected by buyer-supplier interaction (Hakansson 1982).
While the IMP research describes the relationship of factors to performance, no specific

hypotheses were generated.

75

The IMP project characterizes long term buyer-supplier relations as a series of
episodes occurring between parties. These episodes may include the exchange of products
or services (information, financial or social). The project goes on to test a variety of
interactions, environments, atmospheres, and elements. The IMP model suggests that
trust may be a key factor in ongoing relations, and that the relations are influenced by
both individual and organizational characteristics (Smith 1993).

Ford (1980, 1982) takes the IMP model and actually proposes several testable
hypotheses on the subject of relationship development. Ford’s model like the IMP model
examines how industrial buyer/seller relationships develop over time. Ford identifies five
stages of the relationship: the pre-relationship stage, the early stage, the development
stage, the long-term stage, and the final stage. The key finding of Ford is that buyer
perceptions of seller technical and commercial skills are related to their perception of the
closeness between firms. Ford also suggests that key factors may vary across stages of a
relationship. The key implication from this research is again that trust is a key
relationship element, and further, that trust plays a different role depending on the stage

of the relationship.

2. 8.3 Exchange Bargaining/ Power

Dwyer, Schurr and Oh developed a framework of relationship development based
primarily on exchange bargaining and power theories. The authors propose five
progressive stages of relationship development, but unlike Ford recognize dissolution as a
stage requiring management. The four other stages are the awareness, exploration,

expansion, and commitment stages. During the expansion stage, five sub-processes

76

occur including: attraction , communication and bargaining, development and exercise of
power, norm development, and expectation development. These subprocesses deepen the
interdependence among firms and allow them to gauge interfirm compatibility and
integrity. Dwyer, Schurr and Oh’s framework includes the development of trust and
satisfaction at the exploration phase, creating the opportunity for greater risk taking and
motivation. The commitment stage results in the partners ignoring opportunities to
commit with other partners. Dwyer, Schurr and Oh’s framework is useful in that it
suggests research is needed to understand the development of trust in buyer seller

relations.

2. 8. 4 Channel Relationships

Anderson and Narus (1984) looked at the relationship between manufacturer and
distributor from the distributor’s perspective. They develop a model based on the social
exchange theory from Homans (1961) and Thibaut and Kelley (1959). The model states
that outcomes of interaction are evaluated against both expectations based on prior
experience. The model then examines the interaction of knowledge of similar
relationships against the average quality of outcomes available from the next exchange
partner. Anderson and Narus hypothesized that distributor satisfaction is a function of the
level of conflict and communication in the relationship as well as outcomes given
experience with similar prior relationships. While the model incurred measurement
problems, it was key in showing how social exchange concepts are important in

predicting behaviors in working relationships.

77

Anderson and Narus (1990) went on to develop a more comprehensive model of
distributor-manufacturer working partnerships based on exchange theory. The constructs
used included: relative dependence, communications, outcome given expectations, trust,
cooperation, functionality of conflict resolution, conflict, influence over the partner firm,
and influence by the partner firm. The constructs were hypothesized to directly or
indirectly affect both manufacturer and distributor satisfaction. The authors again ran
into some measurement problems despite using multiple informants from both
distributors and manufacturers. Anderson and Narus re-specified the model and
cooperation was shown to be an antecedent to trust. This result demonstrates that trust
needs to be earned (Holden 1990). Also, significant in this study is that trust was found
to be a key construct in Anderson and Narus’ model.

Anderson and Weitz (1989) empirically tested a model which also draws on the
social exchange literature. They posit that to develop effective long term relationships
the parties must expect the relationship to last. The authors found that continuity of a
channel relationship was determined by trust between parties, the age of the relationship,
stakes in the relationship, the imbalance of power, and the reputation of the parties.
Communication was found to be an important determinant of trust but not of continuity.
Other determinants of trust include the: age of the relationship, the reputations of the
parties, cultural similarity, support by partners, and goal congruence. Anderson and
Weitz looked at dyads and found the better relationships were based on trust and that
communication played a key role in building trusting relationships. The authors also
found a non-recursive relationship between trust and communication indicating constant

growth and change in a relationships. Anderson and Weitz analyzed each party’s

78

perceptions in detail. They were specifically interested in the roles of pledges (verbal
commitments) and how other parties react to commitments made and then how
perceptions change based on actions taken on pledges. They found that pledges made and
enactment on the pledges followed by reciprocation play a key role in relationship trust
and development.

Heide and John (1988) tested a model of partner dependence based upon
institutional economics and dependence theory. The premise of their model is that small
channel members protect themselves from exploitation by making investments in
customer relationships. The authors found that joint action (the degree of cross over into
each firrn’s boundaries) is a function of continuity expectations, supplier verification, and
supplier and buyer investments. Like Anderson and Narus, Heide and John found that if
the relationship is expected to continue, there is greater motivation for joint action
especially in uncertain and ambiguous tasks. The implication is that trust is significantly
impacted by a long term form by both firms.

Heide (1994) created the term interorganizational governance in analyzing
channel relationships. Heide created a typology consisting of three different governance
dimensions (relationship initiation, relationship maintenance, and relationship
termination) on the Y axis and three governance forms (market governance,
unilateral/hierarchical and bilateral) on the X axis. Heide goes on to empirically test the
governance structures.

2. 8. 5 Markets and Contracts
Ring and Van de Ven (1992) move beyond the discussion of TCE to a typology

that takes into account the recurrent nature of today’s interorganizational relations. The

79

Figure 2-3 below depicts Ring and Van de Ven’s hypotheses regarding the relationship of
trust, risk, and governance mechanisms in interorganizational relationships. Ring and

Van de Ven posit that as trust grows, a relationship can tolerate larger relative levels of

 

 

Figure 2-3
Risk of Deal
Low High
Low
MARKETS HIERARCHIES

Reliance on

Trust

RECURRENT CONTRACTS RELATIONAL CONTRACTS

High

 

 

 

 

risk. Without trust, companies can tolerate higher levels of risk but high cost monitoring
and contracting mechanisms must be used. The framework States if trust is present,
organizations should strive for recurrent contracts in low risk situations and relational
contracts in high risk situations. In the absence of trust (i.e. opportunism may exist -
Williamson), firms should use markets with low risk and hierarchies in high risk
relations. This dissertation, however, is concerned with Ring and Van de Ven’s high
trust/high risk situation where strategic alliances occur. Inherent in Ring and Van de

Ven’s argument is that trust and risk are separate and both play a key role in relationship

80

governance. The authors also do not argue for a sequential causal relationship between

risk and trust but rather a simultaneous development.

2.9 Summary of Alliance Literature

This sections has reviewed the major models and theories involving
interorganizational relationships. This work has focused primarily on buyer-supplier
relationships, between organizations. While the theories and models provide useful
insights into factors that may be important determinants of effective buyer-suppliers
relationships key antecedents to these factors have not been examined. Relationships
among these factors have also not been examined.

Existing literature recognizes the importance of interaction among parties in
buyer-supplier alliances but has typically focused on only one party’s perspective. Future
research needs to consider the dyad as a unit of analysis. Finally, while attempts have
been made to understand how close relationships form and develop over time, further
clarification and substantiation of this process is required.

Many of these gaps in the buyer-supplier alliance literature are addressed in this
dissertation. The conceptual model developed in this chapter incorporates many of the
key constructs identified by previous empirical research, as well as others, thus expanding
the nomological network of previous works. The research methodology detailed in
Chapter 3 is consistent with the examination of dyadic relationships as the unit of
analysis. The study does not address the gap of understanding how relationships develop
over time. However, the research will suggest management interventions which may

facilitate the development of trust.

81

2.10 Summary of the Entire Literature Review

The literature reviewed in this dissertation provides some consensus on the role of
trust in interorganizational relationships, but large gaps in the literature exist. While there
is literature directed at the antecedents of trust, there is little empirical testing of these
antecedents and in some areas such as asset specificity there are conflicting results. The
strategic alliance literature offers some models from the channels, purchasing, and
organizational theory literature, but the role of trust is not clear in these models, even
though each of these literature streams mentions that trust is a key element to successful
buyer-supplier or channel relations. The shift away from economic based
interorganizational transaction based theory towards long term relationship based theory
further emphasizes the need for clarification of the role of trust. This dissertation
attempts to bridge this gap through a detailed analysis of trust in strategic buyer-supplier
alliances. This dissertation will investigate the specific antecedents to trust development
which then impacts firm performance. The research methodology to accomplish this task

is introduced in the next chapter.

82

 

 

Chapter 3

Pre-Research and Research Design

As presented in chapter 1, the overarching goal of this research study is to reduce
the knowledge gap existing in the literature by investigating the antecedents to trust in
cooperative buyer-supplier relationships. Several of these factors have been selected
based on existing literature addressing interorganizational relationships, especially those
considering buyers and sellers. The scholarly contribution, in addition to narrowing the
existing knowledge gap concerning cooperative long-term buyer-supplier relationships in
the literature, is to synthesize the organizational behavior, marketing, and purchasing
theories regarding trust and the strategic management and policy theories of

interorganizational relationships.

3.1 Proposed Levels of Trust Model

An analysis of the trust and buyer-supplier relationship literature indicates that
there is a hierarchy of actions which can develop or dissolve trust between parties in
relationships. Categorization of the various antecedents to trust has been attempted by
several authors (Hosmer 1995, Mayer et a1. 1995). Authors have also developed models
looking at how organizational trust is formed (Mayer et al. 1995), and several authors

have tried to explain the different kinds of trust that exist (McAllister 1993, Barber 1986,

83

Lewicki & Bunker 1995, Sheppard & Tuchinsky 1995, Williamson 1991) primarily in an
interpersonal context. A study of the antecedents to different types of interorganizational
trust that exist is clearly missing in the literature.

There is agreement that different types of trust exist, but precisely what these
different types are is under debate. Empirical evidence supports the notion that a
goodwill or faith-based affect trust exists and a cognition-based (reliability and
competence) trust exists (McAllister 1993). Hosmer (1995) has brought the significant
issue of context to the forefront by stating that to develop trust it is imperative that both
sides in a relationship look internally as well as externally. The dominant stance in
research and literature on trust is to take the perspective of one side in a relationship and
identify how they trust or distrust the “other” Side. The sides are generally represented in
an interpersonal context (between individual persons). This context will be employed as
the unit of analysis in this dissertation. To effectively study trust, however, a more
objective approach must be taken which combines the perceptions of the both parties in
analyzing trust.

Mayer et al. (1995) present a model of interorganizational trust which begins to
look at both internal and external perceptions and actions. Their model, contains three
antecedents to the development to trust: benevolence, integrity, and ability. These three
areas make up the perceived trustworthiness of the other party.

Similarly This dissertation will investigate three antecedents to trust.

This trust then leads to a risk taking behavior on behalf of the primary party. The first
antecedent will take Mayer et al.’s trustworthiness concept and instead of applying it to a

partner apply it to the trustor. This dissertation will investigate whether a partner can

84

inspire trust in a relationship through their own example. Second, a variable specific to
cooperative buyer-seller relationships is the time orientation of partners. Time orientation
can be short-term, long-term or somewhere in between. The short-term orientation is best
described in the economics literature where partners will engage in relations until they
feel taken advantage of or a better opportunity is found. A similar short-term philosophy
suggests that cooperative buyer-supplier relationships are learning races, where parties
learn as much as they can as quickly as they can and move on. The long-term orientation
suggests that relations are based on mutual loyalty and respect. A partner experiencing
problems is not abandoned but rather offered assistance and every effort is made to
propagate and enrich the relationship. Third, asset specificity is a key antecedent in
interorganizational relationships in the creation of trust. While the role of asset
specificity is not clear from previous research, this dissertation will be the first to explore
varying degrees of asset specificity as well as the philosophy behind asset specificity.

The list of antecedents under study in this dissertation is not meant to be
exhaustive. Other attributes studied previously such as use of contracts, information
sharing, commitment, and communication are important in trust formation. The reason
the three antecedents of this study were chosen was because of their hypothesized critical
importance in the move today towards more cooperative buyer-supplier relationships.
These antecedents were also selected based on their discussion in many conceptual
articles concerning trust, but a corresponding lack of empirical support for their effect as

discussed in chapter 2.

85

3.2 Exploratory Research

In order to locate critical success factors in cooperative buyer-supplier
relationships a survey of firms participating in the Global Procurement and Supply Chain
Benchmarking Initiative was carried out in 1994. This survey was sent to over 200
companies as part of an on-going benchmarking initiative in supply chain management.
Of a sample of 205 firms, 77 responded to the questionnaire (a 38% response rate). Non-
response bias was tested and the reason most commonly given by non-responding firms
was their lack of development of a cooperative buyer-supplier relationship program.
Respondents were located in the United States, Canada, Mexico, Western Europe, and
Australia. Industries represented include industrial products, consumer goods, capital
goods, and service. Average annual sales in 1994 for respondents was $7 billion. The
primary respondent of this survey was the executive responsible for managing buyer-
supplier relationships at each firm. The results of this survey created a model of factors
leading to successful cooperative buyer-supplier relationships. Several of the factors
tested in this study included commitment, information sharing, interdependence, trust and
coordination. Of these four factors the most important attribute in creating successful
cooperative buyer-supplier relationships which emerged was the need to foster and
nurture a sense of trust with a partner (Monczka, Petersen, Handfield and Ragatz 1996).

While trust was found to significantly correlate with relationship success, the trust
construct was measured very broadly (similar to Mohr and Spekman’s 1994 work) as
follows:

0 the feeling of equity towards the deal with the partner

86

o a feeling that the relationship will be beneficial, and

o the degree of harmony in the relationship

These results all test an affective (feeling) dimension of trust which is important
but not comprehensive. The results of this GEBN study corroborate the results of Mohr
and Spekman, and provide further empirical support for the relationship between
affective-based trust and relationship performance. The GEBN study found that trust,
coordination, and interdependence explained 82% of the variance in perceptual measures
of success between least and most successful cooperative buyer-supplier relationships.

The results from this survey further support the findings from the literature review
in Chapter 2. Trust is a critical element of cooperative buyer-supplier relationships, yet
there exists no details regarding the actions that lead to the development of trust. The
next section describes the model used in this dissertation to analyze several key

antecedents to the development of trust in cooperative buyer-supplier relationships.

3.3 Trust Development Model

The GEBN study completed on cooperative buyer-supplier relationships along
with the literature review in Chapter 2 were used to create a model of trust development.
This framework suggests that three antecedents are critical to the development of trust.
Trust is then hypothesized to significantly impact relationship performance.

Figure 3-1 shows a complete model for trust development in cooperative buyer-
supplier relationships. The model in Figure 3-1 addresses many of the antecedents to

trust covered in chapter 2. This model is far too large to be tested and many of the factors

87

such as communication have already been tested and retested. While many technical

issues such as electronic communication, information quality, and coordination are

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 3-1
MAW
Pliloaofllyd’inn ——> Mad emu-um
Town-d Ali-noel
- Mirna] Barefits Reestablish Aw
- Think long term if Nader!
- Tnstwuthimss
- Joinly establish mat-mans and gods
- Establish raou'oe omnitmals
“solution
\ Practice/Alienate
of TM
Intervene Increment
Benin Sun in Resource - (‘alwltnvbased tnst
Conniumnt - Wedge-lined trust
- ltimtifialim—bmed tnst
. . IGcltnfl'the
mm“? ‘— W.)
Perform
. - Cormir Rmms
. WIN - (‘rmte open oamutialion
- Coa
- Delitcry
- Flexibility
- Reasscs Trust

important, they have already been studied. This study will address antecedents
recognized as important to trust development by practitioners and researchers but
untested empirically.

This study is primarily concerned with discerning new critical elements which
significantly impact trust development. The model to be tested is shown in Figure 1-2.
Trustworthiness and relationship planning horizon are new constructs in the study of

cooperative buyer-supplier relationships. The constructs of trustworthiness and

88

relationship planning horizon go beyond describing attributes of relationships such as
information sharing and actually elicit an important philosophy (a system of principles for
conduct) potentially underlying successful relationships. Trustworthiness and
relationship planning horizon are more closely related to principles than asset specificity.
Asset specificity, unlike relationship planning horizon and trustworthiness, is an attribute
(characteristic) of cooperative buyer-supplier relationships. An attribute is defined as a
characteristic of a person or thing while a principle is a doctrine used as a basis for
reasoning or to guide action or behavior (Oxford 1995). A firm would not operate buyer-
supplier relationships on the principle of asset specificity, since asset specificity is an
attribute of relationships. On the other hand, taking a long-term mutually shared view of
relationships is a principle which is a guide for the use of attributes such as asset
specificity. Asset specificity was chosen as a variable for study, however, because prior
studies of asset specificity have yielded mixed results possibly due to the
misunderstanding that asset specificity is not a principle.

Asset specificity is an attribute which can be employed for a variety of reasons.
For instance firms may co-locate personnel so that a new product development initiative
can be improved, or to improve the performance of their worst suppliers. Prior research
on asset specificity shows a distinct difference between the American version of asset
specificity and the Japanese version known as “dedicated” asset specificity. Empirical
research has shown that while Japanese firms and United States firms use co-location,
Japanese firms use it up to seven times more ofien throughout a year (Dyer 1994). This
dissertation will explore this difference to see if trust plays a role in the impact of asset

specificity in cooperative buyer-supplier relationships.

89

3.4 Definition of Factors in Figure 1-2
3. 4. 1 Trust

There are different types or dimensions of trust tested and used in research. An
early distinction in the trust literature is made between reliability-based trust (Deutsch
1958) and a good-will or faith-based trust (Granovetter 1985). Later, the issue of
vulnerability (Akerlof 1973) developed as part of trust. The evolution of trust has
incorporated additional dimensions including loyalty, competence, and character. Beyond
these different dimensions of trust, researchers suggest that not one type but multiple
types of trust exist.

The issue of context was first addressed by Hosmer (1995) who suggests that
there are also multiple contexts of trust. These contexts include interpersonal relations,
individual expectations, economic transactions, and social structures. The introduction of
context creates a matrix of different trust dimensions on one axis with different contexts
on another axis similar to the breakdown of trust described by Cummings and Bromily
(1995) in Table 2-3. The trust construct used in this study will build on the work of
Cummings and Bromily (1995) and Lewicki and Bunker (1995). These two studies were
chosen because of their combination of empirical rigor with strong support for multiple
dimensions of trust. Lewicki and Bunker’s trust dimensions were selected because their
study is one of the first studies to suggest that different types of trust build on one another
over time.

Beyond the work of McAllister (1993) and Cummings and Bromily (1995),

Lewicki and Bunker (1995) and (Tuchinsky & Sheppard (1995) discuss the progression

90

over time of trust through three stages including calculus, knowledge, and identification-
based trust which are defined below:

Calculus-based Trust Trust based on control or assuring that the other party
will do what they say. Trust is sustained through a clear deterrent (punishment)
and likely to occur if trust is violated. The threat of punishment is likely to be a
more significant motivator than promise of a reward. Calculus-based trust often
involves a high degree of monitoring to assess whether a party is being
opportunistic. Calculus-based trust is fragile and therefore any violation of trust
has potential to significantly alter or even end a relationship.

Knowledge-based Trust Knowledge-based trust is grounded in the
predictability - knowing the other party sufficiently well so that the other’s
behavior is anticipatable. Knowledge-based trust relies on information rather than
deterrence as a motivator. Knowledge-based trust develops over time as the
parties develop confidence in the trustworthiness of the other party. In
knowledge-based trust regular communication and courtship are key processes.
Identification-based Trust Identification-based trust is based on identification
with the other party’s desires and intentions. Trust exists because the parties
effectively understand and appreciate the other’s wants. A mutual understanding
develops as each side clearly understands the motivations and problems of the
other side. Each party comes to understand what they must do to sustain the
other’s trust. Identification-based trust is also characterized as having: collective
identity, colocation of personnel, creating joint products or goals, and commonly
shared values.

Figure 3-2 depicts Lewicki and Bunker’s model of how these three types of trust
develop and build on each other over time. The sample of this study is high performing
cooperative buyer-supplier relationships within firms. Prior work on the GEBN reveals
that identification-based trust appears infrequently in relationships whereas, calculus-
based trust appears more frequently because many firms are in the beginning stages of
implementing cooperative buyer-supplier relationships. Knowledge-based trust is the
most frequently seen type of trust seen in relationships analyzed by the GEBN.

Beyond different types of trust there is the issue whether trust is an individual or

organizational construct. Rather than investigate if trust is an organizational or individual

91

Figure 3-2 Lewicki and Bunker’s Model

 

 

 

 

 

 

 

 

 

Identification

Based

Trust
Knowledge :
Trust 5 i
Calculus I i 1
Trust 5 g

F
TIME

construct this dissertation will seek to prove that trust is whether measured individually or
at the organization level will load on the same construct. Prior theory by Ring and Van
de Ven (1992) suggest that buyer—supplier relationships outlast the tenure of the persons
involved which would suggest that organizational trust exists. Does this organizational
trust differ from the trust between individuals? This dissertation will seek to answer this
question. Recent work by Zaheer et al. (1996) using a prior instrument by Rempel and
Holmes (1986) indicates that individual and organizational trust are different constructs.
The key difference between Zaheer et al.’s individual and organization level trust

constructs differ only in their reference to the other party as a contact person or supplier.

92

Trust Antecedents
3. 4. 2 Power and Dependence

Emerson (1962) was the first to note that power and dependence share a reciprocal
relationship. A look at channel and sourcing literature over the past 15 years shows a
movement from analysis of power to dependence. Beyond the area of dependence this
dissertation will look at the interdependence between the buyer and supplier. More
interdependence is hypothesized to increase trust. Power was first described by French
and Raven as existing in five difference forms (referent, expert, coercive, legitimate and
reward). Two of these types of power (coercive, reward and legitimate and expert power

are hypothesized to correlate with knowledge-based and identification-based trust.

3. 4.3 Relationship Time Horizon

Ring (1996) describes two types of trust which depict two different underlying
philosophies toward interorganizational relationships. Fragile trust (short-term
perspective) is based on the economist’s perspective which sees trust as an avoidance of
opportunism (Williamson 1993). Fragile trust as its name implies is easily broken. The
other type of trust is resilient trust which is based on loyalty and openness. The economic
perspective of trust takes a short-term perspective toward relationships and is based on
transactions cost analysis (Williamson 1975) and the Prisoner’s Dilemma. It is no
surprise then that at the first sign of trouble or misunderstanding, firms with a short-term

planning horizon tend to abandon relationships or adopt the “tit for tat” strategy (Gulati &

Nohria 1992).

93

Resilient trust takes a long-term outlook at relationships. The long-term
perspective is best articulated by Gulati and Nohrias’ (1992) concept of a mutually shared
relationship where the relationship is built on faith, respect, and trust. A key to the long-
term horizon is the ability to forego short-term benefits for more important and lasting
long-term gains through stronger trust. This construct will capture a firm’s relationship
planning horizon and determine if a firm has a predominately short-term or long-term

relationship planning horizon.

