a...» an“:
.23 c $9.5»...

may) kill}. v

.. 1293.
it}. fan? 1

dtxa. fl
V)

‘ ‘ a L 434 ,

.. h».

.
.a.

g:§3szx

.1:
a:

1..
a. in“? -
$fi’x?..s|t:

. x£wfin..‘1.
Luflfuvt\h..l 1.143121%:
-$;|’ c n"'

v glint. ,
Ia . ’9
I. .

vm."
t...

‘ ,
u 1:.woatjri.
a ‘
Ila-"Vs. I}...

133..

L114“ ”on”

 

 

1343333

90')";

 

This is to certify that the
dissertation entitled

THE ROLE OF SECURITY IN THE SUPPLIER SELECTION
DECISION

presented by
MICHAEL DOUGLAS VOSS

has been accepted towards fulfillment
of the requirements for the

PhD. degree in Marketing and Supply Chain
, Maflement

 

 

Major Profefior’s Signature
m J/UmVL’L g; Q ()3 C

Date

 

MSU is an Affirmative Action/Equal Opportunity Institution

LIBRARY
Michigan State
University

—_——-—— __,

-- —.-u--- -n-.-o---o-.-----.---a-o-o-o--o-- -.- _.._ _

I
.
I
I
.
I
I
I
I
.
I
I
.
I
I
.
I
.
O

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 p:lClRC/DateDue.indd-p.1

 

 

THE ROLE OF SECURITY IN THE SUPPLIER SELECTION DECISION
By

Michael Douglas Voss

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Marketing and Supply Chain Management

2006

ABSTRACT
THE ROLE OF SECURITY IN THE SUPPLIER SELECTION DECISION
By
Michael Douglas Voss
This research investigates firm trade-offs related to supplier security in the food

industry. This research provides practitioners with empirical evidence to support supply
chain security change management initiatives. This research also informs suppliers of the
trade-offs their customers are willing to accept for increased security. The study further
provides insight regarding the importance of security as a supplier selection criterion.
The results provide academics insight into the importance of security as it relates to price
and delivery reliability, two important supplier selection criteria. The results are also
useful to practitioners to evaluate if their firm Should invest in supply chain security
measures. It is often difficult for practitioners to justify investments in the absence of
government regulation or evidence supporting that the initiative is beneficial. This
research provides supporting, or dissenting, evidence related to a supplier’s ability to reap
rewards from security implementation. The results of the research are also of interest to
public policy makers. The results allow public policy makers to determine if regulation
is necessary to induce firms to increase security, or if there is a demand for increased
supplier security. In the latter case, the free market could substitute for burdensome

regulation.

To Lee Ann, Mason, and any future additions

iii

ACKNOWLEDGEMENTS

As I sit down to write these Simple words of acknowledgement to those who have
made the conclusion of this journey possible, I’m reminded of a scene that occurred at the
conclusion of another journey. As an undergraduate at The University of Arkansas at
Fayetteville I was blessed with a group of friends who were more like brothers. We spent
countless hours together during our first four years in college doing things that we both
should and should not do. The good stories usually came out of those things we should
not have been doing, but I digress. While some of us stayed on longer than four years
(including yours truly) a few members of this group departed Fayetteville “on time”. The
same spring that these friends left, I was also leaving Fayetteville for an internship in
Atlanta. While I would rettu'n to Fayetteville for my last semester of college, I knew that
many of these fiiends would not be there in the same capacity. Similarly, we all knew
that the life, and friendships, we had known would never be the same. Yet this
knowledge was kept in the back of our minds. We didn’t want to admit the impending
change. So sets the stage for a night out in which we all ventured to a local concert hall
to watch Texas musician Robert Earl Keen (I dare say that there is not another
dissertation acknowledgement that mentions Robert Earl Keen). While Robert Earl Keen
has many songs, he closes his shows with one entitled “The Road Goes on Forever zuid
the Party Never Ends.” This song seemed particularly appropriate that night and, in fact,
has remained very appropriate. My group of brothers remains extremely close and the

road truly goes on forever with the proverbial never-ending party in tow.

iv

I also feel that completing the doctoral program is but the beginning of a never-
ending road and a life firll of “partying” where partying is defined as enjoying ones work,
making more money than one deserves (thank you to the taxpayers), and having enough
free time to enjoy the fruits of ones labor with family and friends. It is said that being a
college professor is the second best job to being a rock star. I’m convinced this is the
truth, but I’ve gotten too old for the activities that go along with being a rock star, so I
guess I’ll stick this college professor thing out for a while.