3. 4.4 Asset Specficity

Significant research (Nishiguchi 1994, Dyer 1994, Kamath & Liker 1994,
Handfield Bechtel & Stimpert 1996, Monczka, Petersen, Handfield & Ragatz 1996) and
theory development (Williamson 1975) has been done in the area of asset specificity in
interorganizational relationships. Despite this work the relationship between asset
specificity and trust is unclear. Several authors have found a positive correlation between
asset specificity and trust (Kamath & Liker 1994, Lorenz 1988, Dyer 1994) while other
authors have been unable to find a positive relationship (Campbell 1992, Handfield,
Bechtel, & Stimpert 1995). Some research has even found a negative effect between
asset specificity and trust (Zaheer & Venkatraman 1995, Joshi 1993). Asset specificity is
similar to trust in that there are different types of asset specificity and varying degrees of
each type. Most previous research on asset specificity assumes one gp_e of asset
specificity (usually equipment or co-location) and one Lexi] of asset specificity. Recent
work by Kamath and Liker (1994), Dyer (1994), and Nishiguchi (1994) offer a significant

finding in the asset specificity area. Although Japanese and U. S. auto firms use co-

94

location Japan uses it up to seven times more often. Consequently , while US. firms and
Japanese firms report using asset specific investment there is likely a key difference in
degree of usage. This dissertation will investigate the possible role of trust in success
with the use of dedicated asset specificity. A second issue of interest is the purpose
behind the use of asset specificity. Asset Specificity can be used for a variety of reasons
from collateral to a genuine contribution to genuine learning and problem solving. While
dedicated asset specificity has been shown to impact performance the question of whether
trust also plays a role in that relationship has not been investigated. Must there be
knowledge-based trust for dedicated asset Specificity to occur? This dissertation will seek

to answer this question.

3. 4. 5 Risk/Environmental Uncertainty

Risk has been incorporated into several explanatory models (Mayer et al. 1995;
Zaheer and Venkataraman) of trust and theory-based discussions of trust (F airholm 1994;
Lorenz 1988) but actual empirical investigation of risk is sparse. The similar concept of
uncertainty has been tested (Zaheer and Venkataraman 1994) but risk uncertainty was not
found to create more organization structure or higher levels of trust.

There are two predominant theories on the relationship between risk and trust.
The first states that trust works as a counter to risk (Lorenz 1988; Williamson; 1985) and
the response of most firms to risk is use of contracts or punitive arrangements. In other
words where there is low risk there is likely high trust. The other theory states that to
create trust risk must be present (Lewis 1990; Ring and Van de Ven 1992; Luhmann

1988). Here we would find high levels of both risk and trust in relationships. This

95

dissertation will test cooperative buyer-supplier relationships where high levels of trust
may exist. Therefore, we expect high levels of trust and risk to be simultaneously

present.

3. 4. 6 Trust and Performance

The final factor to be examined is cooperative buyer-supplier relationship
performance. Anderson and Narus (1990) used satisfaction with various parts of a
working relationship among partners as a proxy for partnership success which will also be
used in this dissertation. Satisfaction or dissatisfaction will look at the adequacy of
rewards received through the relationship (e.g. Frazier 1983). Performance is best
measured by items tied to financial, operational and effectiveness dimensions
(V enkatrarnan and Rarnanujam 1986). All three of these dimensions will be part of the

performance measurement in this study.

3.5 Hypotheses

The model of trust development in cooperative buyer-supplier relationships in
Figure 1-2 was developed from the pre-dissertation research and literature. This study
will address the gap in the literature between trust and performance and identify specific
antecedents to trust development.

Pl. Three types of trust exist which are calculus-based, knowledge-based,
and identification-based trust which have a hierarchical relationship.

Figure 3-2 from Lewicki and Bunker shows that over time trust moves from
calculus to knowledge-based over time and eventually to identification-based trust.

Hypothesis one states that knowledge-based trust cannot exist unless calculus-based trust

96

exists and identification-based trust cannot exist unless both calculus and knowledge-
based trust exists.

P2. Organization trust and individual trust are not unique.
Trust has been tested in both individual and organizational units of analysis but there has
yet to be an analysis of whether these units are different. This dissertation will test to see
if these two units of analysis are unique.

HlA. Use of coercive power will be positively associated with calculus-based
trust.

HlB. Use of non-coercive power will be positively associated with knowledge-
based trust.

HlC. Use of non-coercive power will be positively associated with
Identification-based trust.

H2. Dependence between buyer and supplier will be positively associated
with the level of identification-based trust in a supplier.

Interdependence is hypothesized to correlate with identification-based trust because the
firms need each other and therefore have motivation to invest in the relationship. Mohr
and Spekman (1994) found a positive relationship between interdependence and buyer
supplier relationship performance. This dissertation seeks to prove that interdependence
can effect performance but this effect happens through increased trust.

H3A. A long-term buyer-supplier relationship planning horizon will be
positively associated with the level of knowledge-based trust in a supplier.

H3B. A long-term buyer-supplier relationship planning horizon will be
positively associated with the level of identification-based trust in a supplier.

While traditional economic theory had been used to describe interoganizational
relationships it contains three assumptions which are not reflected in cooperative buyer-

supplier relationship practice including the following:

97

o Assumption of no advantages to repeat interactions

0 Assumption of Prisoner’s Dilemma payoff scheme

0 Assumption of Opportunism in all participants
New theory by Gulati & Nohria (1992) suggest that interorganizational relations be
governed by a mutually shared relationship philosophy guided by principles of trust,
generosity, and faith. Gulati and Nohria’s mutually shared paradigm takes a long-term
approach to relations but is yet to be empirically tested. On the other side, firms adopting
the economic short-term relationship planning horizon model will see lower levels of
trust because there is no willingness to forego short-term benefits for long-run gains
because all partners are assumed to be opportunistic. Knowledge and identification-based
trust are hypothesized to have a relationship with planning horizon and not calculus-based
trust because calculus-based trust is likely due to either a short-term economic
perspective or a cooperative relationship that is in its beginning stages.

H4. Asset specificity behavior focusing on information sharing and synergy
will be positively associated with identification-based trust in a supplier.

As discussed earlier, asset specificity can be used for a variety of reasons. Asset
specificity can be a “hostage,” a penalty, or asset specificity can be a genuine
commitment to solving problems in a relationship. Firms using asset specificity as a
hostage (rather than using genuine trust) will not create trust or reap relationship
performance benefits of firms using asset specificity to solve problems (exchange
information and create synergies). Therefore, use of asset specificity to address problems
will produce better results than asset specificity used as a hostage when there is no trust.

Knowledge and identification-based trust are hypothesized to have a relationship with

98

asset specificity and not calculus-based trust because calculus-based trust is likely due to
either a short-term economic perspective or a cooperative relationship that is in its
beginning stages.

H5A. The overall level of asset specificity will be positively associated with
the level of knowledge-based trust in a supplier.

HSB. The overall level of asset specificity will be positively associated with
the level of identification-based trust in a supplier.

As mentioned previously the measurement of asset specificity has not recognized the
existence of varying levels of trust and varying levels of asset specificity. Varying levels
of trust (calculus, knowledge and identification-base) have already been discussed, but
the idea of different levels of asset specificity is relatively new. Recent work by Dyer
(1994), Kamath & Liker (1994), and Nishiguchi (1994) suggest that Japanese firms make
significant investment in “dedicated” asset specificity. In dedicated asset specificity,
suppliers send engineers to work at a customer’s site, locate plants near the customer, or
invest in customized physical assets. The US. has failed to appreciate the benefits of
dedicated asset specificity (Dyer 1994; Nishiguchi 1994). Firms in the United States
report the exchange of personnel and other assets similar to Japanese firms. The
difference however, between typical U.S. asset specificity and Japanese “dedicated” asset
specificity is degree. Japanese automakers average 7,000 face days with suppliers per
year while the average US. automaker averages only 1,000 face days (Dyer 1994). So
there is a key difference between traditional asset specificity and the Japanese version of
asset specificity. This dissertation will seek to make this distinction.

H6. Uncertainty will be positively associated with the level of identification-
based trust.

99

Uncertainty or risk will correlate with the levels of knowledge-based and identification-
based trust. Since respondents will be asked to answer questions about a cooperative
buyer-supplier relationships we expect high levels of trust and risk to exist. Firms are
assumed to create cooperative relationships with suppliers in situations where cooperative
relationships will be beneficial (high risk). Situations where there is low risk don’t
warrant a cooperative relationship and will require less trust.

H7A. Alliance performance will be positively associated with the level of
knowledge-based trust in a supplier.

H7B. Alliance performance will be positively associated with the level of
identification-based trust in a supplier.

Knowledge and identification-based trust will correlate with relationship performance.
Firms with high amounts of openness and concern will have higher relationship
performance. Past empirical research has shown a relationship between trust and
performance (Mohr & Spekman 1994, Monczka, Petersen, Handfield, and Ragatz 1996),

but in both cases only a very general type of affect-based trust was tested.

3.6 Details of Measures to be Used

This study analyzes the gap in research linking the concept of trust development
and performance in cooperative buyer-supplier relationships. While several studies and
theorists have linked higher levels of trust to better relationship performance, trust
remains a fuzzy concept and clear actions or antecedents to trust are untested. Some of
the dimensions and antecedents of trust in this dissertation have been previously
examined in the literature including calculus-based trust, knowledge-based trust, and

identification-based trust along with asset specificity, but measures of these concepts are

100

not well developed. The other constructs in this dissertation which are relationship
planning horizon, trustworthiness, and components of identification-based trust and
concem-based trust are new constructs or elements with little or no empirical testing.
One of the purposes of this dissertation is to attempt to establish valid measures for these
untested constructs. The next sections define measures for calculative, knowledge, and
identification-based trust along with asset specificity, trustworthiness, relationship
planning horizon, and performance. A copy of the actual survey to be used is in

Appendix A.

3. 6. 1 Trust

There are a number of different ways of measuring trust. (McAllister 1993; Rotter
1967; Lazelere & Huston 1980). Studies on trust can be described as falling on a
continuum. On one end of the continuum are studies which ask one party to assess their
feeling of trust with another party. These studies may even break trust down into
different types of trust (McAllister 1993; Barber 1986: Lewicki & Bunker 1995;
Sheppard & Tuchinsky 1995; Williamson 1991). These studies take a unidimensional
perspective where one party looks only at the amount of trust they have with another
party. At the other end of the continuum are studies which take a dyadic or inductive
approach to the study of trust. These studies look at trust in how it exists in a
relationship. The perception of trust by both parties is important in these studies
(Lazelere & Huston 1980; Gabarro 1978). In these studies, parties begin the process of
self realization and begin to look at their own trustworthiness as well as the trust involved

with another party.

101

3. 6.2 Types of Trust

This study looks at five hypotheses centering around the concepts of calculus,
knowledge and identification-based trust. Below are the sources of questions used for the

calculus-based trust construct.

* indicates questions used in questionnaire to measure calculus-based
trust

** indicates questions used in questionnaire to measure knowledge-based
trust

*** indicates questions used in questionnaire to measure identification-
based trust

**** Item of individual trust
Parkhe (1993) developed several measures of opportunism in cooperative relationships

which attained an alpha of .828.

Avoid Opportunism

Questionnaire Item

 

Our partner always provided us a completely truthful picture of their business

 

*Complete honesty does not pay when dealing with our partner

 

*Sometimes my partner alters the facts slightly in order to get what they need

 

My partner carries out their duties even if we do not check up on them

 

*My partner has sometimes promised to do things without actually doing them later

 

*They seem to feel that it is OK to do anything within their means that will help further
their firm’s interest

 

 

 

The two key areas of knowledge-based trust are competence (the ability and resources to

carry out a task) and reliability of supplier’s performance over a period of time.
Competence is essentially the perception of the ability of party B to meet their

commitments set forth and promised to party A. The issues of trust and competence have

been addressed by several authors (Gabarro 1978; Butler 1991; Luhmann 1980; Ghoshal

102

& Bartlett 1995). Gabarro breaks competence-based trust into three key areas. First,
“specific competence” which is trust in the other’s specific function or area. Second,
“interpersonal competence” is the ability of a person to work with people or people skills.
Finally, “business sense” addresses a person’s experience, wisdom, and common sense.

Reliability can be broken down into requisite elements. Reliability is dependent
on prior contact with a party or experience. Several authors (Gabarro 1978; Lorenz 1988;
F airholm 1994) mention time as a key element to reliability-based trust. A second related
major element is repeated interaction among the firms (Gabarro 1978; Good 1988; Gulati
1995; Ring & Van de Ven 1992). Repeated interaction and time (Lorenz 1988) then
leads to levels of confidence, consistency (F airholm 1988 ), and finally trust (Zucker
1995; Lewis 1990). Reliability can then lead to predictability which is confidence in
firture actions (Fairholm 1988). Reliability must be based on integrity or honesty (Fisher
and Brown 1989) to be effective. Reliability based on coercion or stress eventually
creates a suboptimal relationship or total breakdown in the relationship.

Cummings and Bromily describe consistency more as a comfort one party has for
the ability of the other party to complete actions without worry over their performance.
Cummings and Bromily found high reliability on their OTI scale for their consistency
based measures.

For the purpose of this study Cummings and Bromily short version of their OTI
scale will be used. The short OTI version was chosen because the long version contains
over 100 items which is not suitable for the sample of this study. Second, Cummings and
Bromily contend that the short form predictability is just as strong as the long form

predictability. The short form eliminates the intended behaviors in terms of questions

103

which had the lowest item to factor correlations in comparison to the cognitive and

affective elements. Second, the authors chose the affective and cognitive elements with

 

** I think the people in our partner firm tell the truth in negotiations

 

I think that our partner meets its negotiated obligations to our department

 

** In my opinion this partner is reliable

 

I think that the people in our partner firm succeed by stepping on other people

 

** I feel that our partner firm tries to get the upper hand

 

** I think that our partner takes advantage of our problems

 

** I feel that our partner negotiates honestly

 

WNQWPP’P?‘

* * I feel that our partner will keep their word

 

9. ** I think that our partner does not mislead us

 

10. ** I feel that our partner does not try to get out of commitments

 

11. ** I feel that our partner negotiates joint expectations fairly

 

 

12. ** I feel that our jartner takes advantage of people who are vulnerable

 

 

the highest item to factor correlations and use .70 as their minimum cut-off point. The
twelve questions that make up the short form produced a better model fit than the long
form and reliabilities all above .9. In addition the predictability of the short form was
found to be almost identical to the long form. Below is Bromily & Cummings 12
question short OTI form.

Of the twelve questions above, six of the questions tap calculus-based trust
including questions 4, 5, 6, 9 and 12. These questions analyze whether the other party is
opportunistic or not but does not analyze the other parties’ honesty or integrity. The other
questions: 1, 2, 3, 7, 8, 10, and 11 are knowledge-based. Questions 1 and 7 top into the
character of the partner (whether or not they are honest). Questions 2, 3, 8, and 10 tap
into a partner’s reliability and integrity and question 11 addresses fairness. Integrity
addresses whether the partner does what they say they will (keep commitments) and

reliability addresses consistency in performance over time. These seven items look at

104

characteristics that can only be attained through knowledge of the other party. Whether a
partner meets commitments, is honest, or fair are impressions that can only be attained by
having previous experience or knowledge of a partner.

The following seven questions come from an instrument developed by Sheppard
and Tuchinsky (1995) representing trust in buyer-supplier relationships. The construct

had 20 items and achieved an alpha of .86.

 

**** We know this supplier will consider our concerns when making a decision.

 

***This supplier views the world the same way as us.

 

** We have frequent fact to face contact this supplier.

 

 

. **** The confront issues effectively with this supplier.

 

. I understand well the bases on which this supplier is rewarded and compensated.

 

 

1

2

3.

4. ** We have frequent phone conversations with this supplier
5

6

7

. Our supplier understands well the bases on which we are rewarded and compensated.

 

 

The following questions come from McAllister’s construct of affect-based trust

and will be used to create an identification-based trust construct for this dissertation.

 

 

 

 

 

 

 

 

Questionnaire Item Factor Loading
This person approaches his/her job with professionalism and .846
dedication.

"Given this person’s track record, I see no reason to doubt his/her .825
competence and preparation for the job.

I can relay on this person not to make my job more difficult by .736
careless work.

Most people even those who are not close friends of this individual .725
trust and respect him/her as a coworker.

****Other work associates of mine who must interact with this .659
individual consider him/her to be trustworthy.

If people knew more about this individual and his/her background, .643
they would be more concerned and monitor his/her performance more

. closely Qeversed).

 

 

105

The work of Cummings and Bromily also provides factor analysis results and reliabilities

in the discussion of various types of trust.

Affect-Based Trust

 

git—estionnaire Item Factor Loadin
****We have a sharing relationship. We can both freely share our .812

ideas feelings and hopes.
****I can talk freely to this individual about difficulties I am having .722
at work and know that he/she will want to listen.
****If I shared my problems with this person, I know he/she would .720
respond constructively and caringly.
We would both feel a sense of loss if one of us was transferred, and .712
we could no longer work together.
****I would have to say that we have both made considerable .666
emotional investments in our working relationship.

 

 

 

 

 

 

 

 

Two questions were derived from the recent writings of Ring (1995) addressing

integrity and loyalty as components of resilient trust.

 

I 1. ****Our partner acts with integrity. I
I2. *** Our partner is loyal to our alliance. I

 

 

Finally, the following 9 questions were developed by the writer of this dissertation based
on the literature review from chapter 2. Questions 7, 8, and 9 are loosely developed from

an instrument used Kumar (1996) to assess trust in manufacturer-retailer relationships.

 

** We jointly establish goals for our relationship with this supplier.

**** We interact with this supplier in informal as well as formal settings.
**** We regularly communicate with this supplier.

**** This supplier regularly communicates with us.

** This sirpplier is predictable.

** We will take quick action if this supplier does not perform satisfactorily.
** We regularly share information with this supplier.

** We occasionally use fear to motivate this supplier.

** We pass on new information that may be useful to this supplier.

 

 

 

 

 

 

 

 

 

 

 

106

3.7 Antecedents to trust

Not all antecedents to trust development in buyer-supplier relationships will be
addressed by this study. From a purely theoretical consideration, a comprehensive
examination of the antecedents to trust between industrial buyers and suppliers in
cooperative buyer-supplier relationships would have to include all factors which impact
trust. Such a descriptive model would be very difficult to test statistically. The
prohibitively large sample size required to test such a model would likely be unduly
complex, unwieldy, and difficult to explain in a rational and parsimonious manner.
Third, since many of these antecedents might display significant overlap due to
definitional and descriptional incongruities, significant auto-correlation would likely
exist between similar factors which might confound the trust, underlying relationships

between the variables involved.

3. 7.1 Power
Coercive power was first coined by Hunt and Nevin (1974) as power based on
anticipation of the part of possible punishment if there is not compliance. The three

questions below were developed from Kumar (1996).

 

I We will take quick action if this supplier does not perform satisfactorily.
I We occasionally threaten this supplier to motivate them
[We occasionally use fear to motivate this supplier.

 

 

 

 

Non-coercive power is described by Hunt and Nevin as encompassing legitimate,

referent, reward and expert power (French and Raven 1959). This dissertation will

107

question referent power (questions 1-4), and legitimate and exptert power in questions 5-

8.

 

Our supplier feels that we identify strongly with their needs.

Our supplier feels that we have a strong sense of oneness with them.
Our supplier feels that we have become closely associated with them.
We have few problems influencing this key supplier

This key supplier is highly receptive to our ideas

This key supplier respects our industry knowledge

Our key syplier recognizes us as an expert in the industry

This supplier is not receptive to our suggestions

 

 

 

 

 

 

 

 

 

 

3. 7.2 Dependence
Ganesan (1996) and Lusch and Brown (1994) have measured dependence from

both sides of a relationship leading to interdependence. Below are two constructs used by

Lusch and Brown (1994).

 

I 1. We are dependent on this key supplier.

I 2. Our major supplier would be difficult to replace.
I 3. Our major supplier would be costly to lose.
Alpha = .872

 

 

I 1. Our key supplier is dependent on us.

I 2. Our major supplier would find it difficult to replace us.
I3. Our major supplier would find it costly to lose us.
Alpha = .878

 

—_‘l

3. 7.3 Relationship Time Horizon

3.7.3.1 Length of Relationships The construct of Relationship Time Horizon is a
new construct that has been discussed by several authors (Gulati & Nohria 1992) but
remains untested. There are three components which make up this construct. First,

several questions address the key issue of time. Empirical evidence shows a clear

108

relationship between length of relations and time (Gulati 1995). This dissertation asks

two questions about time based on the work of Parkhe in 1993 where the alpha = .64.

 

1. What was the intended duration of this alliance? (l= short-term 1-3 years 3= long-
term over 5 years)

 

2. Perceived likelihood that the likelihood that the alliance will actually last for the
intended duration (1 =low, 5=high)

 

 

 

3.7.3.2 Learning Races A second but related issue to time orientation is the
concept of cooperative relationships as learning races as discussed by Prahalad & Hamel
(1989) and Khanna, Gulati, and Nohria (1994). Firms in learning races strive to learn as
much as they can as fast as they can while being very careful to set limits on information
they divulge. Khanna et al... (1994) look at learning races from a payoff perspective and
found that the ratio of private benefits to common benefits is what is critical in learning

races.

 

1. Alliances are a competition between partners to see who can learn more from the
other.

Once, I have gotten what we need from an alliance we move on.

My primary goal in an alliance is to learn as much from my partner as possible.

I have several controls in place to limit the information a partner can get from us.
Partners will take non alliance information from me if I do not monitor them.

 

 

 

 

 

 

9‘95”!”

 

3.7.3.3 Mutuallv shared paradigm (continued commitment even in adversity)

 

Gulati & Nohria (1992:2) discuss an idea of mutually shared relationships where a
partner’s best strategy may be “Continued Commitment to the relationship even in the
face of perceived cheating by another partner, in order to repair the relationship and
restore mutual assurance.” Trust, generosity, feedback, and repair serve as guiding

principles in mutual assurance games as opposed to retaliation and restoration, the

109

guiding principles in prisoner’s dilemma situations.” Relationship Time Horizon is
computed from three elements above. Relationship Time Horizon forms a short term
economic paradigm if the typical expectations and actual time a firm is in a cooperative
relationship is short, the firm is primarily interested in a learning race where they can
extract as much information as possible while giving up as little information as possible
and the firm sees the relationship as having winners and losers and is unwilling to expend

any effort into fixing a troubled relationship.

 

p—A

I reap unexpected benefits from my alliances.

2. If my partner is gaining more from the alliance than I am I always take steps to end
or alter the agreement.

3. If a partner takes advantage of me, we immediater terminate the alliance.

4. In some of our alliances my partner’s benefit more than we do but it is still in my
best interest to maintain the alliance.

5. I provide significant training or assistance to suppliers whose performance sflps.

 

 

 

 

 

 

 

The mutually shared paradigm that Gulati and Nohria (1992) and Ring’s concept of
resilient trust are based on a firm’s ability to stick with a partner during the tough times.
The questions above were created to elicit respondent’s attitudes toward working through
problems with a partner as opposed to abandoning them.

Finally, Krause (1995) based on the work of Anderson and Weitz (1992) created a
construct for supplier commitment. Anderson and Weitz (1992) define supplier
commitment is the willingness of a supplier to make short-term sacrifices in expectation
of long-term benefits. The only way suppliers will be willing to make commitments is if
they have prior knowledge (knowledge-based trust) in the predictability of the other party.

The four questions below achieved an alpha of .854.

110

 

The Supplier has a strong sense of loyalty to us.

 

This supplier is quite willing to make a long-term investment in helping us.

 

This supplier sees our relationship as a long-term alliance.

 

PP’NT‘

This supplier is willing to dedicate whatever peOple and resources it takes to make us
a satisfied customer.

 

 

 

3. 7. 4 Asset Specificity
Knowledge, human and dedicated asset specificity are hypothesized to impact
trust. Handfield, Bechtel, and Stimpert (1996) developed the following questions on

human and dedicated asset specificity which will be used.

Human Asset Specificity (Alpha of .76)

1. We share our weekly, daily, or hourly production schedules with this supplier.

2. Non-sales personnel from this supplier have visited our plant to see how their
product is used in our process.

3. The supplier works closely with our engineering and design area for new product
development.

4. We Share a great deal of sensitive information with this supplier.

5. Our manufacturing personnel regularly visit this supplier’s facility.

 

 

 

 

 

 

 

 

Dedicated Asset Specificity (Alpha of .78)

1. This supplier has dedicated and reserved equipment and reserved capacity
specifically to maintain our purchasing relationship.

2. This supplier has dedicated personnel specifically to maintain our purchasing
relationship.

3. This supplier has purchased specialized equipment to meet our needs for this key-
input material.

 

 

 

 

 

 

Dyer (1994) also developed the following questions on asset specificity which
will be used in a modified form. The reason for using asset specificity is just as
important as the use of asset specificity itself. U.S. automakers use co-location as a
penalty to suppliers that are having quality or delivery problems while Japanese
manufacturers use co-location primarily as a synergistic problem solving and learning

forum. Japanese firms also use asset specific investment substantially more than US.

111

firms which leads to better performance but does trust play a role in this relationship?

This dissertation will explore the role of trust in asset specificity.

 

1. Total annual man days of face to face contact between supplier salesmen and
engineers and our purchasing personnel and engineers.

 

2. The average number of co-located or guest engineers.

 

3. The extent to which the partner trusts us with confidential (i.e.) technical
information (Likert scale).

 

4. The extent to which the partner shares detailed information on their production
costs (Likert scale).

 

5. The extent to which the I share information with the partner to assist them in
cost reduction (Likert scale).

 

6. The extent to which I share information with the partner to assist them in quality
improvement (reduce defects) (Likert scale).

 

7. The extent to which I share information with the partner to assist them in
improving their delivery and inventory management systems (Likert Scale).

 

 

 

The pre-dissertation research also surveyed firms on asset specificity in terms of
information. Two questions will be used from this study.

GEBN Questions

 

1. In this relationship, it is expected that any information which might help the
other party will be provided.

 

2. The parties are expected to keep each other informed about events or changes
that may affect the other party.

 

 

 

The pre-dissertation research also surveyed firms on asset specificity in terms of
information sharing. Two questions will be used from this study

GEBN Questions

 

1. I require our partner to contribute significant resources so they have less
incentive to leave the relationship

 

2. The primary reason personnel from our partner visit our facility is to learn more
about our products and processes

 

. 3. I do not require significant asset contributions by our partner because we want
the ability to change partners quickly if need be

 

4. I only require our partner to put personnel in our facilities when they have a
significant performance problem

 

 

 

112

 

3. 7.5 Uncertainfl

Risk is hypothesized to play a role in the development of trust in
interorganizational relationships (Luhmann 1988; Ring & Van de Ven 1992). Risk can
fall into several different areas including product risk and pricing risk (Zaheer &

Venkatraman 1994). Alpha for this construct was marginal at .61.

 

I 1. Pricing changes occur often in the area that this cooperative supplier works in J
I 2. New products are introduced often in the industry this cooperative supplier is in. I

 

In addition to these types of risk three new questions will be used addressing risk specific
to the part or service provided by the cooperative supplier in question. These questions

will address the areas technology, supplier switching and overall degree of risk.

 

I 1. Switching from our cooperative supplier to another supplier would be very costly. I
I 2. Technological change is rapid in the industry this cooperative supplier works in I
I 3. We have taken a high degree of risk with this cooperative supplier. I

 

3.8 Performance

Good performance measures will analyze financial, operational, and effectiveness
dimensions as recommended by Venkatraman and Ramanujam (1986). Parkhe (1993)
identified two factors that were important to relationship performance. Fulfillment of

strategic need and indirect performance indicators are reflected below.

 

GEBN questions
' 1. In this strategic supplier alliance/partnership relationship, the parties work together
to solve problems.

 

2. Do you plan on continuing this relationship.

 

3. Please indicate the overall degree of results satisfaction with your most/least
successfitl strategic supplier alliance/partnership.

 

 

 

113

Please indicate the percentage improvements in the following areas attributable to your
partnership.

 

Cycle-time Reduction (from order placement to receipt inclusive) _%
Shared Price Reduction (cost savings due to improvement) _%
Quality Improvement (reduction in PPM, warranties) %
On-time Delivery (ability to deliver within buyer-specified delivery window) _%
Access to new technologies _%
New product development time (from concept to production) _%

1-4 from Krause 1995
5-6 from GEBN 1996
Handfield, Bechtel, and Stimpert 1996 (Alpha of .83)

1. There is a short lead time between the time that the order is placed
with this supplier and when it is received.

 

 

 

The following factors affected the buyer’s decision to use this supplier (strongly

agree/strongly disagree):

 

 

 

I 1. Short lead-times I
L2. Outstanding on-time delivery record I
I 3. Ability to modify their product to meet our requirements I

 

Finally the respondents will be asked to comment as to whether their firm’s ROA ROI
and profitability have increased since the relationship has started or decreased. These
responses will then be checked against Dun and Bradstreet statistics to corroborate the

respondent’s data.

114

3.9 Proposed Research Design

The target sample for this research will be 1,500 National Association of
Purchasing Management (N APM) members. 1,500 members was chosen based on
Cohen’s power formula of sample size required to obtain a specified power, effect size
and confidence level. Based on Cohen’s (1992) power analysis table a minimum sample
size of 84 must be obtained to obtain an effect size of .30 at the .05 confidence level.
1,250 members were chosen because an estimated 20 - 30% response rate will yield a
sample of 200 to 300 responses which is adequate for running the statistical testing for
this dissertation. Titles of the target sample will be Purchasing Manager, Director of
Purchasing, Vice President of Purchasing, and Vice President of Materials Management.
These high level purchasing executives comprise the NAPM Title 1 members. Title 1
members were chosen as the respondent group because they were thought to be informed
about the key issues in the survey instrument.

Several different methodologies could be used in this study including: surveys,
case studies, or interviews. Case studies are often difficult to generalize (Kerlinger
1986). Surveys are useful because standardized measures (that are a necessity for making
comparisons (Fowler and Mangione 1990) can be used across a broad population in order
to make generalizable conclusions. However, if surveys are not conducted properly, they
can miss the depths of many relationships (Brewer & Hunter 1989). For the purpose of
this dissertation, a survey will be used. The reason for a survey is that many of the
problems associated with the use of a survey instrument including: response rate, correct
company contact, sample size - can be avoided by using a sample of Title 1 NAPM

members. The level of analysis for this study will be the interpersonal level. Buyer-

115

supplier relationships are between organizations but organizational relationships are
comprised of a series of interpersonal relationships at the heart of a cooperative buyer-

supplier relationship.

3. 9.1 Data Collection

Data collection will proceed in two stages. Stage one is a preliminary
investigation into the role of trust within cooperative buyer-supplier relationships. This
preliminary stage will consist of constructing a survey instrument to effectively gather
data on the relevant variables. Stage two is to conduct a mail survey of NAPM firms that

have cooperative buyer-supplier relationships.

Stage 1.

Stage one is construction of the survey instrument which will address several key
antecedents to trust developments. The questionnaire will provide the actual investigation
into the strategies of managers and firms; therefore, it is important that the instrument is
constructed properly.

Experts agree that measurement items, should be developed: literature searches,
experience surveys, focus groups involving executives, and analysis of critical incidents
(Nunnally, 1978). Construction of the survey instrument of this study started with a
detailed study and analysis of the trust literature detailed in Chapter Two. The goal of the
literature review was to identify key constructs to trust development that have not been
tested. To test the face validity of the selected items, business school faculty and doctoral
candidates with familiarity in the area of cooperative relationships will be used as expert

judges. This will be done because three of the four antecedent variables have little

116

empirical work to draw on from the past. After several iterations of refining the survey
items, a content analysis performed on the items remaining to determine the overlap of
the remaining items with the conceptual domain of the constructs. The resulting
instrument will then be used to conduct pre-test interviews with cooperative relationship
experienced executives in the Lansing, Jackson, and Detroit area. To identify any
problems with question wording or layout, the interviews will range from one hour to
two and a half hours, which will hopefully yield good suggestions to strengthen the
validity and content of the instrument. A final step to further purify measures is the use
of a Q-sort. A Q-sort involves giving respondents the list of questions and descriptions of
each construct and having them put each question with the construct they feel it
represents. The results of the Q-sort were used to change and delete questions the
respondents did not agree on (Results of the Q-sort are in Appendices A and B). Results
from the Q-sort indicate a clear difference between calculus-based trust and the other two
types of trust. There is not a clear difference between knowledge-based and

identification-based trust.

Stage 2.

Stage two is to take the modified instrument and mail it to 1,250 NAPM (National
Association of Purchasing Management) title 1 members. The strategy will be to send
1,250 packets including a cover letter, survey and a postage-paid return envelope. Two
weeks after the initial mailing reminder postcards will be mailed to those who have not
responded. Non-response bias will be examined comparing the demographic

characteristics of the respondents with the non-respondents and executives returning

117

unusable responses. A one way between groups analysis of variance will determine if
statistical significance exists between the two groups. Chi square differences between the
groups looking at product categories and primary geographic markets can also be tested
for significance.

According to Cook and Campbell, the most representative samples will be those
that are randomly chosen from the population. This sampling strategy is referred to as the
random sampling for representativeness model. A random sample will be drawn from the
total number of title 1 members which is approximately 3,000.

The focus of this study will be the correlated antecedents to trust and a look at
whether trust impacts relationship performance. For this reason, this study is cross
sectional in design and does not attempt to explain the longitudinal development of
buyer-supplier relationships. Rather, this study will Show what tools are effectively

correlated with high levels of trust and whether trust impacts relationship performance.

3. 9.2 Validity Issues

A high response rate from key executives involved with relationships is the
optimal outcome, but even a high response rate will not address several key issues
including common method variance (Campbell & Fiske 1959) or the social desirability
problem (Podsakoff & Organ 1986: 534-536). Using self report data from one source
opens the data to contamination from correlations among variables. Sometimes,
however, key informants provide the only avenue for information desired (Huber &

Power 1985), and “the practical utility of same source self-report measures makes them

118

virtually indispensable in many research contexts (op cit Parkhe 1993: 810, Podsakoff &
Organ 1986: 540).

Cook and Campbell note that there are four types of validity which all research
should be judged upon: statistical conclusion validity, construct validity, internal validity,
and external validity. Statistical conclusion validity is a potential problem for this study
since the power will not be high. Cook and Campbell note that when the sample size is
small it is dangerous to rely solely on statistical significance.

Construct validity is a major problem with case studies, but the statistical
procedures and design of this study should limit construct validity problems. Most of the
measures used will have been previously tested except relationship time horizon,
trustworthiness and asset specificity behavior. This sample will be large enough (Schwab
1980) to perform factor analysis to be sure that all items load on the same underlying
construct. In addition, the design of the study will combat construct validity by the
following: having pre-specified relationships based on an exhaustive search of the
literature, pre-testing and refining of the measures, and collecting information from
multiple sources. This combination of strategies should mitigate some construct validity
problems (Devellis 1991; Yin 1994). This sample will be large enough (Schwab 1980) to
perform factor analysis to be sure that all items load on the same underlying construct.
The construct for knowledge -based trust along with performance and asset specificity
should have some degree of validity because they will be based on prior work.
Relationship time horizon and trustworthiness are new constructs where construct validity

will be an issue. To address this issue, pre-testing field interviews will be performed

119

prior to the study to refine these three measures. In addition, these constructs will be
tested for construct validity using factor analysis.

Internal validity concerns deal with establishing causal relationships between
variables and non-randomized designs Cook and Campbell(l976) suggest that the
researcher must explicitly rule out all possible threats one by one. Many of the possible
threats are not of concern to this study because the design is not interested in the changes
that occur when trust is present but rather is there a strong correlation between specific
antecedent conditions and trust in buyer-supplier relationships. Therefore, problems such
as history, maturation, instrumentation, and statistical regression to the mean may not be
a problem.

The final validity issue is external validity the ability to generalize beyond the
present study. The survey and field studies will cover a wide range of industries and
technologies. Given the variety of firms surveyed and visited this dissertation should

have strong generalizability.

3. 9. 3 Reliability Issues

Reliability will be addressed in several ways. Initial pre-test visits can be used to
refine the survey instrument. Questions that are unclear or that measure constructs other
than those of interest can be changed or adjusted, reducing ambiguity and increasing
reliability (Kerlinger 1986).

The methods addressed above will aid in increasing reliability; however, they are

not measures of reliability and will not guarantee statistical conclusions. In order to

120

ensure that the measures are reliable, reliability will be statistically analyzed using

Cronbach’s alpha.

3.10 Statistics

Structural equations modeling (SEM) will be used to analyze the data. SEM
consists of two parts which are a measurement model and structural equation model. The
measurement model specifies how latent variables and hypothesized constructs are
observed (measured) and describe their measurement properties (reliabilities and validity)
(Schumacker and Lomax 1996). The structural equation model specifies direct and
indirect relationships among latent variables and describes the amount of explained and
unexplained variance (Schumacker and Lomax 1996).

Confirmatory factor analysis will yield the measurement model to assess latent
variables. Convergent validity of the constructs will be determined through the use of
confirmatory factor analysis. This method tests a priori hypotheses regarding the
qualitative and qualitative distinctions underlying each domain. Fit statistics will be
analyzed in creating the measurement model (confirmatory factor analysis). Fit criteria to
be used will include an adjusted goodness of fit index, Bentler and Bonnett’s normal fit
index and Bentler’s comparative fit index. All three of these statistics need to be above
.90 level (Bagozzi and Yi 1988; Byme 1994). Bentler and Bonnett’s normed fit index
tests the proportionate reduction in fitting function when moving the baseline model
(assumes no factors underlie observed variables) to the hypothesized model. Bentler’s

comparative fit index takes into account the means of the sampling distributions in

121

comparing fit. Fit statistics can be compared between the comfirmatory factor analysis
and a-priori model to indicate if relationships were left out of the a—priori model.
Figure 3-3 shows the a-priori path relationships specified in the hypotheses.

COERPOW = construct for coercive power

NCOERPOW = construct for non-coercive power

DEPEN = construct for dependence

TIMHOR = construct for relationship time horizon

SPECBEH = construct asset specificity behavior

SPECDED = construct for dedicated asset specficity

RISK = construct for risk

CALCU = construct for calculus-based trust

KNOWL = construct for knowledge-based trust

IDENTI = construct for identificaiton-based trust

RELPERF = construct for relationship performance

B1 = path coefficient relating coercive power to calculus-based trust

B2 = path coefficient relating non-coercive power to knowledge-based trust

B3 = path coefficient relating non-coercive to identification-based trust

B4 = path coefficient relating interdependence to identification-based trust

B5 = path coefficient relating alliance time horizon to knowledge-based trust
B6 = path coefficient relating alliance time horizon to identification-based trust

B7 = path coefficient relating asset specificity behavior to identification-based
trust

122

 

B8 = path coefficient relating dedicated asset specificity to knowledge-based trust
B9 = path coefficient relating dedicated asset specificity to identification-based
B10 = path coefficient relating risk to identification-based trust

B1 1 = path coefficient relating knowledge-based trust to relationships
performance

B 12 = path coefficient relating identification-based trust to relationship

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

performance
Coercive Pom ”9"”
(mmwr) [31
152
WWW
(tummy)
(Halts-Based
Trust (CALCU)
Damian [34
(mm
B11
Relationslip [’5 > Tasman.) ——> W”
. . Pcrfmrnru:
Plarnrgihizm [B (RHJ’ERF)
(1le)
B12
AssetSpecificity [37
Beirm'u lderritieumeascd
(spams Tm (11mm
138
Muted
(Stamp)
B10
Rid:
(RISK)

 

 

 

Finally, the significance of each path coefficient is tested using the critical ration which is

the ration of regression weight to standard error. Under assumptions of independent

123

observations and multivariate normal distributions this ration approximates the standard
normal distribution and is used to test significance (Bollen 1989).

After a measurement model has been specified and indicates that the latent
variables are measured well (valid and reliable) by observed variables it is time to specify
a structural model. A structural model allows for relationships among the latent variables
depicted by lines of arrows. Since we do not have a measurement model yet there is not

structural model.

3.11 Limitations of This Research

There are several potential limitations to this dissertation. First, the population is
drawing its sample from the National Association of Purchasing Management (NAPM)
which has a purchasing specific membership. The conclusions will be from a purchasing
perspective which may include inherent power asymmetries not found in all cooperative
relationships. Future studies may benefit from studying trust in relationships from
perspectives other than purchasing. Second, the sample Size used to collect the data from
the target population may adversely affect the statistical analyses used to assess the data.
The marginal cost of increasing the sample size is extremely high , which limits the
number of cooperative buyer-supplier relationships that can be used.

Third, non respondent bias may be a problem. No data will be collected from
non-respondents so it will be impossible to detect this potential bias. However, it is
expected that most respondent firms that agree to participate in this study will have

favorable attitudes regarding cooperative buyer-supplier relationships since they already

124

engage in at least one of these relationships. Conversely, non-respondents may be
negative or neutral towards the use of cooperative buyer-supplier relationships.

A fourth source of potential bias could be introduced through the buying firm’s
choice of a cooperative buyer-supplier relationship on which they will base their
response. No constraints will be placed on the selection criteria other than the supplier
“partner” chosen to participate must have been involved with this particular buying firm
in a cooperative, buyer-supplier relationship as defined here for at least one year. We will
ask the buying firms to choose their best supplier and outline clearly what is meant by
“best”. Even by outlining these criteria of what the best supplier is firms may choose
their best supplier based on their emotions or feelings towards who their best supplier is.
Other factors including personal relationships with members of the supplier firm and

length of relationship with the supplier firm may impose bias on this study.

125

Chapter IV

Analysis of Data

4.1 Overview

The research findings are categorized into four general sets of results that are
consistent with the methodological plan described in chapter III. First, the results of a
general analysis of company profiles and frequency information is examined, including a
look at normality. Second, the results of propositions investigating whether different
types of trust exist is examined through a confirmatory factor analysis. Third, Individual
versus Organizational dimensions of trust are analyzed by assessing a set of paired t-tests.
Finally, the full structural model specifying the antecedents to different types of trust
(discovered in propositions l and 2) is tested.
4.1.1 Description of the Sample

Respondents at 1,445 organizations were mailed surveys and 359 usable surveys
were received, resulting in a 24.8% response rate. Fourteen surveys were not included in
the final sample in order to improve multivariate kurtosis problems (discussed later in this
chapter).

Title one members of NAPM (the National Association of Purchasing
Management) were identified as the sampling frame. Title one members include

purchasing managers and all other higher ranking purchasing executives including, vice

126

president of procurement, vice presidents of materials management, and director of
purchasing. Approximately 3,000 title one members exist. This study focuses on
industrial respondents and did not include non-industrial members. Industries
represented in the sample included: chemicals & related products, electrical/electronic
equipment, fabricated metal products, food and related products, furniture, miscellaneous
manufacturing, paper and related products, primary metal, printing and allied industries,
rubber and plastics products and textile mill products.

As shown below in Table 4-1, the respondents were primarily high ranking
purchasing executives, including: 150 directors of purchasing, 49 Vice-Presidents of
Purchasing, 45 Purchasing Managers, 30 Vice-Presidents of Materials Management and
70 respondents falling in the other category. Of these 70, 37 respondents were in

ownership, president or similar roles at their firms. Nineteen of the respondents were

Table 4-1 Respondent’s Titles

 

 

 

 

 

 

 

 

 

 

 

Title Frequency Percentage
Director of Purchasing 150 44%
Vice-President of Purchasing 49 14%
Purchasing Manager 45 13%
Vice-President of Materials Management 30 9%
Other 70 20%
TOTAL 344 100%

 

127

buyers and the remaining 14 were in middle management roles such as director of
systems or vice-president in charge of operations. Respondents had an average of 12
years of experience with their companies.

Table 4-2 presents the frequencies and percentages of respondent’s companies’
gross annual sales. The three largest categories, which include companies with gross

annual sales over $100 million and above, comprise 65% of the sample. Thus, the

respondent group was comprised primarily of medium to large firms.

Table 4-2 Respondent Companies’ Gross Annual Sales

 

 

 

 

 

 

 

 

 

 

 

 

 

Frequency Percentage
Less than $5 Million 11 3%
$5 million to $10 million 8 2%
$10 million to $50 million 48 14%
$50 million to $100 million 50 15%
$100 million to $500 million 77 22%
$500 million to $1 billion 42 12%
Over $1 billion 108 31%
TOTAL 344 Approx. 100%

 

4.2 Non-Response Bias

The best method to minimize non-response bias is to increase response rates. As
noted in chapter 111, the Dillman method, (requiring three separate mailings to the target
sample,) was used in an effort to maximize the response rate. One way of testing for non-

response bias is to test for statistically significant differences in the responses of early and

128

late waves of returned surveys (Armstrong & Overton 1977). The non-response bias test
used ten randomly selected survey items, and t—tests were performed on the two groups of
50 respondents each. The groups represented the first 50 respondents and last 50
respondents of the 346 responses used. The t-tests yielded no statistically significant
differences among the ten survey items tested. Thus, the data analysis proceeded with

analysis of the measurement and structural models as discussed in chapter 111.

4.3 Missing Data

There are several different strategies for dealing with missing data including
listwise deletion, pairwise deletion (Wilks 1932), mean replacement (Afifi & Elashoff
1966), regression replacement (Gleason & Staelin, 1975), maximum likelihood
estimation (Alison 1987) and similar response pattern imputation (Joreskog & Sorbom,
1993). In using structural equation modeling it is crucial that there be no missing data be
handled appropriately (Bentler 1993).

Listwise deletion was used to handle missing data on the dependent variable.
Listwise deletion was appropriate because a large percentage of the cases with missing
data on the dependent variable were missing a substantial portion of the dependent
variable data. Several respondents failed to fill out the final page of the questionnaire
which included all the information on individual level trust. Mean replacement was used
for replacement of data on the independent variable. Mean replacement was selected for
two reasons. First, mean replacement on the independent variable avoids contaminating
the correlation between independent and dependent variable error terms in the phi matrix.

Second, in this sample the number of missing data point for the independent variables is

129

much less than for the dependent variable. Only 22 missing data points were replaced on

the independent variable.

4.4 Initial data analysis
4. 4. I Univariate Distributions

After the initial data collection was completed, the univariate distributions were
examined to detect non-normality. One of the assumptions necessary for the use of
maximum likelihood (ML) function is that the variables are normally distributed.
Excessive kurtosis usually eliminates asymptotic efficiency and renders the estimated
asymptotic covariance matrix and chi-squared estimator potentially inaccurate (Browne,
1982). When nonnorrnality or excessive kurtosis threatens the validity of the ML
significance tests, Bollen (1989) recommends using transformations of the variables that
approximate multinorrnality or that remove excessive kurtosis. Table 4-3 below shows

the univariate distributions for all variables in this study.

Table 4-3 - Univariate Distributions of all Variables

 

 

 

 

 

 

 

Mean Std Dev. Kurtosis Skewness
Coercive Power 4.887 1.617 -.743 -.492
V1
Coercive Power 5.364 1.531 -.260 -.796
V2
Non-Coercive Power 2.991 1.362 .023 .757
V7
Non-Coercive Power 2.555 1.157 .887 .889
V8
Dependence 3.787 1.700 -.896 .164
V15
Dependence 2.828 1.459 -.072 .785
V17
Time Horizon 4.773 1.373 -.504 -.283
V25

 

 

 

 

 

 

 

130

Table 4-3 -Continued

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Mean Std Dev. Kurtosis Skewness
Human Asset Spec. 2.458 1.698 .987 1.342
V37
Human Asset Spec. 2.708 1.554 .203 .886
V38
Dedicated Asset Spec. 2.808 1.657 -.227 .804
V43
Dedicated Asset Spec 3.581 1.931 -l.09l .336
V45
Uncertainty 4.395 1.643 -.899 -.131
V49
Calculus-Based Trust 5.930 1.253 1.827 -1.407
V57
Calculus-Based Trust 5.724 1.398 .429 -1.108
V58
Calculus-Based Trust 5.852 1.295 2.310 -1.511
V60
Knowledge-Based Trust 2.099 1.141 2.779 1.487
V66
Knowledge-Based Trust 2.224 1.195 2.617 1.468
V67
Identification-Based Trust 2.590 1.216 .106 .676
V78
Identification-Based Trust 2.462 .842 .106 .676
V79
Identification-Based Trust 2.430 1.099 1.023 .923
V80
Performance 2.015 1.017 2.209 1.208
V85
Performance 2.541 1.157 1.040 .935
V86

 

 

 

 

 

 

Some of the variables display excessive kurtosis and/or skewness. In particular,
the coefficients of skewness and kurtosis were large for variables 57, 60, 66, 67, and 85.
For variables that are negatively skewed (variables 66, and 67) a log transformation is
recommended (West, Finch and Curran 1996) and was applied. Variables that are

positively skewed should be raised to a power greater than one in their transformation

131

(West, Finch and Curran 1996) which was done (to variables 57 and 60). Below are the

descriptive statistics for the transformed variables.