Many have contributed to the successful completion of my doctoral program and
thus my ability to party down this road. First, I would like to thank God for the many
blessings he has bestowed upon my family and 1. Without Him, no great accomplishment
is attainable, but with Him all things are possible.

Second, my loving wife. I would like to thank my wife Lee Arm for all of her
love and support. There can be no better person in the world than her. There is also
absolutely, positively, without a doubt, no chance that I would have finished the program
without her. Her hard work as an extraordinarily skilled nurse also allowed uS to do so
many things during our time in Michigan that we never would have been able to do
otherwise. I will always remember our time here and the special memories that we will
keep for the rest of our lives together.

Next, my family. My parents are the two hardest working people I know and they
sacrificed a great deal over the years to make sure that I never wanted for anything. They
also made doctoral program success possible by providing me with love, a strong work
ethic, guidance, and a desire to do the right thing. I would also like to thank my two

grandmothers, Jean Voss and Martha Oliver, for their love and support before, during,

and after my undergraduate, masters, and doctoral programs. Further, I would be remiss
if I didn’t acknowledge my “new” family: Warren, Pat, and Jason Campbell. While
Warren left us too early, I would like to thank he and Pat for raising such a wonderful
daughter, who I love with all my heart, for taking a risk and allowing Lee Arm to venture
up to the “Frozen Tundra”, and for being such fantastic people. I would also like to thank
Jason for keeping Lee Ann in line before I arrived on the scene.

Where would I be without faculty guidance? I’ll begin by thanking John Ozment
of the University of Arkansas. The way in which John treats everyone, including his
doctoral students, was not only directly of great assistance to me as a student at Arkansas
but also indirectly as a doctoral student at Michigan State through Scott Keller. As my
fiiend and mentor at Michigan State, Scott treated me as an equal while also taking
untold hours out of his schedule to educate me in the explicit and tacit nuances of
academic success. Without Scott, I’m certain my doctoral career would have ended
before I even took my comps. Scott and John have been, and will remain, valued friends
and colleagues. Hopefully I can pass on their tradition to any doctoral students under my
tutelage. I would also like to thank Roger Calantone. Over the past five years Dr.
Calantone has taken an interest in my success and gone out of his way to provide
instruction, advice, assistance, conversation, and laughter. The chair of my dissertation
committee, Dave Closs, also deserves the highest acknowledgement. Dr. Closs’
guidance, patience, and giving nature have allowed me to not only successfully complete
my dissertation, but also have made me a better scholar and allowed my family and I

comfort during our time at Michigan State. Two other members of my dissertation

vi

committee merit mention: Cheri Speier and Keith Helferich. Dr.’s Speier and Helferich
were an invaluable part of my committee and I cannot thank them enough.

Fifth, a man is nothing without his friends. Specifically, I would like to thank the
friends I referenced in the opening of this section and others that have joined our merry
band of malfeasants. In no particular order: Scott McDaniel, Clay Campbell, Brandon
Timbes, Barrett Hunter, Chad Smith, Steven Bauer, Cliff Pumphrey, Stephen Terrill, J.D.
Robbins, and Ben Dorman. I could truly never hope for better friends/brothers and I
thank you all for your support over the years. Specifically, with regard to doctoral
program success, I would like to thank Bauer for all of his help with my dissertation.
Bauer took time out of his schedule and shared his contacts with me such that I was able
to complete my work. Another friend that I must mention is my office mate for my first
four years at Michigan State, Gilbert Nyaga. Gilbert was as true a friend as I could hope
for and we certainly went through many trials and tribulations together. However,
through all these trials and tribulations, Gilbert was always positive, supportive, and the
perfect office mate. Through our time together I learned a great deal about the Kenyan
culture, ate some incredible Kenyan food, and met some wonderful friends of his as well
as his wonderful wife Lucy. I would also like to thank Gilbert for putting up with me. I
know it was not easy. Finally, thanks to Drew and Holly Shull for their friendship and
the lamp they sent me a few weeks after I arrived in Michigan. I used it every day, Shull!