Table 4-4 - Transformed Univariate Distributions of all Variables

 

 

 

 

 

 

 

 

 

 

 

Mean Std Dev. Kurtosis Skewness

Calculus-Based Trust 36.733 12.839 -.222 -.819
V57

Calculus-Based Trust 35.916 12.858 -.064 -.815
V60

Knowledge-Based Trust .266 .218 -.697 .246
V66

Knowledge-Based Trust .292 .218 -.571 .164
V67

Dependence 3.787 1.700 -.896 .164
V15

Identification-Based Trust .253 .212 -.994 .137
V85

 

 

Another problem that can be encountered in data analysis is a lack of multivariate

normality. Kurtosis is a statistic that describes peakedness of a frequency distribution

compared to a normal distribution (Hair et al. 1992). Mardia’s coefficient and

normalized estimates are statistical indicators of multivariate normality. Mardia’s

coefficient for this set of data was 78 and the normalized estimate was 49 which were

quite high. Ideally Mardia’s coefficient normalized estimate should be as low as possible.

EQS indicates which cases should be removed to lower kurtosis. Based on these

suggestions 15 cases were eliminated and Mardia’s coefficient was lowered to 38.74 or

nearly 50% and a normalized estimate of 23.19, a 52% decrease.

The preceding analyses served to meet preliminary assumptions of normality

required to use the structural equation methodology. In the next section the measurement

properties of the latent constructs are examined to demonstrate evidence of convergent,

discriminant and nomological validity.

4.5 Confirmatory Factor Analysis
Before embarking upon further analysis the measurement properties of the four
latent variable constructs were examined. In the analysis the propositions regarding

differences among the three different types of trust (proposition 1) and differences

Table 4-5 Statistics for Manifest Variables of Trust (Measurement Model)

Latent Variables and Their Factor Standard
Manifest Variables Loadings Errors t-Value r

Calculus-Based Trust
Question # 57: We feel that this key supplier 0.928 0.539 22.099 0.373
takes advantage of us.
Question # 58: We feel that this key supplier 0.896 0.060 20.853 0.445
tries to get the upper hand.
Question #60: I think this key supplier takes 0.798 0.591 17.146 0.615
advantage of our vulnerabilities
Knowledge-Based Trust
Question #66: I feel that this key supplier 0.917 0.009 21.681 0.582
will keep its word.
Question #67: I think that this key supplier 0.917 0.009 21.703 0.526
does not mislead us.
Identification-Based Trust
Question #78: We think this key supplier is 0.863 0.053 19.849 0.505
open in describing their strengths and
weaknesses in negotiating joint projects.
Question #79: We know this key supplier 0.896 0.047 21.1 10 0.444
will consider our concerns when making a
decision.

Question #80: I feel that our key supplier 0.945 0.045 23.113 0.328
negotiates joint expectations fairly.
Performance

Question #85: In this relationship, the parties 0.766 0.01 1 14.934 0.643
work together to solve problems.
Question #86: This relationship has met our 0.594 0.061 11.339 0.805
profitability projections thus far.

 

 

 

 

 

 

 

 

 

 

 

* All loadings significant at p<.01

133

between individual and organizational trust (proposition 2) were also examined. These
analyses were done to ensure consistency and reliability among each of the measures. A
short discussion of the practical significance of the changes to the measurement model is
also included with each analysis.

Based upon these results, several modifications to the original structural model
were made. These respecifications were substantively based on the two measurement

models subsequently discussed.

4. 5 . 1 Trust Measurement Model

The first measurement model tested proposition 1 stating that three different types
of trust exist (calculus, knowledge and identification-based). Figure 4.1 below shows the
model tested in proposition one as well as the results. Initially the model with all of the
hypothesized indicators was tested but because of significant crossloadings among
manifest variables a more parsimonious model was used.

An important criteria in supporting a strong measurement model are indicators of
convergent and discriminant validity (Anderson & Gerbing 1988) among the latent
variables. Bagozzi, Yi, and Phillips add that a test for convergent validity iinvolves
testing the significance of all factor loadings. T-values will indicate whether factor
loadings are significantly different from zero. In the measurement model in Figure 4-1,
all factor loadings were found to be significant providing evidence of convergent validity.

Discriminant validity can be established in two ways. First, examining the phi
matrix can help to determine if the exogenous latent variables are correlated strongly with

each other. Large significant correlations may indicate a lack of discriminant validity in

134

the model. The high correlation between knowledge-based and identification-based trust

((1)23: .817), however, does not support discriminant validity between those two latent

Hgne41'lhstNhsrmmlMxH

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

variables. The high correlation between knowledge-based and identification-based trust
would suggest the possibility of these two constructs having a common second order
construct underlying them. One way to test for this possibility is to run a second order
factor model and a single factor model and compare the results with a two factor model.

The results are Shown in Table 4-6.

135

Table 4-6 Trust Measurement Model Fit Statistics

 

 

 

 

 

 

 

Criterion One Factor Two Factor 2"d Order
Bentler & Bonett Nonnorrned Fit Index .764 .996 .992
Bentler and BOnett’s Norrned Fit Index .855 .994 .994
Bentler’s Comparative Fit Index .858 .998 .997
Chi-Square (x2) 279 1 1.84 11.840
Degrees of Freedom 9 8 6
Probability of a > Chi-Square p<.001 .15849 .06563

 

 

 

 

 

 

A comparison of the chi-squared fit statistics suggest the nature of the relationship
between knowledge-based and identification-based trust. Table 4-6 above compares the
fit statistic among the three different models run. The fit statistics for the one factor
model’s (combination of knowledge-based and identification-based variables into one
factor) are all substantially below the threshold value (.90) so that option is eliminated.
The fit statistics for the second order and two factor models show an excellent fit in both
cases suggesting the existence of a second order factor.

Nomological validity of a construct exists when a latent variable relates to other
constructs in a way that is consistent with underlying theory. In chapter II, our theoretical
model predicted a hierarchy of correlation between the three types of trust (Lewicki &
Bunker 1995) and performance. The deeper or stronger the level of trust the stronger the
relationship to performance. The factor loadings from Figure 4-1 indeed support this
theory. The correlations are -.825 for calculus-based trust, .958 for knowledge-based
trust and .968 for identification-based trust. The negative correlations between calculus-

based trust and knowledge-based trust (-.779) and identification-based trust (-.718)

136

TRUST

Figure 4-2

Lewicki & Bunker’s Model

 

Identification-Base
Trust

 

 

Knowledge-Based
Trust

 

 

 

Calculus-Based
Trust .

 

 

TRUST

 

  
 

Calculus-Based

New Model Supported by this Study

  
   
 
 

Identification-Based
Trust

Knowledge-Based
Trust

 

TIME

Trust

137

suggest a major modification to Lewicki and Bunker’s model of trust development as
shown below in Figure 4-2. Rather than knowledge-based and identification-based trust
building on it appears that calculus-based trust should be avoided altogether. The
avoidance of calculus-based trust possibly provides a foundation for knowledge and
identification-based trust. Moreover, out model suggests that organizations establishing
new relationships with suppliers can “jump” into a strong relationship by establishing
stronger ties immediately. The new model in Figure 4-2 shows this new relationship
between the three types of trust which only partially supports Proposition 1 stating that
three unique types of trust exist(it appears that only two forms of trust exist). The
implications of this model will be discussed further in chapter V.

Nomological validity is also supported in analyzing the difference in discriminant
validity between the three types of trust. As mentioned above the difference between
knowledge-based and identification-based trust is not strongly supported through the
discriminant validity analysis. This is not surprising because at both of these levels of
trust a significant investment is made by both parties in the form of information sharing
and emotional investment in the relationship. Sorting out the knowledge-based
components from the emotional or identification-based components may not be possible
because both types of trust occur simultaneously. A more detailed description of the
theoretical implications of this issue are addressed in chapter V. In conclusion, all
relationships in the model support nomological validity.

Table 4-7 shows the goodness of fit indices and the chi-squared statistic for the
confirmatory factor analysis of the trust measurement model. As can be seen in Table 4-

7, all three fit indices indicate an acceptable fit of the measurement part of the model.

138

Table 4-7 Trust Measurement Model Fit Statistics

 

 

 

 

 

 

 

 

Criterion Value Desirable
Ranges

Bentler & Bonett Nonnorrned Fit Index .984 >090
Bentler and Bonett’s Norrned Fit Index .981 >090
Bentler’s Comparative Fit Index .990 >090
Chi-Square (x2) 58

Degrees of Freedom 29

Probability of a > Chi-Square p<.001 >0.05
Largest Standardized Residual .053 <2.00

 

 

 

 

 

Each of the fit indices is well above the desired minimum value of .90. The largest
standardized residual was .053 which is lower than the 2.00 standard espoused by
Bagozzi and Yi (1988) and is not unacceptably large. Table 4-5 shows the factor
loadings, standard errors, t-test, and R2 statistics for the four latent variables tested in the

trust measurement model.

4. 5.2 Individual versus Organizational Trust

Proposition 2 states that no significant difference between individual and
organizational trust is expected. With an acceptable trust measurement model in place
proposition two was analyzed which tests whether individual and organizational trust are
unique. A search of the literature on trust in chapter 11 revealed that there is little research
analyzing potential differences between individual and organizational trust. This
proposition is tested using paired t-tests of the means of similar questions on calculus,

knowledge, and identification-based trust at the organizational and individual level.

139

 

Table 4-8 Analysis of Individual versus Organizational Trust

Questions t-test p-value Sippificance

 

1a. We have a sharing relationship, where my -1.7022 .0896 No
primary contact and I can both freely share
ideas, concepts, or information.

85. In this relationship, the parties work
to ether to solve problems.

 

2a. I can talk freely to this person about - .0000 Yes
difficulties we are having to know that 12.094
he/she will listen. 4

78. We think this supplier is open in describing
their strengths and weaknesses in

negotiating joint project.

 

6a. This primary contact acts with integrity. -6.5281 .0000 Yes
73. I feel our key supplier does not try to get out
of commitments.

 

7a. I regularly communicate with the 2.1342 .0335 Yes
communicates with us.

65. We have frequent fact to face contact with
this supplier.

 

10a. The quality of work obtained from this 4.3334 .0000 Yes
primary contact is only maintained by
diligent monitoring.

59. We monitor this key supplier closely, so
that they cannot take advantage of us.

 

2a. I have faith in my contact person to look 4.5492 .0000 Yes
out for my best interest even when it is
costly to do so.

88. This key supplier is genuinely concerned
about our company’s welfare.

 

 

 

 

 

Table 48 above summarizes the t-test results. Seven of the eight t-tests were
found to be significant at the 0.05 alpha level. Only the t-test between questions 1a and
78 were not found to be Significant (This result is actually significant at the p<.10 level).
These results do not support proposition 2 and imply that respondents perceived a

difference between individual and organizational trust. This result has important

140

 

implications for research on trust, as it counters previous theory described by Ring and

Van deVen. This will be discussed firrther in chapter V.

4.6 Measurement and Structural Model Testing of Trust Antecedents

Then next step of this analysis involved testing hypotheses 1 through 7 which
posits associations between a set of antecedents and the three types of trust. The factor
loadings, standard errors, t-values and r2 statistics for the trust antecedents are shown in
Table 4-9 below. Each of the individual items remaining in the analysis were significant
indicators of their respective factors with t—values greater than 1.96, p < 0.05 and the
coefficient signs were all positive as expected. The overall fit of the antecedent
measurement model which is summarized below shows fit indices all above the
acceptable fit of the measurement model with the exception of the Adjusted Goodness-of-
F it index. However, the AGI was only slightly below the threshold value of 0.90 judged

acceptable enough to proceed with a test of the structural model.

4.7 Testing of Structural Models

The difference between confirmatory and structural equation modeling is that the
latter employs structural equations to test relationships among the latent factors in the
model. The relationships between the observed or manifest variables and their
corresponding constructs that constitute the measurement model of this study is brought
out through an analysis of structural relationships.

Structural equation modeling combines multiple regression and factor analysis to

simultaneously test a series of dependent relationships (Hair et al 1992). Hypotheses l to

141

Table 4-9 Statistics for Manifest Variables of Trust Measurement Model

Latent Variables and Their
Manifest Variables

Standard
Loadings Errors

Factor
t-Value r2

 

Coercive Power

Question #2: We will take quick punitive
measures (e.g. reduce volumes) if this key
supplier does not meet our performance
expectations.

Question #3: We occasionally use fear to
motivate this key supplier.

0.868

0.504

0.129

0.175

6.322

7.599

0.864

0.496

 

Non-Coercive Power

Question #7: We have few problems
influencing this key supplier.

Question #8: This key supplier is highly
receptive to our ideas.

0.742

0.803

0.057

0.056

17.488

18.656

0.670

0.596

 

Supplier Dependence

Question #15: This key supplier is dependent
on us.

Question #17: This key supplier would find
it costly to lose us

0.811

0.760

0.080

0.076

17.441

19.565

0.585

0.650

 

Planning Horizon

Question #25: If our key supplier is gaining
more from the relationship than we are, we
always take steps to end or alter the
agreement.

1.000

0.055

16.764

0.000

 

Human Asset Specificity

Question #37: Non-sales personnel from our
key supplier have visited our plant to see
how their product is used in our process.
Question #38: Personnel from our key
supplier work closely with our engineering
and design area in developing new products.

0.659

0.798

0.079

0.080

16.125

15.596

0.752

0.603

 

Dedicated Asset Specificity

Question #43: Our key supplier has
dedicated equipment or capacity specifically
to maintain our relationship.

Question #45: Our key supplier has
purchased specialized equipment to meet our
needs of this key-input material.

0.869

0.791

0.080

0.075

17.246

13.345

0.496

0.612

 

 

Uncertainty
Question #47: Technological change is rapid
in the industry this key supplier works in.

 

0.613

 

0.112

 

14.394

 

0.790

 

I42

 

6 involve the testing of three separate structural models (one for each type of trust).

Figures 4-3 to 4-5 below depict the three structural models tested.

Table 4-10 Calculus-Based Trust

 

 

 

 

 

 

 

 

Criterion Value Desirable
Ranges

Bentler & Bonett Nonnormed Fit Index .983 >090
Bentler and Bonett’s Normed Fit Index .994 >090
Bentler’s Comparative Fit Index .997 >090
Chi-Square (x2) 4.722

Degrees of Freedom 2

Probability of a > Chi-Square .09434 >0.05

 

 

 

 

The first model analyzes calculus-based trust which was hypothesized to be

positively associated with coercive power. Fit statistics for the model are in Table 4-10

below. All three fit indices are above the .90 level which shows a good fit. The Chi-

square statistic is significant at the or = .05 level which supports Hypothesis la.

HlA

 

Coercive Power

 

 

 

.395‘ Trust (Organizational)

Figure 4-3

 

+ Calculus-Based

 

 

 

* Indicates significant at p<.01

143

Figure 4-3 above shows the relationship between coercive power and calculus-
based trust. The equation below indicates the structural equation model for calculus-
based trust. The top number is the unstandardized beta weight, the second number is the
standardized beta weight, the third number the standard error and the final number the t-
value. Hypothesis 1A postulated an association between calculus—based trust and coercive
power. The structural equation below shows a significant relationships between these

two constructs.

Calculus-Based = .714 Coercive + 1.0D
Trust .714 Power + .919D
.133
5.360

Figure 4-4 Knowledge-Based Trust

 

Non-coercive

 

 

 

 

 

 

 

 

 

 

 

 

 

Power H 13

.070‘

H3A. -.279

Relationship ’ Knowledge-Based
Planning Horizon Trust (Organizational)
.080
HSA
Dedicated

 

 

Asset Specificity

 

* Indicates significant at p<.01 level

144

The second structural model in Figure 4-4 analyzes the antecedents of knowledge-
based trust. All three fit indices as seen in Table 4-11 below are above the .90 accepted
cutoff level.

Analysis of the structural model indicates support for hypotheses HlB, however
hypotheses H3A and H5A are not supported. Moreover, there is evidence to support that
the use of non-coercive power leads to knowledge-based trust, but that asset specificity

behavior and dedicated asset specificity do not. These results will be discussed in

 

 

 

 

 

 

 

 

chapter V.
Table 4-11 Knowledge-Based Trust
Criterion Value Desirable Ranges

Bentler & Bonett Nonnormed Fit Index .952 >090
Bentler and Bonett’s Normed Fit Index .927 >090
Bentler’s Comparative Fit Index .962 >090
Chi-Square (x2) 48

Degrees of Freedom 11

Probability of a > Chi-Square p<.001 >0.05
Largest Standardized Residual <2.00

 

 

 

 

 

The equation below indicates the structural equation model for knowledge-based
trust. The top number is the unstandardized beta weight, the second number is the
standardized beta weight, the third number the standard error and the final number the t-
value. Only Non-coercive power was found to have a significant relationship with
knowledge-based trust. Both dedicated asset specificity and relationship planning

horizon while close to the 1.96 t-statistic cutoff are not statistically significant.

145

Knowledge-Based = .010 Dedicated Asset + .156 Non-Coercive + -.145 Planning + 1.0D

Trust .080 Specificity .702 Power -.279 Horizon .650
.006 .015 .084
1.757 10.297 -1.724

Table 4-12 Identification-Based Trust

 

 

 

 

 

 

 

Criterion Value DesirableRanges
Bentler & Bonett Nonnormed Fit Index .758 >090
Bentler and Bonett’s Normed Fit Index .800 >090
Bentler’s Comparative Fit Index .824 >090
Chi-Square (7(2) 299
Degrees of Freedom ‘ 48
Probability of a > Chi-Square P<.001 >0.05

 

 

 

 

 

The final structural model tested relationships between six different antecedents
and identification-based trust. Table 4-12 above lists the fit indices for the identification-
based structural model. The fit indices indicate a poor fit. All three indices are
substantially below the .90 cutoff level.

A comparison of the identification-based structural model fit indices with the
measurement model fit statistics which were all above .90 indicate the presence of
relationships not specified in the identification-based trust model.

Figure 4-5 above shows the relationships among identification-based trust and its
antecedents. The equation below indicates the structural equation model for
Identification-based trust. The top number is the unstandardized beta weight, the second
number the standardized beta weight, the third number is the standard error, and the final
number the t-value. Dedicated asset specificity (H5B), non-coercive power (H1 C),

human asset specificity (H4), and uncertainty (H6, negative relationship) were found to

146

MEWM

 

 

 

 

 

 

 

 

 

 

 

 

 

 

lchtifica’nr-Bml
Trim (0311le ' ‘

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

*hxlmrssigiimntmmlewl

have significant associations with identification-based trust. Dependence (H2) and

relationship planning horizon (H3B) were not found to have significant associations.

Identification-Based = .182 Dedicated Asset + .632 Non-Coercive + .009 Dependence +
Trust .282 Specificity .688 Power .013
.037 .067 .039
4.922 9.438 .234
.144 Human Asset + -.297 Uncertainty + -.024 Time + .559D
.174 Specificity -.320 -.035 Horizon .560
.047 .077 .029
3.073 -3.833 -.809

147

Table 4-13 Additional Lagrangian Pathways Identified

 

 

 

 

 

 

 

 

 

 

 

 

 

Parameter Parameter Lagrangian Multiplier
Dependence Dedicated Asset Specificity 74.263
Human Asset Specificity Dedicated Asset Specificity 46.010
Non-Coercive Power Dedicated Asset Specificity 34.578
Uncertainty Dedicated Asset Specificity 21.584
Dependence Non-Coercive Power 17.398
Uncertainty Dependence l4. 1 3 3
Human Asset Specificity Dependence 13.895
Uncertainty Non-Coercive Power 12.428
Human Asset Specificity Non-Coercive Power 11.404
Uncertainty Human Asset Specificity 8.851
Time Horizon Human Asset Specificity 4.800

 

 

 

EQS output provides a list of Lagrangian Multipliers chi squared statistics which

tests whether fixed relationships in the model are in fact nonzero in the population and

would be better treated as free parameters in the future. Table 4-13 shows the following

additional paths were statistically significant in the Lagrangian Multiplier test of the

identification-based trust model.

Figure 4-6 below shows a modified model of identification-based trust and its

antecedents. Several paths were added and removed based on the results of the

Lagrangian multiplier test.

The fit indices for this modified model are all above .95 indicating a strong fitting

model. The implications of Figure 4-7 below will be discussed in chapter VI.

Table 4-14 summarizes the findings among all of the propositions and hypotheses tested.

148

 

Figure 4-6 Modified Identification-Based Trust Model

.459‘

 

.281'

 

l

 

 

 

.242‘

l

 

C
De endence 259. A55“ SPCCIDCIIY .160 Identification-Based Dedicated
p _—> Behavior _—> Trust (Organizational) —-> Asset Specificity

 

 

 

 

Non-coercive
Power

 

‘Indicates significant at p<.01 level

Table 4-14 Propositions and Hypotheses

l

 

Pl. Three types of trust exist which are calculus-based,
knowledge-based, and identification-based trust.

Supported

 

P2. Organization trust and individual trust are not unique

Not Supported

 

HlA. Use of coercive power will be positively associated with
calculus-based trust.

HlB. Use of non-coercive power will be positively associated
with knowledge-based trust.

HlC. Use of non-coercive power will be positively associated
with identification-based trust.

Supported
Supported

Supported

 

H2. Dependence between buyer and supplier will be positively
associated with the level of identification-based trust in a
supplier.

Not Supported

 

H3A. A short-term cooperative buyer-supplier relationship
planning horizon will be associated with the level of
knowledge-based trust in a supplier.

H3B. A short-term cooperative buyer-supplier relationship
planning horizon will be associated with the level of
Identification-based trust in a supplier.

Not Supported

Not Supported

 

 

H4. Asset Specificity behavior focusing on information sharing
and synergy will be positively associated with identification-
based trust in a supplier

Supported

 

 

149

 

Table 4-14 (Continued)

 

H5A. The level of dedicated asset specificity will be positively Not Supported
associated with the level of knowledge-based trust.
H5B. The level of dedicated asset specificity will be positively Supported
associated with the level of Identification-based trust.
H6. Environmental uncertainty will be positively associated Supported
with the level of identification-based trust
H7A. Cooperative buyer-supplier relationship performance Supported
will be positively associated with the level of knowledge-
based trust in a supplier.

H7B. Cooperative buyer-supplier relationship performance Supported
will be positively associated with the level of Identification-
based trust in a supplier.

 

 

 

 

 

 

4.8 Summary and Conclusions

The goal of this study was to investigate the interorganizational dynamics of trust
in ongoing buyers-supplier relationships. The outcomes of the data analysis reported in
this chapter are measures for three different types of trust, performance, and the key
antecedents to trust.

A second contribution of this research is the validation of existing factor scales
and development of new factor scales. Table 4-15 lists the final sets of items for each
factor and includes the coefficient alpha for each purified scale.

Finally, many of the a-priori relationships were found to be positive and
significant. Key findings include establishment of measures for the three types of trust
and strong relationships between the three types of trust and performance. In testing the
trust measures the possibility of identification and knowledge-based trust also being
components of a second order trust construct was established. Key results include:

- Coercive power positively associated with calculus-based trust

0 Non-coercive power positively associated with identification-based trust

150

0 Asset specificity behavior positively associated with identification-based trust

0 Dedicated asset specificity is positively associated with identification-based
trust

0 Uncertainty is negatively associated with identification-based trust.
Chapter V will provide an overview of the research and discussion of the findings

in this chapter.

Table 4-15 Statistics for Manifest Variables of Trust Measurement Model

 

Standardized Latent Variables Coefficient
and Their Manifest Variables Alpha
Calculus-Based Trust .842

Question # 57: We feel that this key supplier takes advantage of us.
Question # 58: We feel that this key supplier tries to get the upper hand.
Question #60: I think this key supplier takes advantage of our
vulnerabilities

Knowledge-Based Trust .914
Question #66: I feel that this key supplier will keep its word.
@estion #67: I think that this key supplier does not mislead us.
Identification-Based Trust .902
Question #78: We think this key supplier is open in describing their
strengths and weaknesses in negotiating joint projects.

Question #79: We know this key supplier will consider our concerns
when making a decision.

Question #80: I feel that our key supplier negotiates joint expectations
fairly.

Performance .605
Question #85: In this relationship, the parties work together to solve
problems.

Question #86: This relationship has met our profitability projections
thus far.

Coercive Power .608
Question #2: We will take quick punitive measures (e.g. reduce
volumes) if this key supplier does not meet our performance
expectations.

Question #3: We occasionally use fear to motivate this key supplier.
Non-Coercive Power .758
Question #7: We have few problems influencing this key supplier.
Question #8: This key supplier is highly receptive to our ideas.

 

 

 

 

 

 

 

 

 

151

Table 4-15 (Continued)

 

Standardized Latent Variables Coefficient
and Their Manifest Variables Alpha
Supplier Dependence .866

Question #15: This key supplier is dependent on us.

Question #17: This key supplier would find it costly to lose us
Planning Horizon -
Question #25: If our key supplier is gaining more from the relationship
than we are, we always take steps to end or alter the agreement.
Human Asset Specificity .687
Question #37: Non-sales personnel from our key supplier have visited
our plant to see how their product is used in our process.

Question #38: Personnel from our key supplier work closely with our
engineering and design area in developing new products.

Dedicated Asset Specificity .809
Question #43: Our key supplier has dedicated equipment or capacity
specifically to maintain our relationship.

Question #45: Our key supplier has purchased specialized equipment to
meet our needs of this key-input material.

Uncertainty -
Question #47: Technological change is rapid in the industry this key
supplier works in.

 

 

 

 

 

 

 

 

* All loadings significant at p<.01

152

CHAPTER V

SUMMARY AND DISCUSSION

In this chapter, the results of the measurement and structural model analyses are
interpreted and summarized. This chapter will also discuss implications of this study in
relation to previous research and suggest directions for future research. Finally, the
limitations of this current study are outlined. This chapter is organized into six sections.
Section one provides an overview of the findings of the study. It reviews the results
regarding the research questions, propositions, and hypotheses. Section two discusses the
research findings among each proposition and hypothesis tested. Section three discusses
the contributions of this study to prior theory in the areas of trust and interorganizational
relationships, and section four looks at the implications of this study on professional
practice. Section five presents limitations of the study, and section six provides

directions for future research.

5.1 Overview of Research

This section provides an overview of the methodology, research questions,
propositions, and hypotheses. This study involved an empirical investigation into the
definition of trust and identification of the key antecedents that develop trust. In this
study, three different types of trust were tested including: calculus, knowledge, and

identification-based trust. Factors that were potentially important antecedents to trust

153

development were identified through a thorough search of the trust and buyer-supplier
relationship literature. Based on the literature, a model of the significant antecedent
factors to trust development was constructed. This model (Figure 1-3) was tested using
data from a written survey instrument.

A survey methodology was used because prior research on trust was done
primarily in other disciplines such as economics or organizational behavior, and in order
to generalize these finding to buyer-supplier relationships, the rigorous use of statistical
testing from a large population of firms was required. A pretest of the survey instrument
was done to help ensure clarity, reliability, and validity of questions. Title one members
of the NAPM (National Association of Purchasing Management) membership database
were used because these upper level purchasing executives would have access to the data
sought on the questionnaire.

The research questions were developed with the a-priori model in Figure 1-3.
Three research questions that guide this investigation as follows:

1. What are the different types of trust experienced by purchasing executives in
cooperative buyer-supplier relationships?

2. What are the key antecedents that affect the level of trust experienced by
cooperative buyer-supplier relationships?

3. Are there significant differences between the three different types of trust and
cooperative buyer-supplier relationship performance?

The research questions above address specific relationships among factors that
were identified in the literature as important to the development of trust in buyer-supplier
relationships. The first two research questions explore the definition of trust in the realm

of buyer-supplier relationships and the relationship of trust to key antecedents. The final

154

research question addresses the relationship between different types of trust and

performance.

5.2 Discussion of Research Findings and the Theoretical Implications of Findings

As noted in chapter two, the study of trust has been long and rich (Deutsch 1958;
Solomon 1960; Strickland 195 8). While many studies on trust have been done (Mayer et
al. 1995; Hosmer 1995), the use of the term remains inconsistent (Barber 1983; Hosmer
1995; Zucker 1986). Trust has been associated with concepts including: reliability,
competence, goodwill, vulnerability, loyalty, and honesty, yet has never been specifically
measured and assessed within the context of buyer-supplier relationships. Recent
literature has explored the existence of multiple forms of trust which may or may not be a
function of time (McAllister 1993; Bromiley & Cummings 1996; Ghoshal & Bartlett
1995). These multiple forms of trust help to explain the wide range of terms or concepts
associated with trust.

A recent theory on trust from Lewicki & Bunker (1995) states that three types of
trust exist that have a hierarchical relationship as they develop as shown in Figure 3-2.
This dissertation is one of the first to test the existence of calculus-based, knowledge-
based, and identification-based trust theorized by Lewicki and Bunker. In addition, this
study is one of the first to explore the key antecedents associated with these three
different types of trust. The contributions of this study as they apply to previous studies
are presented in the following paragraphs. The following sections address the areas of
trust, power, dependence, asset specificity, uncertainty, relationship time horizon, and

performance against the existing literature on trust and trust development.

155

5. 2.1 Calculup-based. Knowledge-based. and Identification-baped Trust

Proposition 1 addresses the different types of trust present in key buyer-supplier
relationships and the results of proposition 1 are summarized in Table 5-1 below. The
three types of trust tested were calculus, knowledge, and identification-based trust. Prior
theory suggests that these three types of trust exist in a hierarchical relationship
progressing from calculus to knowledge and finally to identification-based trust.
However, not all relationships will make it to the knowledge or identification-based
stages. This study sought to verify the presence of the three types of trust, but because
this study is cross-sectional, the temporal relationships among the types of trust were not

tested.

Table 5-1 Propositions

 

 

 

Propositions/Hypotheses Result Implications

P1. Three types of trust exist which are Confirmed Three different types of

calculus-based, knowledge-based, and trust exist: Calculus,

identification-based trust. Knowledge and
Identification based
trust.

P2. Organizational trust and individual Not No support for

trust are not unique Supported proposition P2.
A significant difference
between organizational
and individual-based
trust exists.

HlA. Use of coercive power will be Supported Coercive power is

positively associated with calculus- positively associated

based trust. with calculus-based

HlB. Use of non-coercive power will Supported trust.

be positively associated with Non-coercive power is

knowledge-based trust. positively associated

H1 C. Use of non-coercive power will Supported with both knowledge

be positively associated with and identification-

identification-based trust. based trust.

 

 

 

 

156

 

Proposition 1 was tested by running a confirmatory factor analysis among the
three different types of trust and performance in an attempt to establish convergent,
discriminant, and nomological validity. Convergent and nomological validity were
established between the three types of trust and performance; however, discriminant
validity between knowledge-based and identification-based trust was questionable. A phi
value between identification and knowledge-based trust of over .80 indicated the
possibility that knowledge-based and identification-based trust are part of a second order
construct.

The high correlation between knowledge-based and identification-based trust is

not totally unexpected. Table 5-2 shows that knowledge-based trust addresses the areas

Table 5-2: Comparison of theory with results of this study

 

 

Type of Trust Lewicki & Bunker This Study
Calculus-Based 0 Fear 0 Fear
0 Punishment - Punishment

0 Monitoring

0 Opportunistic

 

 

Knowledge-Based o Predictability 0 Honesty
- Information 0 Not misleading
0 Regular
Communication
Identification-Based o Empathy o Openness

0 Mutual Understanding o Empathic
- Shared values 0 Fair

0 Joint goals

 

 

 

 

 

157

of honesty (keeping your word) and being straightforward in representing facts, while
identification-based trust addresses being open, fair, and empathetic. These two types of
trust address areas that are very similar. It would not be surprising to find respondents
simultaneously feeling that trust includes being honest, straightforward, open, fair, and
empathetic. The plausibility of this explanation led to investigation of identification and
knowledge-based trust possibly being components of a second order factor. A second
order factor occurs when two latent independent variables define a third latent variable.
A second order factor analysis was performed to indicate whether knowledge-based and
identification-based trust are two independent latent variables that make up a third latent
variable possibly called character-based trust. The second order factor analysis yielded
strong fit statistics indicating the possibility of a second order factor being present.

An alternative explanation to the high correlation between knowledge and
identification-based trust is the nature of the relationships surveyed. Respondents were
asked to direct their thoughts in responding to a key supplier providing a high expenditure
and strategic service or part. Since respondents commented on key suppliers, it is
possible that many respondents described a relationship that is mature. Results from
Question #23 of the survey which asked how many years the respondent has bought from
the key supplier supports this possible explanation. The average age of the relationship
used by respondents to answer the survey was just under 13 years. This indicates that
most respondents are looking at well established relationships that are deep into the
knowledge and identification-based trust stages. The lack of short or mid-term length
relationships in the sample could possibly explain why there is not strong discriminant

validity between knowledge and identification-based trust.

158

While the sample was slanted toward long and mid term length relationships the
relationship between length of relationship and trust was tested. Despite the bias of the
sample toward long-term relationships which would make it more difficult to see a
relationship between length of relationship and trust, t-tests found a significant
relationships. These results suggest that trust does not plateau after 10 years but actually
can continue to grow and develop as relationships become 15, 25 or even 35 years old.

The scale items used for these measures were developed from previous studies by
McAllister 1995, Bromiley and Cummings 1995, and Kumar 1996. In studies by
McAllister and Bromily and Cummings, multiple forms of trust were found. In
McAllister’s case, different antecedents were associated with two different types of trust.
McAllister’s study investigated the existence of three types of trust on an
interorganizational level. The three types of trust studied are different from any of the
multiple trust types developed in prior theory.

Calculus-based trust is similar to a construct tested by Bromily and Cummings
called “avoid taking excessive advantage of.” Calculus-based trust, however, goes
beyond Bromily and Cumming’s concept and is centered on conformance through threat
of punishment or fear. The fact that calculus-based trust emerged as a latent variable is
important, but maybe an even more significant finding was the impact of calculus-based
trust on performance. Recent work by Kumar 1996 shows empirical support for the
negative effects of fear on a buyer-supplier relationship, and this study supports Kumar’s
theory and findings. Threats and fear have been part of the buyer supplier landscape for
many years, but partnership and Kereitsu (Nishiguchi 1994) relationships suggest a move

toward more cooperative relationships. Skeptics, however, maintain that being tough and

159

using fear is still necessary, but this study strongly suggests that fear actually reduces
performance in key buyer-supplier relationships.

In developing the measures for trust, a tradeoff often encountered by researchers
between content validity and convergent validity was experienced (Dunn, Seaker, and
Waller 1994). In order to achieve high factor loadings, certain components of the three
trust constructs were eliminated, and consequently, the content validity of the knowledge-
based trust construct was narrowed to include addressing the areas of honesty and not
misleading another party. Lewicki and Bunker’s theoretical description of knowledge-
based trust includes predictability, reliance on information, and communication
information exchange. In this dissertation information and communication exchange fell
out of the knowledge-based construct in order to achieve convergent validity. Likewise, a
trade-off between convergent and content validity occurred with the identification-based
trust construct which was defined as involving identification with the other party,
understanding, joint goals, and having a shared collective identity. The identification-
based trust latent variable for this study ended up with three manifest variables involving
openness, fairness, and empathy. As a result, fairness, which is an area thought to
possibly be an element of knowledge-based trust, is actually a part of identification-based
trust. In addition, joint goal setting and shared identity did not load high enough on the
factor and were not included as part of identification-based trust. Table 5-2 below shows
the key elements of the three trust types according to Lewicki and Bunker (1995) and this
study.

Table 5-2 shows that the measurements in this study are consistent content-wise

with the theory posited by Lewicki & Bunker (1995); however, all three types of trust

160

used in this study are more narrowly defined. Calculus-based trust is essentially fear of
punishment, knowledge-based trust involves being honest and forthright, and
identification-based trust involves being empathic, open, and fair. One surprising finding
is that identification-based trust includes the concepts of fairness and openness. The prior
theory by Lewicki and Bunker (1995) suggest that these two concepts would more likely
develop at the knowledge-based stage.

In conclusion, this study establishes the existence of calculus-based trust and
possibly the existence of knowledge and identification-based trust. The possibility that
knowledge and identification-based trust form a single type of trust (character- based

trust) also exists.

5. 2.2 Individual versus Organizational Trust

Proposition 2 looked at whether a difference exists between trust at an
organizational level versus trust at an individual level. The apriori proposition stated that
these two types of trust would not be unique.

However, six of the seven statistical tests analyzed showed a significant difference
between trust at the individual and organizational levels. These findings suggest that
respondents do in fact see a difference between trusting an individual person and trusting
an organization. This result is interesting in that it indicates that while purchasing
executives trust the individual they are working with, this trust does not necessarily
extend to the entire organization. This result supports a strong emphasis on stability in
relationships among buyers and suppliers; however, the results suggest that too much

stability may possibly be negative. In other words, the supplier becomes too comfortable

161

with the relationship manager from the buyer, so that eventually when this manager
leaves the supplier feels anxiety because their trust of the organization is not strong.
Therefore, it may be important that rather than individual relationships between members
of the buyer and supplier that groups or teams from both parties work together which
would mitigates dependence on a small number of relationships and possibly increase
each party’s trust of one another.

Looking at theory, the work of Ring and Van de Ven 1984; Dore 1983;
Granovetter 1985; Eccles 1981; and Lincoln 1985 found that long-term personal
relationships develop standards of behavior at the same level or better than rules created
by a hierarchical relations. Ring and Van de Ven also go on to point out that one of the
major advantages to having high levels of trust is that trust effectively mitigates risk.
While trust may provide many of the benefits noted in this study (avoid use of contracts,
mitigate risk, increase information and resource exchange), a key question that arises is
whether firms are effectively taking advantage of the benefits that trust provides and
translating these benefits into performance improvement. This question was answered by
statistically testing to see if there is a relationship between length of relationship and
performance. Length of relationship and performance were found to have a significant
relationship which supports the Ring and Van de Ven’s, Dore’s, Eccle’s and
Granovetter’s theory findings. Therefore, it is essential that firms have a reason for
creating trust (avoid costly contracts, take more risks, and share resources) because
without a plan trust may be created but not leveraged into results.

Most prior studies on trust have analyzed trust at an individual level (McAllister

1993; Cummings and Bromiley 1995), but in the area of buyer-supplier relationships,

162

trust at the organizational level is also important. The question arises, is trust of an
employee of a firm the same as trust of the organization itself? Intuitively, the answer
would appear to be yes. If a purchasing agent has a good relationship with an employee
at the supplier, the purchasing agent would then likely say he or she has a good
relationship with the organization as well.

Results from this study suggest that trust of an individual does not automatically
translate into trust of an organization. Many possible explanations exist regarding why
trusting an individual would differ from trusting the organization they represent, and

future research analyzing this issue could prove to be interesting.

5. 2.3 Coercive and Non-coercive Power

Hypothesis 1A states there is a positive association between coercive power and
calculus-based trust. Calculus-based trust is trust based on fear of punishment, and
coercive power is use of punishment as influence, so it is not surprising that a strong
significant relationship was found between these two variables.

Hypothesis 1B states that there is a positive association between non-coercive
power and knowledge-based trust. Non-coercive power is the use of rewards, persuasion,
and other non-coercive measures to influence a supplier. Thus the a-priori model posits
that non-coercive power will be present with the stronger more developed forms of trust.
Support was found for Hypothesis lB. Hypothesis 1C states that there is a positive
association between non-coercive power and identification-based trust and support was
found for Hypothesis 1C also. These findings support the stream of research created by

Deutsch in 1960 and extended by Gaski and Nevin through the I980S suggesting that

163

non-coercive and coercive power have different effects on levels of performance. This
study went further in hypothesizing that power impacts trust which is then correlated with
performance. Implications of this finding are signficant. Firms that are impatient and
resort to the use of coercive power will likely see the performance of their key supplier
deteriorate whereas firms using non-coercive power will have a much better chance of

seeing a supplier’s performance improve.

Table 5-3 Hypotheses 2-3

 

 

 

 

 

Propositions/Hypotheses Result Implications
H2. Dependence between buyer and Not Dependence of supplier
supplier will be positively associated with Supported on the buyer may lead to
the level of identification-based trust in a exploitive rather than
supplier. trust building behavior

by buyer.

H3A. A short-term cooperative buyer- Not Measurement of
supplier relationship planning horizon will Supported relationship planning
be associated with the level of knowledge- horizon using proxies
based trust in a supplier. such as the length of a
H3B. A short-term cooperative buyer- Not firm’s average key
supplier relationship planning horizon will Supported supplier relationship
be associated with the level of may be more accurate.
identification-based trust in a supplier.

 

 

As noted on page 44 of chapter two, there appears to be some type of relationship
between Lewicki and Bunker’s three types of trust and Raven’s (1958) three types of
power (coercive, expert and referent). Prior empirical work in the area of power has
looked at coercive and non-coercive power (Gaski & Nevin 1985) in channel
relationships, and Gaski and Nevin’s work was extended in this study to buyer-supplier
relationships. Non-coercive power according to Gaski & Nevin and Hunt & Nevin

(1974) is referent, legitimate, and reward power, but for the purpose of this study, non-

164

coercive power is the non-coercive influence of the buyer on the supplier which could be
due to rewards, identification, status, or other non-coercive sources.

Little research has been done on the relationship between different types of power
and trust. Gaski (1986) found some support for a relationship between fear and
punishment and dissatisfaction in a relationship. The strong positive relationship
between calculus-based trust and coercive power confirms Gaski’s earlier findings. This
study found coercive power to have a strong positive association with calculus-based trust
which in turn had a strong negative correlation with performance. Research on non-
coercive power and trust is scarce so this study is one of the first to analyze this issue.
Both knowledge and identification-based trust were found to have a positive association
with non-coercive power. It should be noted however, that non-coercive power in this
study addressed the issue of supplier influencability and not the specific type of influence
(legitimate, reward, identification, etc.). Therefore, influencability of a supplier is heavily
correlated with trust, but how that influence is attained is a question requiring further

study.

5. 2. 4 Dependence

Hypothesis 2 analyzed the association between dependence and identification-
based trust. Dependence addresses how dependent the supplier is on the buyer. The
relationship between identification-based trust and supplier dependence was not
statistically significant. A possible explanation for the lack of significance is the possible
exploitation by large buyers of the supplier who is dependent on them. As previously

mentioned, the survey respondents were predominantly from very large industrial firms.

165

 

The size of these firms combined with suppliers who feel vulnerable or dependent may
lead to one sided relationships where the supplier does whatever they must to keep the
buyer as a customer.

Dependence analyzes the reliance of the supplier on the buyer and how this
associates with trust. Originally, this study attempted to consider interdependence, but the
measures did not provide significant convergent validity to include buyer dependence
along with supplier dependence as a good measure of interdependence. Previous work
on dependence has looked at the relationship between power and dependence (Emerson
1962; Frazier, Spekman and O’Neal 1988) and the effect of power asymmetries on a
relationship (Williamson 1975; Heide 1994; Lusch & Brown 1996). The measures in
this study directly address how costly, and hence, how dependent the supplier is on the
buyer. The expectation is that a dependent supplier would be more amenable to creating
a good relationship with the buyer which would generate in higher levels of trust.

Dependence was theorized to have a positive association only with identification-
based trust primarily because of the findings of Mohr and Spekman who found a positive
correlation between interdependence and trust in buyer-supplier partnerships. As noted
above, this study tested supplier dependence and not interdependence between buyer and
supplier. Interestingly though, in this study, supplier dependence was not found to have a
significant association with identification-based trust; however, prior theory supports this
finding. Several prior studies (Frazier, Spekman and O’Neal 1987; Williamson 1975;
and Etgar & Valency 1983) suggest that in situations where dependence exists, the more

powerful party will almost always exploit the relationship. It is not surprising then to find

166

that perhaps in this study some firms use a supplier’s dependence to gain more favorable
conditions which could have a negative impact on trust.

Future research may benefit from looking at restraint from the use of power in a
dependence or power asymmetric relationship. Lusch & Brown (1996) and Heide &
Miner (1992) found significant relationships between restraint of the use of power and
various satisfaction and expectation measures of channel relationships. This study would
likely have had a better opportunity of seeing a positive association between restraint
from the use of power and trust rather than dependence and trust. Exercise of power is an
important issue because prior research (Emerson 1962; Buchanan 1992 ) found that in
relationships with power asymmetry a significant difference in relational norms, of which
trust is a part of, exists. Therefore, it would be beneficial to specifically find
relationships where power asymmetry exists and compare relationship performance
between buyers that exploit power asymmetry versus relationships where there is restraint

of power.

5. 2. 5 Relationship Planning Horizon

Hypothesis 3A tested the association between relationship planning horizon and
knowledge-based trust. The association was not statistically significant. Scale items for
the relationship planning horizon construct were created specifically for this study and
combined with prior measures in an attempt to create a latent construct. Convergent
validity was established, but in creating the measurement model, it was difficult to
establish discriminant validity between relationship planning horizon and several of the

other constructs including: dedicated asset specificity, knowledge-based trust and

167

identification-based trust. Problems establishing discriminant validity are not surprising
since the advanced levels of trust and dedicated asset specificity are both constructs that
usually require a long-term perspective or long-term investment.

Because a long-term perspective is part of these constructs, discriminant validity
became a problem and the relationship planning horizon construct was only measured by
one manifest variable. This lone manifest variable used looks more at the issue of
perceived equity in the relationship rather than the planning horizon. Hypothesis 3B
testing the association between relationship planning horizon and identification-based
trust was not found to be statistically significant.

While perceived equity in the relationship was not found to be significantly
related to trust, length of relationship which is perhaps a closer proxy for relationship
time horizon was found to be significantly related to trust. Firms that have the capacity to
have a 15 or 25 year relationship with one supplier shows an ability and emphasis of a
firm on developing long-term relationships which could be a skill that relationship
managers have in or develop in that company.

As noted in chapter II, the theory on relationship planning horizon can be split
into two schools of thought. The first is the economic short-term planning horizon idea
put forth by Williamson 1975, Hirschman 1982, and discussed by Gulati and Nohria
1992. The economic short-term perspective is the opposite of the mutually shared long-
term philosophy put forth by Gulati and Nohria 1992, Fisher and Brown 1989, and others.
This study hypothesized that a long-term orientation would be positively associated with
knowledge-based and identification-based trust. Neither association was found to be

statistically significant.

168

A problem with testing the relationship planning horizon latent variable with trust
was the difficulty in attaining discriminant validity between relationship planning horizon
and trust, and also between relationship planning horizon and asset specificity. This
difficulty is not completely surprising because both trust and asset specificity are each
constructs that would seem to imply a long-time time horizon. For example, asset
specificity is not likely to take place between a buyer and supplier doing business for the
first time. Asset specificity generally occurs later in a relationship and will escalate if
performance and trust increase (Kamath & Liker 1994).

A logical explanation of why discriminant validity was difficult with the
relationship planning horizon construct is the lack of prior research using this construct.
While there are a few studies that have used similar concepts, no established definition or
reliable and valid construct for relationship planning horizon exists. This study was
hoping to establish a reliable and valid construct but was unable to do so. The only way
relationship planning horizon was usable in the structural model was to use one question
which addresses the perceived equity in the relationship. Not surprisingly, this question
on perceived equity was not found to have a significant relationship with either
knowledge-based or identification-based trust.

In conclusion, the results of the analysis on buyer-supplier relationship planning
horizon are misleading. Neither relationship was found to be significant; however,
perceived equity and not relationship planning horizon was tested. In fact, long-term
planning manifest variables were tied so closely to trust that discriminant validity could
not be established. The inability to establish discriminant validity may indicate that

relationship planning horizon is actually a part of the latent variable of knowledge-based

169

 

trust, identification-based trust, or asset specificity. While relationship planning horizon
did not fall out as part of the different types of trust in the confirmatory factor analysis, an
implicit time or planning horizon element is evident when comparing calculus-based trust
versus knowledge and identification-based trust. Knowledge and identification-based
trust imply a relationship that has taken time to develop. The parties know each other
well and can predict how the other party will react to certain situations. Calculus-based
trust is more short term and associated with actions aimed at getting quick results
possibly at the expense of long-term trust development. The prior post hoc analysis
mentioned that length of relationship was found to be significantly associated with trust

which supports the link between planning horizon and trust.

5. 2. 6 Human Asset Specificity

Asset specificity has traditionally been the dedicated equipment or capacity made
by one firm to a relationship. Recent research by Kamath & Liker 1994; Nishiguchi
1994; indicated that asset specificity can be broken into multiple types including
components of co-location, information exchange and dedication of sites. This study
broke down asset specificity into two types, asset specificity behavior and dedicated asset
specificity. Hypotheses 4 looks at the association between asset specificity behavior and
identification-based trust.

Asset specificity behavior builds on the work of Dyer (1994) who looked at the
higher incidence of personnel and knowledge transfer between Japanese automotive firms
in comparison to American automotive firms. This study postulated that a higher

incidence of visitation of key supplier personnel to the buyer’s facilities would be

170

 

Table 5-4 Hypotheses 4-5

 

Propositions/Hypotheses Result ImJilications
H4. Asset Specificity behavior focusing Supported Co-location and
on information Sharing and synergy will information sharing by
be positively associated with the supplier has a
identification-based trust in a supplier positive relation to

identification-based trust.
(Knowledge-based trust
was also found to have a
significant relationship
with asset specificity

 

 

 

 

 

behavior).
H5A. The level of dedicated asset Not Dedicated asset
specificity will be positively associated Supported specificity is a significant
with the level of knowledge-based trust. influence on
HSB. The level of dedicated asset Supported identification-based trust
specificity will be positively associated but not on knowledge-
with the level of Identification-based trust. based trust. Firms use
dedicated assets only at
advanced levels of trust
development.

 

associated with higher levels of identification-based trust. The relationship between asset
specificity behavior and identification-based trust was found to be positive and
statistically significant.

These results strongly support the increase use of co-location and coordination
that many buying firms are using with suppliers. Those who speak against the use ofco-
location state the dangers of sharing important information and ideas with suppliers, but
these fears have yet to influence performance in buyer supplier relationships. In fact,
possibly the opposite happens, and suppliers are able to take information and ideas

learned from a buyer they have co—located with and apply these ideas to their own

171

suppliers. In this way ideas from the original equipment manufacturer are slowly
disseminated throughout the supply base which makes the entire supply base better.

5.2.6.1 Asset specificityI behpvior Asset specificity behavior addresses the more
human and behavioral aspects of asset specificity. These behavioral aspects include co-
location, information sharing, sharing of personnel, and frequency of visits by personnel.
Previous theory on asset specificity and trust has been contradictory. Research by
Kamath & Liker 1994; Nishiguchi 1994, and Dyer 1995 indicate a positive relationship
between asset specificity and trust. However, research by Joshi 1993, Campbell 1992,
and Zaheer & Venkatraman 1995, have found a negative relationship between trust and
asset specificity. In many of the previous studies on asset specificity and trust, the
distinction between different types of asset specificity has not been made which may
explain the conflicting results.

Asset specificity behavior in this study emerged as personnel from the supplier
(primarily non-sales) visiting the buyer’s facilities. These visits by supplier personnel
would facilitate communication and information exchange among the companies, but
primarily this type of asset specificity functions as a way for the supplier to provide better
service to the buyer. Because asset specificity behavior involves co—location which is a
significant commitment on behalf of a supplier, asset specificity behavior was
hypothesized to only have an association with identification-based trust.

Results from this study support the findings of Nishiguchi, Kamath & Liker, and
others stating a positive association between trust and asset specificity behavior exists. A
possible explanation of why this study found a positive association is that when asset

specificity is used in a proactive developmental fashion, trust is likely to develop.

172

 

However, some firms use asset specificity in a punitive fashion, and in these cases, the
level of trust is probably not high because a supplier has been forced to dedicate material

or human resources to a relationship.

5. 2. 7 Dedicated Asset Sjecificity

Hypotheses 5A and 5B analyze the association between dedicated asset specificity
and knowledge and identification-based trust. Dedicated asset specificity was
hypothesized to have a positive association with both types of trust. Previous research
(Kamath & Liker 1994; Nishiguchi 1994) suggest that higher levels of asset specificity
are associated with higher levels of trust and responsibility in suppliers. Neither
Hypotheses 5A and 5B were statistically supported. This result is not surprising because
as noted in chapter two there is little agreement on the relationships between asset
specificity and trust. In fact, research by Zaheer & Venkatraman 1995 and Joshi 1993
found significant results supporting a negative association between dedicated asset
specificity and trust.

The other more traditional type of asset specificity is dedicated asset specificity.
Dedicated asset specificity involves the dedication of assets involved in a relationship
ranging from plant capacity to specific pieces of specialized equipment. Dedicated asset
specificity is similar to asset specificity behavior in that it can be used to help develop
relationships or to punish a poorly performing supplier. Several authors (Borys &
Jemison 1989; Williamson 1983; Gulati & Nohria 1992; Cooper et al. 1990) have noted
the tendency of firms to use dedicated assets as mutual hostages so that parties will be

less likely to be opportunistic. This study defined dedicated asset specificity as

173

specialized equipment, dedicated equipment, or dedicated capacity specifically set aside

 

for the relationship or purchased by the supplier for the relationship.

Identification-based trust was found to have a significant positive association with
dedicated asset specificity; however, knowledge-based trust was just below the threshold
for significance. These results would again support the work of Kamath & Liker and
Nishiguchi who formed a theory making specific distinctions in both dedicated and
behavioral asset specificity that occur over four stages as a buyer-supplier relationship
matures. Therefore, dedication of equipment and capacity would not occur until later
mature stages where identification-based trust exists. This study definitely supports the
use of dedicated asset specificity in a positive way which builds a relationship rather than
as a punishment to poor performing suppliers.

A key issue that arises, however, is whether trust or asset specificity arises first.
Previous research suggests that trust builds the use of asset specificity as well as the use
of asset specificity builds trust. This study specifically found that both knowledge and
identification-based trust are positively associated with human and dedicated asset
specificity. The structural model tested implies that asset specificity causes trust but an
interesting finding emerged from the identification-based trust model. In that model
identification-based trust was found to influence dedicated asset specificity. This finding
is important because it suggests that trust is developed for some reason (mitigate use of
contracts, develop human asset specificity, and develop dedicated asset specificity) and
not merely developed as an end result. A danger in developing trust is that managers in
firms will try to develop a “positive atmosphere” or “good feelings” between the parties.

This response may possibly be in due to previous poor performance or a directive from

174

upper management to develop strategic alliances or partnerships. The problem with this
approach is that trust may be developed, but it is not translated into any actions or
behaviors that will impact performance. The use of trust to develop dedicated asset
specificity would avoid the problem of developing trust without maximizing this action’s

impact on performance.

5. 2.8 Environmental Uncertainty

Prior theory addressing trust and uncertainty is scant. Hypothesis 6 posits a
positive association between identification-based trust and risk. Hypothesis 6 was found
to not be statistically significant. However, it should be noted that uncertainty was
measured with only one manifest variable which addressed the area of technological
uncertainty in the supplier’s industry. Only one manifest variable was used due to
difficulties in establishing discriminant validity between uncertainty and other
antecedents to trust.

Implications of the inability to find discriminant validity indicate that perhaps in a
large database using many different industries that uncertainty is significantly different
across different industries. This study looked at several different kinds of uncertainty
including cost, technology and introduction of many new products. Future studies may
benefit from focusing on a certain type of uncertainty rather than all three types.

Uncertainty involves the uncertainty in the environment of the buyer and
supplier. Problems were encountered in achieving convergent validity among the
questions asked on environmental uncertainty. Environmental uncertainty was

hypothesized to have a positive association with identification-based trust only.

175

 

Table 5-5 Hypotheses 6—7

 

 

Prowitions/Hypotheses Result Implications

H6. Environmental uncertainty will be Supported Technological uncertainty

positively associated with the level of in the buying firm’s

identification-based trust environment is positively
related to identification-
based trust.

H7A. Cooperative buyer-supplier Supported Relationship performance

relationship performance will be is positively associated

positively associated with the level of with both knowledge and

knowledge-based trust in a supplier. identification-based trust.

H7B. Cooperative buyer-supplier Supported

relationship performance will be

positively associated with the level of

identification-based trust in a supplier.

 

 

 

 

 

Uncertainty was not hypothesized to have a association with knowledge-based trust
because prior theory on uncertainty suggests that uncertainty and trust move together
which means we would expect to see higher levels of uncertainty with higher levels of
trust.

Hypothesis H6 was found to be significant in associating the technological
uncertainty of the supplier’s environment with identification-based trust. Even more
interesting is that the relationship was found to be a negative association. This result
suggests that higher levels of trust are associated with lower levels of technological
uncertainty experienced by suppliers. Intuitively, this result is not surprising. A turbulent
risky environment may not allow for longer term key buyer-supplier relationships to
unfold and develop.

A significant problem in interpreting this question is the sample which is an
aggregation of respondents across all levels of technological uncertainty. This

aggregation makes it difficult to see what the true nature of the relationship between

176

I.

environmental uncertainty and identification-based trust is. Future research on
uncertainty and trust would probably benefit from making distinctions between industries.
For example, in the highly volatile computer industry where change is rampant, is
uncertainty closely associated with trust? The answer is likely yes. However, in more
mature industries such as the automotive or chemical industries where channel and supply
chain relationships are more developed, the opposite may be true. In this study, it is
possible that there are low levels of uncertainty in many of the industrial industries
surveyed which would possibly explain the negative association between uncertainty an
trust. In addition, future research on uncertainty addressing the supplier’s technological
environment may benefit from dyadic studies where the supplier as well as the buyer can
provide data on the relationship between uncertainty and trust. Supplier perceptions
could then be combined with buyer perceptions to add face validity to the uncertainty

COI'lStI'UCt.

5. 2. 9 Performance

Hypotheses 7A and 7B hypothesize a positive association between knowledge and
identification-based trust with relationship performance. Both hypotheses 7A and 7B
were found to have strong statistical significance. These findings were slightly surprising
because of the strength of the relationship between trust and performance. Both
knowledge and identification-based trust were found to have estimated structural
coefficients higher than .95 with performance which is exceptionally strong. This means

that a movement of one standard deviation in one variable would result in a movement of

I77

.95 of a standard deviation in the other variable. These results suggest that trust is an
essential element in the performance of a key buyer-supplier relationship.

Not only is trust an essential element but firms need to look closely at where they
invest resources including money, people and equipment. The results from hypotheses
7A and 78 suggest that trust building may be an exceptionally profitable area of
investment for firms. Firms can use extra resources for a variety of reasons ranging from
upgrading information systems to training and hiring of personnel. Results of this study
suggest that trust development with key suppliers is an area that firms should look at
closely Hypotheses 1-6 then provide more specific information on what investment areas
have a significant impact on trust.

An additional interesting finding was the strong negative relationship between
calculus-based trust and performance suggesting that use of fear or punishment has a
significant negative impact on performance. Prior to this study, extensive research on the
relationship between trust and performance already existed as noted in chapter 11.
Research clearly shows an association between trust and performance. Since this study
looked at new theory on trust, performance was included in the analysis to fiirther explore
nomological validity of the new trust theory. What is interesting about the findings of
this study is the strength of the relationships between trust and performance. Both
knowledge-based and identification-based trust were found to have positive associations
above .95 which is extremely strong. The strength of these findings further support the
critical importance of trust in key buyer-supplier relationships. An interesting finding
also in the performance area was the large negative association between calculus-based

trust and performance. This large negative association indicates that the use of fear and

178

 

punishment as noted recently by Kumar 1996 does not produce better performance but in
fact lowers performance.

The major conclusion to draw from the relationship between trust and relationship
performance is that trust deveIOpment is an important and profitable investment for firms.
Managers have a variety of choices in how they can spend budgeted monies but this study
clearly shows that resource investment in trust development is an investment likely to
bring strong returns in performance.

Just as important, but far less expensive of an investment is the finding that fear
and punishment have a strong negative impact on performance. The clear advantage of
this finding is that changing managerial thought processes can be less costly in terms of
resources such as money or equipment. This is good new for executives in firms with
limited budgets. However, while taking fear and threat out of a relationship doesn’t cost
much in resources, it requires something difficult to achieve. What it requires is a
fundamental change in thinking which is not achieved quickly in a firm that has used fear
and threat of punishment as supplier “motivation” in the past. What makes this change
even more difficult is that fear works in the short run. Threats to a supplier will force the
supplier to change; however, these changes have come at a tremendous expense. That
expense is the long-term trust (knowledge and identification-based) discussed in this
study. In fact, it may be more difficult to change a thought process in an organization as
opposed to expending resources to obtain information or equipment.

5.3 Practical Implications
The findings of this study suggest that definitely two and possibly three types of

trust exist in buyer relationships with key suppliers. A calculus-based trust exists which

179

uses punishment and threat to attain compliance from the supplier. Also, a knowledge-
based trust exists where honesty and directness in representing facts exists. Finally, there
is identification-based trust which is built on being open, fair, and empathic. In addition
to establishing these three types of trust, this study confirms the strong association of
trust with relationship performance. Purchasing executives need to look hard at their
relationships with their key suppliers to see if they are developing high levels of
knowledge and identification-based trust. Possibly, just as important is that purchasing

executives avoid developing high levels of calculus-based trust.

Figure 5-1 Typical Trust Development

Typical Trust Development in a

Cooperative Buyer-Supplier Relationship
High

 

 

 

 

 

7
6
5
4
3
2
1
LOW 1 year 2 Years 3 Years
Time

Figure 5-1 is based on actual data supplied by a respondent analyzing data from

all of his key suppliers. What the data shows is the frustration this purchasing executive

180

 

was experiencing in trying to get key supplier relationships to prosper beyond the initial
honeymoon period. Results of this study provide several possible explanations and
solutions to the occurrence in Figure 5-1.

One possible explanation for the drop off in key supplier relationships that occurs
after the relationship is 1-2 years old is that initial levels of knowledge-based trust are not
nurtured into identification-based trust. Knowledge-based trust is primarily the honesty
and integrity of one party, but identification-based trust goes further to where parties can
actually predict and gauge how each other may react to situation. Identification-based
trust may require higher levels of asset specificity behavior and dedication of assets than
knowledge-based trust.

In other words, there is high energy as the buyer-supplier relationship starts as
information and coordination occurs, but after a year of new projects and priorities are
emerging it is easy to slowly put less attention and resources in an cooperative buyer-
supplier relationship causing a decrease in trust.

A second cause for a decrease in trust after a year is possibly the introduction of
calculus-based trust building behaviors such as the use of coercive power to improve
performance. The use of coercive power may be a natural reaction by a buying firm if a
supplier has run into performance problems. As this study found, however, trust and
performance will likely decrease and not increase with the use of coercive power.

This study offers several possible solutions to the problem depicted in Figure 5-1.
The first solution is that firms creating key buyer-supplier relationships realize that there
are different levels of trust. Beyond this realization, firms must take actions that take a

relationship beyond knowledge-based trust to identification-based trust. This movement

181

 

to identification based trust requires taking specific actions outlined below in Table 5-6

 

 

 

below.
Table 545 Key Actions of Trust

TYPE OF TRUST Knowledge-Based Trust Identification-Based Trust

0 Tell the truth 0 Be as open as possible

0 Never exaggerate 0 Ensure perceived fairness
KEY ACTIONS 0 Never misrepresent by partner

0 Have actions back up 0 Be empathetic to the

promises other side’s needs
i

 

 

 

 

Table 5-6 shows that a relationship needs to move beyond the honesty and
integrity level to a relationship built on openness, fairness, and empathy. This point in
time is where many firms can stall as shown by Figure 5-1. Executives will attend
seminars stating the many benefits in creating partnerships and alliances so these
executives extol the virtues of being honest and acting with integrity with suppliers but
they often fail to raise the relationship to the next level (idenfication-based trust). At the
next level, honesty is translated into openness where firms share important information
and personnel which creates stronger trust and better performance. Similarly, not only
must the buyer avoid misrepresentation, but it is important that the supplier feel that they
are treated fairly by the buyer which in turn requires empathy and concern on the part of
the buyer.

The emergence of fairness as a key component of identification-based trust cannot

be underestimated. There has been substantial prior research on fairness in the areas of

182

 

 

procedural and distributive justice (Kumar 1995; Kim and Mauborgne 1991). Procedural
justice is perceived fairness of procedures and processes in a relationship. Distributive
justice address fairness in reception of outcomes or earnings in a relationship. Kumar
found a strong relationship between procedural justice and relationship quality which
included broke character-based trust into two components including honesty and

benevolence. This study supports these prior findings of Kumar by including procedural

 

fairness as a part of identification-based trust. ".

It is also essential that a relationship manager make a distinction between key
suppliers and non-key suppliers. Without this distinction, relationship managers may find
themselves spending their time with “the squeaky wheel” or the supplier with many
problems. Unfortunately, the supplier with many problems may not be a key supplier.
Consequently, there is a huge opportunity cost from opportunities foregone with key
suppliers because the relationship manager has squandered time or cost savings with a
non-key supplier.

Knowledge of what type of relationship a manager is involved with is also
important because it is impossible in most firms to create identification-based trust with
all suppliers. This study, however, says nothing about how non-key suppliers should be
treated. This study does NOT imply that it is correct to use coercive power or calculus-
based trust development with non-key suppliers. This study only addressed key suppliers,
and therefore, only conclusions can only be drawn on buyer relationships with key
suppliers.

This research study also suggests several key antecedents that help to develop the

three types of trust tested in this study. Use of coercive power should be avoided by

183

purchasing executives. Coercive power which uses threats and fear to influence another
party was strongly associated with calculus-based trust. Non-coercive trust, however, is
a strategy that executives can use to increase trust and hence performance in a
relationship. Non-coercive trust uses rewards, persuasion, and other non-punitive or non-
fear provoking measures to influence another party.

Asset specificity can also provide higher levels of trust with key suppliers. Asset
specificity behavior where key non-sales personnel from the supplier spend substantial
time at the buyer’s facilities has potential for developing higher levels of knowledge-
based trust. Co-location of key non-sales personnel has grown substantially over the
years as firms grasp for ways of increasing supplier performance. This study verifies that
co-location, if done correctly can substantially impact trust and consequently
performance.

The use of dedicated assets (equipment or capacity) also has potential to increase
identification-based trust but is less effective at the knowledge-based trust level.
Executives must be careful in using dedicated asset specificity as a strategy. Some
suppliers may see this strategy as manipulative rather than the forging of a relationship. It
is important in the area of asset specificity that both sides demonstrate voluntary
commitment to the relationship.

One of the more interesting results of this study is the finding concerning supplier
dependence and trust. In pre-test interviews with buying firm executives, several
executives mentioned that they only created strong relationships with key suppliers where
the buying firm was a major customer to the supplier. By being a major customer, the

buying firm felt that the supplier was dependent on them, and consequently, it would be

184

easier for the buying firm to exert influence or power over the supplier. This study found
no relationship between supplier dependence and trust which indicates the possibility that
many purchasing executives use supplier dependence to exert coercive power and hence
knowledge and identification-based trust is not built.

The revised model also presents some very interesting implications for

 

relationship managers as shown in Figure 5-2. The most important finding addresses the

issue of whether trust or asset specificity come first in relationship development. The

 

modified model suggests that in key buyer-supplier relationships trust precludes dedicated
asset specificity, but asset specificity behavior is strongly associated with the

development of identification-based trust. The results suggest that it is important to get

Figure 5-2 Modified Identification-Based Trust Model

.459‘

 

28l‘

 

 

 

 

 

C
D nd 259‘ Asset Specificity dentiiEatiomBased '242 Dedicated
epe cnce ——-——> Behavior -——'> Trust (Organizational) —-> Asset Specificity

 

 

 

 

 

 

 

 

 

 

 

 

.186‘

.753‘
24l’

 

Non-coercive
Power

 

 

 

* Indicates significant at p<.0l level

185

personnel and information from the supplier integrated into the buyer’s facilities as soon
as possible. Dedicated asset specificity may require a waiting period until both sides
exhibit behaviors indicating the presence of identification-based trust including openness
a sense of fairness and empathy.

After the demonstration of identification-based trust, a firm should investigate the
use of dedicated assets in a key buyer-supplier relationship. The modified model also
shows the importance that non-coercive power plays in developing identification-based
trust. The model suggests that asset specificity behavior where information and personnel
are coordinated and shared may actually be interpreted by the supplier as a non-coercive
method of influence. Again, since trust is such an important influence on performance

relationship managers may want to direct resources to methods of non-coercive power.

5.4 Limitations

This research study has several limitations. First, the sample was drawn from the
membership of the National Association of Purchasing Management. Thus, the results are
only generalizable to a group to that extent that the group resembles the National
Association of Purchasing Management database.

Second, all of the data is obtained from purchasing executives. Some of the
constructs ask about supplier dependence and other feelings of the supplier toward the
buyer. While prior studies have used perceptions of one party to assess relationship
attributes, this practice still remains problematic. Are the perceptions the buyer has
actually indicative of the supplier’s feelings or situation? Future research needs to

corroborate buyer perceptions with data from the supplier.

186

Third, the sample was not stratified by industry. All of the respondents were from
manufacturing firms; however, there are over 10 different SIC codes represented in the
sample. Within each SIC code, up to 3 different industries can be represented.
Consequently, the results of this study represent analysis of an aggregate of industries.
Future studies breaking a sample into different industries could possibly produce
different results from the ones in this study.

Fourth, several measures for various reasons were represented by only one
manifest variable. The constructs defined by only a single manifest variable lost
considerable content validity. In fact, the constructs were considerably different from

their names.

5.5 Directions For Future Research

First, and most important, a dyadic study needs to be done to follow up on several
of the interesting findings of this study. The key to a dyadic study would be to
investigate if three types of trust exist from the perspective of suppliers as well as buyers.
Also, several of the antecedents such as supplier dependence could be tested with data
from the supplier rather than relying on buyer perceptions. A dyadic study could also
help to clarify the area of asset specificity where dedication of assets to a relationship can
be seen as both a commitment to a relationship or as punishment.

A second direction for future research would be to investigate trust in a time series
study to examine the possible hierarchical relationship among calculus, knowledge and
identification-based trust. A time series study would also allow investigation into the

possible temporal relationship among various antecedents and trust development.

187

 

i

Finally, a study analyzing successful versus unsuccessful key buyer-supplier
relationships would be beneficial. This study would hypothesize that unsuccessful
relationships would see higher levels of threat and fear while successful relationships

would have more openness and honesty.

 

 

188

 

 

APPENDIX A

189

Please take time to complete this questionnaire by circling the appropriate answer in either pen or pencil.

o All responses to this questionnaire will be held strictly confidential. Only aggregate results will be
reported.

0 If you are unsure of the answer to a question, please provide your best estimate. Incomplete questions
create serious problems in data analysis.

The purpose of this study is to gain an understanding of your experiences with a key supplier.
Questionnaire

Please direct your thoughts to a key supplier that provides a high expenditure (total dollar amount) and
strategic service or part. Please respond to each statement by circling the number which best reflects the
extent of your agreement or disagreement with the statement.

Please answer the following questions with this key supplier in mind.

Strongly Strongly
Agree Disagree

ll|2[3l4[5|6 7]

 

 

1. We occasionally threaten this key supplier to motivate them. 1 2 3 4 5 6 7
2. We will take quick punitive measures (e.g. reduce volumes) 1 2 3 4 5 6 7
if this key supplier does not meet our performance expectations.
3. We occasionally use fear to motivate this key supplier. l 2 3 4 5 6 7
4. Our key supplier feels that we identify strongly with their needs. 1 2 3 4 5 6 7
5. Our key supplier feels that we have a strong sense of oneness l 2 3 4 5 6 7
with them.
6. Our key supplier feels that we have become closely associated l 2 3 4 5 6 7
with them.
7. We have few problems influencing this key supplier. l 2 3 4 5 6 7
8. This key supplier is highly receptive to our ideas. I 2 3 4 5 6 7
9. This key supplier respects our industry knowledge. I 2 3 4 5 6 7
10. Our key supplier recognizes us as an expert in the industry. 1 2 3 4 5 6 7
l 1. This key supplier is not receptive to our suggestions. 1 2 3 4 5 6 7
12. We are dependent on this key supplier. l 2 3 4 5 6 7
13. This key supplier would be difficult to replace. 1 2 3 4 5 6 7
14. This key supplier would be costly to lose. 1 2 3 4 5 6 7
15. This key supplier is dependent on us. 1 2 3 4 5 6 7
16. This key supplier would find it difficult to replace us. 1 2 3 4 5 6 7
17. This key supplier would find it costly to lose us. I 2 3 4 5 6 7
18. We expect our key supplier relationships to last more than 1 2 3 4 5 6 7
three years.
19. We reap unexpected benefits from our key suppliers. 2 3 4 S 6 7
20. Once we have gotten what we need from a key supplier, l 2 3 4 S 6 7
we end the relationship and then move on.
21. We rarely have to motivate or reward this key supplier. l 2 3 4 5 6 7
22. This key supplier regularly comes up with performance 1 2 3 4 5 6 7
enhancing ideas on their own.
23. How many years has your firm purchased products or services from this key supplier.
Years

190

 

 

24.

25.

26.

27.

28.

29.
30.

31.
32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.
49.

50.
51.

Strongly

Agree

111213l4lslsl71

If a key supplier takes advantage of us, we immediately
terminate the relationship.

If our key supplier is gaining more from the relationship than
we are, we always take steps to end or alter the agreement.
In some of our relationships, our key suppliers benefit more
than we do, but it is still in our best interest to maintain the
relationship.

We provide resources and training to key suppliers if their
performance slips.

This key supplier is quite willing to make a long-term
investment in helping us.

This key supplier sees our relationship as long-term.

This key supplier is willing to dedicate whatever people and
resources necessary to make us a satisfied customer.

This key supplier is loyal to our relationship.

Key supplier relationships are a competition between 15 and
our key supplier to see who can learn the most the fastest.
Our primag goal in a relationship is to learn as much

from our key supplier as possible.

We have several controls in place to limit the

information our key supplier can get from us.

This key supplier will take information unrelated to the
relationship if we do not monitor them.

We share our weekly, daily, or hourly production schedules
with this key supplier.

Non-sales personnel from our key supplier have visited

our plant to see how their product is used in our process.
Personnel from our key supplier work closely with our
engineering and design area in developing new products.
We share a great deal of sensitive information with this key
supplier.

Our manufacturing personnel regularly visit our key
supplier’s facilities.

In this relationship, it is expected that any information which
might help the other party will be provided.

The parties are expected to keep each other informed about
events or changes that may affect the other party.

Our key supplier has dedicated equipment or capacity
specifically to maintain our relationship.

Our key supplier has dedicated personnel specifically to
maintain our relationship.

Our key supplier has purchased specialized equipment to
meet our needs for this key-input material or service.
Switching from our key supplier to another supplier

would be very costly.

Technological change is rapid in the industry this key
supplier works in.

We have taken a high degree of risk with this key supplier.
Pricing changes occur often in the area that this key
supplier works in.

Key suppliers should not have to be rewarded to perform well.

We reward key suppliers exhibiting excellent performance.

191

l

l

2

2

N

N

3

3

4

4

Strongly
Disagree

 

 

52.
53.
54.
55.
56.
57.
58.
59.
60.

61.
62.

63.
64.
65.
66.
67.
68.
69.
70.

71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.

83.
84 .

85.

Strongly Strongly

 

 

Agree Disagree
Ulzl3l4lslém

New products are introduced often in the industry this 1 2 3 4 5 6 7
key supplier works in.
Complete honesty does not pay when dealing with this key 1 2 3 4 5 6 7
supplier.
Sometimes, this key supplier alters the facts slightly in order 1 2 3 4 5 6 7
to get what they need.
This key supplier has sometimes promised to do things 1 2 3 4 5 6 7
without actually doing them later.
This key supplier seems to feel that it is OK to do anything 1 2 3 4 5 6 7
within their means that will help further their firm’s interest.
We feel that this key supplier takes advantage of us. 1 2 3 4 5 6 7
We feel that this key supplier tries to get the upper hand. I 2 3 4 5 6 7
We monitor this key supplier closely, so that they cannot take 1 2 3 4 5 6 7
advantage of us.
1 think that this key supplier takes advantage of our 1 2 3 4 5 6 7
vulnerabilities.
This key supplier is predictable. l 2 3 4 5 6 7
Given this key supplier’s track record, 1 see no reason to 1 2 3 4 5 6 7

doubt their competence and preparation for the job.

In my opinion, this key supplier is reliable. I 2 3 4 5 6 7
I feel that this key supplier negotiates honestly. I 2 3 4 5 6 7
We have frequent face to face contact with this key supplier. 1 2 3 4 5 6 7
I feel that this key supplier will keep its word. 1 2 3 4 5 6 7
1 think that this key supplier does not mislead us. 1 2 3 4 5 6 7
We have frequent phone conversations with this key supplier. 1 2 3 4 5 6 7
We regularly share information with this key supplier. 1 2 3 4 5 6 7
This key supplier carries out their duties even if we do not 1 2 3 4 5 6 7
check up on them.

We feel confident that this key supplier won’t take advantage 1 2 3 4 5 6 7

of us.
We think the employees with this key supplier tell the truth in l 2 3 4 5 6 7
negotiations.

I feel that our key supplier does not try to get out of 1 2 3 4 5 6 7
commitments.

We pass on new information that may be useful to this 1 2 3 4 5 6 7
key supplier.

We jointly establish goals for our relationship with this 1 2 3 4 5 6 7
key supplier.

This key supplier views the world the same way as we do. 1 2 3 4 5 6 7
This key supplier has a strong sense of loyalty to us. 1 2 3 4 5 6 7

We think this key supplier is open in describing their strengths 1 2 3 4 5 6 7
and weaknesses in negotiating joint projects.

We know this key supplier will consider our concerns when l 2 3 4 5 6 7
making a decision.

I feel that our key supplier negotiates joint expectations fairly. 1 2 3 4 5 6 7
We confront issues effectively with this key supplier. l 2 3 4 5 6 7
Our key supplier relationship is characterized by having 1 2 3 4 5 6 7
short lead-times.

This key supplier has an outstanding on-time delivery record. I 2 3 4 5 6 7
This key supplier has the ability to modify their products 1 2 3 4 5 6 7

to meet our requirements.
In this relationship, the parties work together to solve problems. 1 2 3 4 5 6 7

192

 

 

 

86.
87.
88.

89.

90.

91.

92.

93.

94.

95.

Strongly Strongly

 

 

Agree Disagree
III21314|516IU

This relationship has met our profitability projections thus far. I 2 3 4 5 6 7

We are disappointed in the performance of this relationship. I 2 3 4 5 6 7

This key supplier is genuinely concerned about our company’s 1 2 3 4 5 6 7
welfare.

Rewards have little impact on our key supplier’s behavior. 1 2 3 4 5 6 7

How would you characterize your firm’s ROI over the past two years in comparison to your

industry’s average. Below Averagel 2 3 4 5 6 7 Above Average

How would you characterize your firm’s ROA over the past two years in comparison to your
industry’s average. Below Average 1 2 3 4 5 6 7 Above Average

How would you characterize your firm’s profits over the past two years in comparison to your
industry’s average. Below Average 1 2 3 4 5 6 7 Above Average

a) Please circle the answer that best describes your position in your organization:
1. VP. of Purchasing

2. VP. of Materials Management

3. Director of Purchasing

4. Purchasing Manager

5. Other:

 

b) Number of years with this organization

What are your company’s annual gross sales dollars? (circle your answer)

1. Less than $5 million 4. $50 million to $100 million 7. Over $1 billion
2. $5 million to $10 million 5. $100 million to $500 million
3. $10 million to $50 million 6. $ 500 million to $1 billion

Please estimate the total number of person days of face to face contact between key supplier
salesperson and engineers with your purchasing personnel and engineers over the past year.
Person Days

96. Please estimate the average number of person days co-Iocated (or guest engineers or technical support

personnel) your key supplier have had at your firm over the past year.

Person Days

97. What is the intended duration of this relationship in years. Years

98. Please indicate the percentage improvements realized by your firm in the following areas attributable

to your firrn’s relationship with this key supplier.

 

 

 

 

 

Cycle-time Reduction (from order placement to receipt inclusive) %
Shared Price Reduction (cost savings due to improvement) %
Quality Improvement (reduction in PPM, warranties) %
On-time Delivery (ability to deliver within buyer-specified delivery window) %
Access to new technologies %
New product development time (from concept to production) %

 

193

 

 

Who (name and title) is the primary contact at this key supplier or the person you consider to be the key
supplier’s main representative.

Name: Title:

 

 

Note: Questions 1-13 below refer to the primary contact indicated above.

 

 

 

 

Strongly Strongly
Agree Disagree
111213 I415I6I7J

I. We have a sharing relationship, where my primary contact 1 2 3 4 5 6 7
and I can both freely share ideas, concepts, or information.

2. I can talk freely to this person about difficulties we are 1 2 3 4 5 6 7
having and know that he/she will listen.

3. If I shared my problems with this person, I know he/she l 2 3 4 5 6 7
would respond constructively and caringly.

4. 1 would say the primary contact and I have both made a l 2 3 4 5 6 7
considerable emotional investment in this relationship.

5. I interact with this primary contact informally as well as 1 2 3 4 5 6 7
formally.

6. This primary contact acts with integrity. 1 2 3 4 5 6 7

7. I regularly communicate with this primary contact. I 2 3 4 5 6 7

8. My primary contact regularly communicates with us. 1 2 3 4 5 6 7

9. Other people in our firm who have interacted with this 1 2 3 4 5 6 7
primary contact consider him/her trustworthy.

10. The quality of work obtained from this primary contact is 1 2 3 4 5 6 7
only maintained by diligent monitoring.

11. My primary contact regularly offers suggestions to enhance I 2 3 4 5 6 7
the performance of our relationship.

12. I have faith in my contact person to look out for my best 1 2 3 4 5 6 7
interest even when it is costly to do so.

13. I would feel a sense of betrayal if my primary contact 1 2 3 4 S 6 7

person’s performance was below my expectations.

What other factors do you believe are important in developing a trusting relationship with a key
supplier?

 

 

 

 

Thank you for your time. Again, if you would like a copy of the
executive summary for this study, please attach a business card to this
questionnaire when you return it.

194

 

APPENDIX B

195

March 15, 1998

Name
Position
Street
City

An essential foundation for successful buyer-supplier relationship is a strong sense of
commitment between the two firms. However, little is known about what are the key
factors that lead to strong buyer-supplier relationships as well as what can be done to
maintain them. The National Association of Purchasing Management (NAPM)
and the Department of Marketing and Supply Chain Management at Michigan State
University are supporting this project on relationship development between buyers
and suppliers. Your input is critically important in supporting this research.

You are one of a small group of executives selected from the NAPM membership list
to provide your input on these matters. The questionnaire included is concise and
should take roughly 15 minutes to fill out. Also, a return envelope with postage has
been provided. Please be assured that your responses are completely confidential.
The questionnaire has been tested and respondents found it informative and easy to
complete.

Ifyou would like a copy of the executive summary of this study, please include a
copy of your business card in the enclosed envelope. Ifyou have any questions,
please call Christian Bechtel at (517) 353-6381 ext. 271.

Thank you for your assistance.

Sincerely,
Christian Bechtel Robert B. Handfield, PhD.
PhD. Candidate Associate Professor
Marketing and Supply Chain Management Marketing and Supply Chain
Michigan State University Management
Michigan State University
Dissertation Chair

196

 

 

 

 

 

APPENDIX C

197

April 10, 1998

Name
Position
Company
Address

I recently mailed you a questionnaire as part of my dissertation project, which I
believe will provide insight into ways companies are developing trust to improve
performance in their key suppliers. If you have already returned the questionnaire,
thank you for your participation. Ifyou have not yet had a chance to complete the
survey, could you please take a few minutes and fill out the enclosed survey now? A
completed survey from your company is invaluable to the success of this study.

Michigan State University has one of only two Ph. D. programs in the country that
specifically addresses the concerns of the purchasing firnction. In addition, this
research is supported by the National Association of Purchasing Management
(N APM) which also provided the membership list form which your name was
selected.

Your responses are completely confidential and if you would like to receive a copy
of the executive summary of results simply include a copy of your business card in
the enclosed stamped envelope.

If you have any concerns, please call Christian Bechtel at (517) 353-6381 ext. 271.

Christian Bechtel

PhD. Candidate

Marketing & Supply Chain Management
Michigan State University

198

 

BIBLIOGRAPHY

199

Bibliography

Achrol, Ravi Singh; Reve, Torger; and Stern, Louis W., “The Environment of Marketing
channel Dyads: A Framework for Comparative Analysis,” Journal of Marketing,
Vol. 47, pp. 55-67, 1983.

Aguayo Rafael, Dr. Deming: “The American who Taught the Japanese about Quality,”
Simon & Schuster, New York, 1991.

Akerlof, G. A., “The Market for ‘Lemons’: Quality Uncertainty and the Market
Mechanism, Quarterly Journal of Economics, Vol. 84, pp. 488-500, 1970.

Anderson, James C. and Narus, James A., “ A Model of Distributor Firm an (1
Manufacturer Firms Working Partnerships,” Journal of Marketing, Vol. 54 1990,
45-58, 1990.

Anderson, James C. and Narus, James A., “Partnering as a Focused Market Strategy,”
California Management Review, pp. 95-113, Spring, 1991.

Anderson, Erin and Barton Weitz, “Determinants of Continuity in Conventional
Industrial Channel Dyads,” Marketing Science, Vol 8, No. 4, pp. 331-323, 1989.

Anderson, Erin, and Weitz, Barton, “ The Use of Pledges to Build and Sustain
Commitment in Distribution Channel Dyads,” Journal of Marketing Research,

Vol. 29, pp. 18-34, 1992.

Armstrong, J. Scott, and Terry 8. Overton. “Estimating Nonresponse Bias in Mail
Surveys.” Journal of Marketing Research 14 (August 1977): 396-402.

Arrow, K., The Limits of Organization, Norton, New York, 1974.
Axelrod, R. The Evolution of Cooperation, Basic Books, New York, 1984.
Bagozzi, Richard P., Alice M. Tybout, Samuel C. Craig, “The Construct Validity of the

Tripartite Classification of Attitudes.” Journal of Marketing Research, Vol. 16, 1,
(February 1979): 88-95.

200

 

 

Bagozzi, R. P. and Y. Yi, “On the evaluation of structural equation models.” Journal of
Academy of Marketing Science, 16 1, 1988, 74-94.

Barber, Bernard, The Logic and Limits of T rust, Rutgers University Press, New
Brunswick, NJ. 1983.

Barnes, Louis, B., “Managing the Paradox of Organizational Trust,” Harvard Business
Review, March/April, 1981.

Barney, Jay, B. and Hansen, Mark, H. “Trustworthiness as a Source of Competitive
Advantage,” Strategic Management Journal, Vol. 15, pp. 175-190, 1994.

Bateson, Patrick, “The Biological Evolution of Cooperation and Trust,” In Gambetta,
Diego, (Eds) Trust Basil Blackwell, Inc., New, York, NY, pp. 14-30, 1988.

Bentler, P. and D. Bonnett “Significance tests and goodness-of-fit in the analysis of
covariance structures.” Psychological Bulletin 88 (1980): 588-600.

Birou, Laura, The Role of the Buyer-Supplier Linkage in an Integrated Product
Development Environment, Unpublished dissertation, E. Lansing, Michigan,
Michigan State University, December, 1993.

Bollen, K. Structural Equations with Latent Variables. Wiley and Sons, New York 1989.
Borys, Bryan and Jemison, David B., “Hybrid Arrangements as Strategic Alliances:
Theoretical Issues in Organizational Combinations,” Academy of Management

Review, vol. 14, no. 2, pp. 234-249, 1989.

Boss, R. W. “Trust and managerial problem solving revisited,” Groups and Organization
Studies, pp. 331-343, 1978.

Boyle, R., and Bonaich, P., “The development of trust and mistrust in mixed-motive
games,” Sociometry, 23, pp. 123-139, 1970.

Bradach, J. L., & Eccles, R. G., “Price Authority and trust: from ideal types of plural
forms,” Annual Review of Sociology, 15, pp. 97-1 18, 1989.

Brewer, John, and Hunter, R. Multimethod Research: A Synthesis of Styles, Sage
Publications, Newbury Park, CA. 1989.

Britton and Stallings, Leadership is Empowering People, University Press of America,
Lanham, MD, 1986.

201

Bromily, Philip, and Cummings L. L., “The Organizational Trust Inventory (OTI):
Development and Validation, in Kramer, Roderick, M., and Tyler, Tom, R., (Eds)
Trust in Organizations: Frontiers of Theory and Research, Sage Publications,
Thousand Oaks, California, 1995, pp. 302-330.

Butler, J. K., Jr. “Toward understanding and measuring conditions of trust: Evolution of a
conditions of trust inventory,” Journal of Management, 17, pp. 643-663, 1983.

Butler, John, K. “Toward Understanding and Measuring Conditions of Trust: Evolution
of a Conditions of Trust Inventory,” Journal of Management, vol. 17, pp. 643-663,
1991.

Butler, J. K. and Cantrell, R. S., “A Behavioral Decision Theory Approach to Modeling

Dyadic Trust in Superiors and Subordinates,” Psychological Reports, Vol. 55, pp.
19-28, 1984 .

Byme, Barbara M. Structural Equation Modeling with EQS and EQS/Windows: Bsic
Concepts, Applications, and Programming. Thousand Oaks: Sage Publications,
1994.

Cadotte, Ernest R. and Louis Stern, “ A Process Model of Interorganizational Relations in
Marketing Channels,” Research in Marketing, Vol. 2, pp. 127-158, 1979.

Campbell, Alexandra J ., The Antecedents and Outcomes of Cooperative Behaviour in
International Supply Markets, Unpublished dissertation, Toronto, Canada,
University of Toronto, September, 1994.

Churchill Jr. Gilbert A., “A Paradigm for Developing Better Measures of Marketing
Constructs.” Journal of Marketing Research, 26, 1 (February 1979): 64-73.

Cook, Thomas D. and Campbell, Donald T., Quasi-Experimentation: Design and
Analysis Issues for Field Settings, Rand McNally College Publishing Company,
Chicago, 1979.

Cooper, R. W., Dejong, D. W. Forsythe, R. and Ross, T. W., “Selection Criteria in
Coordination Games: Some Experimental Results,” American Economic Review,
pp. 218-233, (1990).

Cummings, L. L. & Bromily, Philip, “The Organizational Trust Inventory, (OTI):
Development and Validation,” in Kramer, Roderick, M., and Tyler, Tom, R.,
(Eds.)T rust in Organizations: Frontiers of Theory and Research, Sage
Publications, Thousand Oaks, California, 1995, pp. 302-323.

Dasgupta, Partha, “Trust as a Commodity,” In Gambetta, Diego, (Eds.) Trust, Basil
Blackwell, Inc., New, York, pp. 49-72, 1988.

202

 

 

 

Deutsch, M. , “Trust and suspicion,” Journal of Conflict Resolution, vol. 2, 265-279,
1958.

DeVellis, Robert R. Scale Development: Theory and Application. Newbury Park, CA.:
Sage Publications, 1991.

Dillman, Don A. Mail and Telephone Surveys: The Total Design Method. New York:
John Wiley, 1978.

Dixit, Avinash, K., and Nalebuff, Barry, J., Thinking Strategically: The Competitive Edge
in Business, Politics, and Everyday Life, W. W. Norton & Company, New York,
1991.

Dore, R. Goodwill and the Spirit of Market Capitalism, British Journal of Sociology, Vol.
34, pp. 459-482 (1983).

Dunn, John, “Trust and Political Agency, ”In Gambetta, Diego, (Eds) Trust, Basil
Blackwell, Inc. New York, pp. 73-93, 1988.

Dwyer, F. R., Schurr, P. H., & Oh, 8., “Developing buyer-seller relationship,” Journal of
Marketing, vol. 51, pp. 11-27, 1987.

Dyer, Jeffrey H., Four Papers on Governance, Asset Specialization, and Performance: A
Comparative Study of Supplier-A utomaker Relationships in the US. and Japan,
Unpublished dissertation, Los Angeles, California, University of California, Los
Angeles, February, 1994.

Eisenhardt, Kathleen, M., “Building Theories from Case Studies,” Academy of
Management Review, Vol. 14, No. 4, pp. 118-126, 1989.

Ellram, Lisa, M., “A Managerial Guideline For the Development of Implementation of
Purchasing Partnerships,” International Journal of Purchasing and Materials

Management, pp. 2-8, (1991).

Emerson, Richard, “Power-Dependence Relations,” American Sociological Review, Vol.
2, pp. 31-41, February, 1962.

Etgar, Michael, and Valency Aharon, Journal of Retailing, Vol. 59, Iss 4, PP. 81-92,
Winter 1983.

Fairholm, Gilbert, W., Leadership and the Culture of Trust, Praeger Publishers, Westport,
Connecticut, 1994.

Farris, G., Senner, E., and Butterfield, D. “Trust, Culture and Organizational Behavior,”
Industrial Relations, vol. 12, pp. 144-157, 1973.

203

 

 

Fisher, Roger, and Brown, Scott, Getting Together, Penguin Books, New York, 1988.

Fowler, F loys J. Jr., and Mangione, Thomas W., Standardized Survey Interviewing:
Minimizing Interviewer-Related Error, Sage Publications, New York, 1990.

Frazier, Gary, Interorganizational Echange Behavior in Marketing Channels: A
Broadened Perspective,” Journal of Marketing, pp. 68-78, Fall 1983.

Frazier, Gary C., Spekman, Robert E. and O’neal Charles R. “Just-in-Time Echange
Relationships in Industrial Markets,” Journal of Marketing, Vol. 52, Iss. 4, pp. 52-
67, Oct 1988.

French, John and Raven Bertram, “The Bases of Social Power,” Studies in Power,
Dorwin Cartwright, ed, Ann Arbor, University of Michigan Press, pp. 150-167,
1959.

Friedland, N. “Attribution of control as a determinant of cooperation in echange
interactions,” Journal of Applied Psychology, Vol. 20 pp. 303-320, 1990.

Gabarro, John, J,. “The Development of Trust, Influence, and Expectations,” In Athos,
Anthony and Gabarro, John, J ., (Eds), Interpersonal Behavior Communication and
Understanding in Relationships, Prentice-Hall, Inc., Englewood Cliffs, New
Jersey, pp. 290-303, 1978.

Gambetta, Diego, “Mafia: the Price of Distrust,” In Gambetta, Diego, (Eds) Trust, Basil
Blackwell, Inc. New, York, pp. 158-175, 1988.

Gambetta, Diego, “Can We Trust Trust?,” In Gambetta, Diego, (Eds) Trust, Basil
Blackwell, Inc. New, York, pp. 213-238, 1988.

Ganesan, Shankar, “Determinants of Long-terrn Orientation in Buyer-Seller
Relationships,” Journal of Marketing, Vol. 58 pp. 1-19, April 1994.

Garrity, Mary A., Measuring Trust Within Organizational Culture, Unpublished
dissertation, Hattiesburg, Mississippi, The University of Southern Mississippi,
1991.

Gaski, John R., The Theory of Power and Conflict in Channels of Distribution.” Journal
of Marketing, 48, 3, (Summer 1984): 9-29.

Gaski, Joh F., and John R. Nevin, The Differential Effects of Exercised and Unexercised

Power Sources in a Marketing Channel.” Journal of Marketing Research, 22, 2,
(May 1985): 130-142.

204

Ghoshal, Sumantra, and Bartlett Christopher, A., “Linking Organizational Context and
Managerial Action: The Dimensions of Quality Management, Strategic
Management Journal, Vol. 15, pp. 1091-1120.

Good, David, “Individuals, Interpersonal Relations, and Trust,” In Gambetta, Diego,
(Eds) Trust, Basil Blackwell, Inc., New, York, pp. 31-48, 1988.

Granovetter, M. “Economic Action and Social Structure: A Theory of Embeddedness,”
American Journal of Sociology, vol. 91, pp. 481-510, 1985.

Granovetter, M. Problems of Explanation in Economic Sociology, in N. Nohria & R.
Eccles (Eds) Networks and Organizations: Structure Form and Action, pp. 25-56,
Boston: Harvard Business School Press, 1992.

 

 

Gulati, Ranjay, “Does Familiarity Breed Trust? The Implications of Repeated Ties For
Contractual Choice in Alliances,” Academy of Management Journal, vol. 38, no. 1,
pp. 85-112, 1995.

Gulati, Ranjay, and Nohria, Nitin, “Mutually Assured Alliances,” Academy of
Management Proceedings, pp. 17-21, 1992.

Hahn, Chan. K., Kyoo J. Kim and John S. Kim, “Costs of Competition: Implications for
Purchasing Strategy,” Journal of Purchasing and Materials Management, Vol. 22,
Iss. 4, pp. 2-7, Fall 1986.

Hair, Jr., Joseph F., Rolphy B. Anderson, Ronald L. Tatham, and William C. Black.
Multivariate Data Analysis with Readings. 3d ed., New York: Macmillan
Publishing Company, 1992.

Hakansson, Hakan, Marketing and Purchasing of Industrial Goods: An Interaction
Approach,” Wiley, New York, 1982.

Hall, Mark C., An Empirical Investigation of the Relationship Between Buyer-Seller
Interactions, Buying Situations, and Perceive Risk in Organizational Purchasing,
Unpublished dissertation, Fayettville, Arkansas, University of Arkansas, May,
1991.

Hamel, Gary, Doz, Yves L., and Prahalad, C. K., “Collaborate with Your Competitors-
and Win,” Harvard Business Review, vol. 67, no. 1, pp. 133-139, 1989.

Handfield, Robert, “A Resource Dependence Perspective of JIT Purchasing,” Journal of
Operations Management, Vol. 11, Iss. 3, pp. 289-311, Spring 1993.

Handfield, R., Bechtel, C. and Stimpert, J. “An Empirical Investigation into Supplier
Alliance Communication,” Working Paper, Michigan State University, 1996.

205

 

Hart, Kerry M., Capps, H. R., Cangemi, Joseph, P, and Caillouet, Larry, M., “Exploring
Organizational Trust and Its Multiple Dimensions,” Organization Development
Journal, vol. 4, No. 2, pp. 31-39, 1989.

Harvey, Jerry, B. “Trust and Organizational Effectiveness,” Technology Review, pp. 271-
277, May/June 1989.

Heide, Jan B., “Organizational Governance in Marketing Channels.” Journal of
Marketing, 58, 1, (January 1994): 71-85.

Heide, Jan B. and John, George, “The Role of Dependence - Balancing in Safeguarding
Transaction Specific Assets in Convention Channels, Journal of Marketing, Vol
52, pp. 20-35, (1988).

Heide, Jan B. and John, George, “Alliances in Industrial Purchasing: The Determinants of
Join Action in Buyer-Supplier Relationships,” Journal of Marketing Research, 27,
pp. 24-36, 1990.

Hirsch, Social Limits to Growth, Cambridge, MA. Harvard University Press, 1975.

Hirschman, Albert, 0., Shifting Involvement: Private Interest and Public Action,
Princeton, NJ. Princeton University Press, 1982.

Holden, Reed K., An Exploratory Study of Trust in Buyer-Seller Relationships,
Unpublished dissertation, Boston, Massachusetts, Boston University, September,
1 990.

Holmstrom, B., “Moral Hazard and Obvservability,” Bell Journal of Economics, Vol. 10,
pp. 74-91.

Homans, George, Casper, Sentiments and Activities, Transaction, New Brunswick, NJ.
1961.

Hosmer, LaRue, Tone, “Trust: The Connecting Link Between Organizational Theory and
Philosophical Ethics,” The Academy of Management Review, vol. 20, no. 2, pp.
379-403, April, 1995.

Hovland, E. I., Janis, I. L., and Kelley, H. H., Communication and Persuasion, New
Haven, CT. Yale University Press, (1953).

Huber, G, “Temporal Stability and Response-order Biases in Participant Descriptions of

Organizational Decisions,” Academy of Management Journal, vol. 28, pp. 943-
950, 1985.

206

Huber, George P. and Power, Daniel F ., “Retrospective Reports of Strategic-Level
Managers: Guidelines for Increasing Their Accuracy,” Strategic Management
Journal, vol. 6, no. 2, pp. 171-180, 1985.

Hunt, Shelby D. and John R. Nevin, “Power in Channel of Distribution: Sources and
Consequences,” Journal of Marketing Research, 11 (May 1974): 186-193.

Johnson-George, C, & Sway, W. C. “Measurement of specific interpersonal trust:
Construction and validation of a scale to assess trust in a specific other, Journal of
Personality and Social Psychology, vol. 43, pp. 1306-1317, 1982.

Joshi, Ashwin, W., Eflects of Dependence and Relational Norms on Commitment and
Opportunism in Buyer-Seller Relationships, Kingston, Canada, 1993.

Kamath, R. and Liker, Jeffrey, “A Second Look at Japanese Product Development,”
Harvard Business Review, vol. 72, pp. 154-170, 1994.

Kerlinger, Fred N., Foundations of Behavioral Research, third edition, Holt, Rinehart,
and Winston, Inc., Chicago, 1986.

Klein B. , R. A. Crawford, and A. A. Alchian “Vertical Integration, appropriable rents,

and the competitive contracting process,” Journal of Law and Economics, Vol. 21
pp. 297-326, 1978.

Klimoski, Richard, J ., and Karol, Barbara, L., “The impact of Trust on Creative Problem
Solving Groups,” Journal of Applied Psychology, vol. 61, no. 5, pp. 630-633,
1976.

Kogut, Bruce, “On Designing Contracts to Guarantee Enforceablity; theory and evidence
from East- West Trade,” Journal of International Business Studies, Vol. 17, Iss. 1
pp. 47-61, Spring 1986.

Kreps, D. M., Corporate Culture and Economic Theory, In J. Alt and K Shepske (Eds),
Perspectives on Positive Political Economy, Cambridge University Press, New
York, pp. 1190-1243, 1990.

Kumar, Nirmalya, The Power of Trust in Manufacturer-Retailer Relationships, Harvard
Business Review, November-December 1996, pp. 92-110.

Lamming, Richard,_Beyond Partnership: Strategies For Innovation and Lean Supply,
Prentice Hall, New York, 1993.

Landeros, Robert, An Empirical Study of Buyer/Seller Relationships in US.

Manufacturing Firms, unpublished dissertation, East Lansing, Michigan: Michigan
State University, 1988.

207

 

 

 

Larzelere, R. E., & Huston, T. L., “The Dyadic Trust Scale: Toward understanding

interpersonal trust in close relationships,” Journal of Marriage and the Family,
vol. 42, pp. 595-604, 1980.

Lascelles, D. M. and Dale, B G. “The Buyer - Supplier Relationship in Total Quality
Management,” Journal of Purchasing and Materials Management, Vol. 25, Iss. 2,
pp. 10-19, Summer 1989.

Lewicki, Roy, J ., and Bunker, Barbara, B., “Developing and Maintaining Trust in Work
Relationships,” in Kramer, Roderick, M., and Tyler, Tom, R., (Eds) Trust in
Organizations: Frontiers of Theory and Research, Sage Publications, Thousand

Oaks, California, 1995, pp. 114-139.
Lewis, Jordan, D., Partnerships For Profit, The Free Press, New York, 1990.
Lewis, Jordan, D., The Connected Corporation, The Free Press, New York, 1996.

Lewis, J. D., & Weigert, A. J ., “Social atomist, holism, and trust,” The Sociological
Quarterly, vol. 26, pp. 455-471, 1985a.

Lewis, J. D., & Weigert, A. J ., “Trust as a social reality,” Social Forces, vol. 63, pp. 967-
985, 1985b.

Lieberman, J. K., The Litigous Society, New York, Basic Books, 1981.

Lorenz, Edward, H., “Neither Friends nor Strangers: Informal Networks of
Subcontracting in French Industry,” In Gambetta, Diego, (Eds) Trust, Basil
Blackwell, Inc., New, York, pp. 194-210, 1988.

Luhmann, Niklas, “Familiarity, Confidence, Trust: Problems and Alternatives,” In
Gambetta, Diego, (Eds) Trust, Basil Blackwell, Inc., New, York, pp. 94-108,
1988.

Luhmann, Niklas, Trust and Power, (H. Davis, J. Raffan, & K. Rooney, Trans), Wiley,
New York, 1979.

Lusch, Robert F., “Sources of Power: Their Impact on Intrachannel Conflict.” Journal of
Marketing Research, 13, (November 1976): 382-390.

Lusch, Robert F., and James R. Brown, “Interdependency, Contracting, and Relational
Behavior in Marketing Channels.” Journal of Marketing, 60, 4, (October 1996):
19-38.

Macaulay, S., “Non-Contractual Relations in Business,” American Sociological Review,
Vol., 28, pp. 55-70, (1963).

208

 

Marsden, P. V. Introducing Influence Processes into a System of Collective Decisions,
American Journal of Sociology, vol. 86, pp. 1203-1235, 1981.

Mayer, Roger, C., Davis, James, H., and Schoorman, F. David, “An Integrative Model of
Organizational Trust,” The Academy of Management Review, vol. 20, no. 3, pp.
709-734, July, 1995.

Mcallister, Daniel J ., Interpersonal Trust, Citizenship Behavior, and Performance: A
Relational Analysis, Unpublished dissertation, Irvine, California, University of
California, Irvine, September, 1993.

Mcfall, Mike, “Epert Systems- Computer aided Quality Assurance,” Quality, Vol. 26,
Issue 9 , pp. 14-16, September, 1987.

Miles, Matthew B., and Huberrnan A. Michael, Qualitative Data Analysis: An Expanded
Sourcebook, Sage Publications, Thousand Oaks California, 1994.

Mintzberg, Henry, Raisinghani D. and Theoret, A., “The Structure of ‘Unstructured’
Decision Processes,” Administrative Quarterly, vol. 21 pp. 246-275, 1976.

Mishra, Aneil, K., “Organizational Responses to Crisis: The Centrality of Trust,” in
Kramer, Roderick, M., and Tyler, Tom, R., (Eds) Trust in Organizations:
Frontiers of Theory and Research, Sage Publications, Thousand Oaks, California,
1995, pp. 261-287.

Mohr, Jakki, and Job R. Nevin, “Communication Strategies in Marketing Channels: A
Theoretical Perspective.” Journal of Marketing, (October 1990): 36-51.

Mohr, J ., & Spekman, R., “Characteristics of Partnership Success: Partnership Attributes,
Communication Behavior, and Conflict Resolution Techniques,” Strategic
Management Journal, vol. 15, pp. 135-152, 1994.

Monczka, Robert, M., Petersen, Kenneth, J ., Handfield, Robert, B., and Ragatz, Gary, L.,
“A Study of Determinants of Successfirl Strategic Supplier Alliances,” Working
Paper, Michigan State University, 1996.

Morse, Janice, Critical Issues in Qualitative Research Methods, Sage Publications,
Thousand Oaks, CA. 1991.

Narus, James A. and Anderson, James C., “Distributor Contributions to Partnerships with
Manufacturers,” Business Horizons, pp. 34-42, 1987.

Newman, Richard G. “Single Sourcing: Short-term Savings versus Long-terrn Problems,”

Journal of Purchasing and Materials Management, Vol. 25, Iss. 2, pp. 20-25
,Summer 1989.

209

Nishiguchi, Toshihiro, Strategic Industrial Sourcing: The Japanese Advantage, New
York: Oxford University Press, 1994.

Noordevvier, Thomas G, John, George, and Nevin, John R., “Performance Outcomes of
Purchasing Arrangements in Industrial Buyer-Vendor Relationships,” Journal of
Marketing, vol. 54, pp. 80-93, 1990.

Nunnally, Jim C., Psychometirc Theory, 2nd editions, McGraw-Hill Publishing
Company, New York, 1978.

Parkhe, Arvind, “Strategic Alliance Structuring: A Game Theoretic and Transaction Cost

Examination of Interfirm Cooperation,” Academy of Management Journal, vol. 36,
no. 4, pp. 794-829, 1993.

Pennings, Johanne, & Woiceshyn, M., New Technology as Organizational Innovation:
The Development and Diflerence of Micro-electronics, Ballinger Publication
Company, Cambridge, MA, 1987.

Pilling, Bruce K. and Zhand, Li, “Cooperative Exchange: Rewards and Risks,”
International Journal of Purchasing and Materials Management, pp. 2-9, Spring,
1992.

Pisano, Gary P. “The R&D Boundaries of the Firm: An Empirical Analysis,”
Administrative Science Quarterly, Vol. 35, Iss. 1, pp. 153-176, March 1990.

Powell, W. W., “Hybrid Organizational Arrangements,” California Management Review,
Vol. 30, No. 1, pp. 67-87, 1987.

Prescutti, William, D., “The Single Source Issue: US. and Japanese Souring Strategies,
International Journal of Purchasing and Materials Management, Vol. pp. 2-9,
1992.

Provan, Keith G. and Skinner, Steven J ., “Interorganizational Dependence and Control as
Predictors of Opportunism in Dealer-Supplier Relations,” Academy of
Management Journal, vol. 32, no. 1, pp. 202-212, 1989.

Ramanujam, Vasudevan, and Varadarajan, P. “Research on Corporate Diversification: A
Synthesis,” Strategic Management Journal, Vol. 10, Iss. 5, pp. 523-551,
November/December, 1989.

Rempel, J. K., and Holmes, J. G., How do I Trust Thee? Psychology Today, Vol. 20, No.
2, pp. 28-34.

Rempel, J. K., Holmes, J. G., & Zanna, M., “Trust in Close Relationships, Journal of
Personality and Social Psychology, vol. 49, pp. 163-168, 1985.

210

Ring, Peter Smith, “Fragile and Resilient Trust and their Roles in Economic Echange,
Business and Socieity, Vol. 35, Iss. 2, pp. 148-175, June, 1996.

Ring, P. S. & Van de Ven, A. H., “Structuring Cooperative Relationships Between
Organizations,” Strategic Management Journal, vol. 49, pp. 95-1 12, 1992.

Ring Peter Smith, and Van de Ven, “Developmental Processes of Cooperative
Interorganizational Relationships,” Academy of Management Review, Vol. 19,
Issue 1, pp. 90-118, January, 1994.

Rogers, Carl, B., On Becoming a Person, Houghton Mifflin, Boston, Massachusetts,
1964.

Rosen, B. and Jerdee, T. H., “Influence of Subordinate Characteristics on Trust and use of
Participation Decision Strategies in a Managerial Simulation,” Journal of Applied
Psychology, Vol. 62, pp. 628-631, 1977.

Rotter, J. B., “A New Scale For the Measurement of Interpersonal Trust,” Journal of
Personality, vol. 35, pp. 651-665, 1967.

Rousseau, Denise M., Promises in Action: Psychological Contracts in Organizations.
Newbury Park, CA: Sage Publications, 1995.

Sabel, C. F ., “Studied Trust: Building New Forms of Cooperation in a Volatile
Economy,” Human Relations, Vol. 46, No. 9, pp. 1133-1170, 1993.

Schmitz, Judith, M., Frankel, Robert, and F rayer, David, J ., “ECR Alliances, A Best
Practices Model,“ Joint Industry Project on Eflicient Consumer Response, Grocery
Manufacturers of America, pp. 54-55, 1995.

Schumacker, Randall E., and Lomax, Richard G., Beginner’s Guide to Structural
Equation Modeling, Lawrence Erlbaurn Associates, Mahwah, New Jersey, 1996.

Schurr, P. H., & Ozanne, J. L., “Influences on Exchange Processes: Buyers’
Preconceptions of a Seller’s Trustworthiness and Bargaining toughness,” Journal

of Consumer Research, vol. 11, no. 4, pp. 939-953, 1985.

Schwenk, C. “Laboratory research on ill-structured decision aids: The case of dialectical
inquiry,” Decision Sciences, pp. 140-144, 1983.

Shapiro, S., “The Social Control of Interpersonal Trust,” American Journal of Sociology,
vol. 93, pp. 623-658, 1987.

Shapiro, D. L., Sheppard, B. H., and Cheraskin, L., In Theory: Business on a Handshake,
Negotiation Journal, Vol. 8, pp. 365-377, 1992.

211

Sheppard, Blair H., and Tuchinsky, Maria, “Micro-OB and the Network Organization,” in
Kramer, Roderick, M., and Tyler, Tom, R., (Eds.)Trust in Organizations:
Frontiers of Theory and Research, Sage Publications, Thousand Oaks, California,
1995, pp.140-165.

Sinatar, Marsha, “Building Trust into Corporate Relationships,” Organizational
Dynamics, pp. 73, Winter, 1988.

Silver, A., “Trust in Social and Political Theory,” In G. D. Suttles & m. N. Zald (Eds),
The Challenge of Social Control, Norwood, New Jersey: Ablex, 1985.

Smith, J ., Brock, Horizontal Selling Alliances: The Eflect of Organizational Distance and
Mutual Trust on Dyadic Working Relationships, Unpublished dissertation, Ontario,
Canada, University of Western Ontario, August, 1993.

Solomon, L., “The Influence of Some types of Power Relationship and Game Strategies
Upon the Development of Interpersonal Trust,” Journal of A bnormal and social
Psychology, vol. 61, pp. 223-330, 1960.

Stephens, James, A., An Expectational View of Marketing Relationships and the Role of
Trust and Disclosure, Unpublished dissertation, Tempe, Arizona, Arizona State
University, May, 1993.

Strickland “Surveillance and Trust,” Journal of Personality, Vol. 126, pp. 245-250, 1958.

Taylor, R. G., “Trust and Influence in the workplace,” Organization Development
Journal, vol. 8, pp. 333-36, 1990.

Tesler, L. G., “A Theory of Self-enforcing Agreements,” Journal of Business, vol. 53, pp.
27-44, 1980.

Thibaut, J. and Kelly, H. H., The Social Psychology of Groups, New York, John Wiley
1959.

Treleven, Mark, “Single Sourcing: A Management Tool for the Quality Supplier,”
Journal of Purchasing and Materials Management,” Vol. 23, Iss. 1, pp. 19-24,
1 987.

Tway, Duane Converse Jr., A Construct of trust, Unpublished Dissertation, Austin,
Texas, University of Texas, 1994.

Van de Ven, Andrew, and Walker, Gordon , Administrative Science Quarterly, Vol. 29
Iss 4, pp. 598-621, December 1984.

212

'l

Van Gigch, Joseph P., Applied General Systems Theory 2nd edition, New York: Harper
& Row, Publishers, 1978.

Vedder, James, N., “How Much Can We Learn From Success,”_A cademy of Management
Executive, vol. 6, no. 1, pp. 56-65, 1992.

Venkatesh, R., Ajay K. Kohil, and Gerald Zaltman, “Influence Strategies in Buying
Centers.” Journal of Marketing, 59, 4, October 1995: 71-82.

Walter, Elaine, Wallster, G., William, Berscheid, Elllen, Equity: Theory and Research,
Allyn and Bacon Inc., Boston, MA, 1978.

Whitney, John, O., The Trust Factor, McGraw-Hill, Inc., New York, 1994.

Wiley, Kee, H., & Knox, R., “Conceptual and methodological considerations in the study
of trust and suspicion,” Journal of Conflict Resolution, vol. 13, pp. 357-366, 1970.

Williams, Bernard, “Formal Structures and Social Reality,” In Gambetta, Diego, (Eds)
Trust, Basil Blackwell, Inc., New, York, pp. 3-13, 1988.

Williamson, O. E., Markets and Hierarchies, : Analysis and Antitrust Implications, The
Free Press, New York, 1975.

Williamson, 0. E., “Transaction Cost Economics: The Governance of Contractual
Relations, Journal of Law and Economic, Vol. 22, pp. 233-261, 1979.

Williamson, O. E., Credible Commitments: Using hostages to support exchange.
American Economic Review, vol. 73, pp. 519-540, 1983.

Willaimson, O. E., The Economic Institutions of Capitalism, Free Press, New York, 1985.

Williamson, O. E. Calculativeness, Trust and Economic Organization, Journal of Law
and Economics, Vol 36, pp. 453—486, 1993.

Yin, Robert, K., Case Study Research: Design and Methods, second edition, Sage
Publications, Thousand Oaks, California 1994.

Yoshino, Michael Y. and Rangan, U. Srinivasa, Strategic Alliances: An Entrepreneurial
Approach to Globalization, Boston, MA, Harvard Business School Press, 1995.

Zaheer, Akbar, & Venkatraman, N. , “Relational Governance as an Interorganizational

Strategy: An Empirical Test of the role of Trust in Economic Exchange, ” Strategic
Management Journal, vol. 16, pp. 373-392, 1995.

213

Zaj ac, Edward J. and Olsen, Cyrus P., “From Transaction Cost of Transactional Value
Analysis: Implications for the Study of Interorganizational Strategies,” Journal of

Management Studies, Vol 30, No. 1, pp. 131-145, 1992.

land, Dale, E., “Trust and Managerial Problem Solving,” Administrative Science
Quarterly, 17, p. 229-239, 1972.

Zucker, L. G. “Production of Trust,” In B. M. Staw & L. L. Cummings (Eds), Research
in Organizational Behavior, vol. 8, Greenwich, CN. pp. 53-111, 1986.

214

"‘illllllllllllllES