Finally, I would like to extend my sincere gratitude to the wonderful ladies who
work in the Marketing and Supply Chain main office. Laurie Fitch, Cheryl Lundeen,
Sherry Danner, Kathy Mullins, Kathy Waldie, as well as others who have come and gone

have all been a source of assistance, laughter, and generosity. I appreciate them greatly.

vii

In sum, Lord knows I could not have completed this journey on my own. The
above parties have made possible any success I have had. To each of them: Thank you

from the bottom of my heart. May the road go on forever and the party never end.

One more thing:
Pure as the dawn on the brow of thy beauty
Watches thy soul from the mountains of God
Over the Fates of thy children departed
Far from the land where their footsteps have trod.
Beacon of hope in the ways dreary lighted;
Pride of our hearts that are loyal and true;
From those who adore unto one who adores us—
Mother of Mothers, we sing unto you.
Hit that line, hit that line, keep on going
Take that ball right down the field
Give a cheer, RAH!! RAHl!
Never fear, RAHI! RAH!!
Arkansas will never yield
On your toes, Razorbacks, to the finish
Carry on with all your might
For it’s A-A-R-K-A-N-S-A-S for Arkansas
Fight! Fight! Fight!

WOOOOOOOOOOOOOOOOOO PIG SOUIEEEE, RAZORBAC KS!

viii

TABLE OF CONTENTS

LIST OF TABLES ..................................................................... xi
LIST OF FIGURES ................................................................... xvi
CHAPTER 1: INTRODUCTION .................................................. 1
1.0 Introduction ......................................................................... 1
1.1 The Need for Increased Security ................................................. 1
1.1.1 Terrorist Motivations ........................................................ 2
1.1.2 Impacts of Terrorism ......................................................... 3
1.2 Supply Chains and Supply Chain Security ..................................... 3
1.2.1 Consequences of Supply Chain Security Failure ......................... 4
1.2.2 Vulnerability of the United States Supply Chain ......................... 4
1.2.3 Supply Chain Security Management ....................................... 5
1.3 The Importance of Food and Food Supply Chain Security .................. 7
1.4 Impact of Security on Supplier Evaluations .................................... 8
1.5 Research Objectives ............................................................... 10
1.6 Research Questions ................................................................ 10
1.7 Scope of Research .................................................................. 12
1.8 Contributions to Theory and Management Practice ........................... 13
1.9 Research Organization ............................................................ 14
CHAPTER 2: LITERATURE REVIEW ......................................... 15
2.0 Introduction ......................................................................... 15
2.1 Overview of Supply Chain Security ............................................. 15
2.1.1 Pressures Leading to Security Implementation ........................... 17
2.1.2 The State of Supply Chain Security Best Practice ....................... 19
2.1.3 Cost and Service Implications of Security Implementation ............ 23
2.2 Food and Food Supply Chain Security .......................................... 27
2.2.1 Food Vulnerability ............................................................ 29
2.2.2 Food Supply Chain Security Best Practice ................................. 31
2.3 Supplier Selection Criteria ........................................................ 33
2.4 Synthesis ............................................................................ 36
2.5 Statement of Problem .............................................................. 37
2.5.1 Research Questions and Hypotheses ....................................... 37
2.6 Conclusion ............................................................................ 43
CHAPTER 3: METHODOLOGY ................................................. 44
3.0 Introduction ......................................................................... 44
3.1 Research Purpose and Objectives ................................................ 44

ix

3.2 Research Methodology ............................................................ 45

3.2.1 Unit of Analysis .............................................................. 45
3.2.2 Sampling Procedure ......................................................... 45
3.2.3 Data Collection and Implementation ....................................... 47
3.2.4 Instrument ..................................................................... 50
3.2.5 Data Analysis .................................................................. 56
3.3 Conclusion ........................................................................... 57
CHAPTER 4: ANALYSIS OF DATA ............................................. 58
4.0 Introduction ......................................................................... 58
4.1 Preliminary Analysis ............................................................... 59
4.1.1 Mailing Procedures and Final Sample Size .............................. 59
4.1.2 Characteristics of Sample ................................................... 60
4.2 Hypotheses Tests ................................................................... 64
4.3 Summary ............................................................................ 124
CHAPTER 5: IMPLICATIONS AND FUTURE RESEARCH ............. 128
5.0 Introduction ......................................................................... 128
5.1 Implications ............................................................................ 128
5.1.1 Managerial Implications .................................................... 130
5.1.2 Public Policy Implications ................................................... 137
5.1.3 Theoretical Implications .................................................... 138
5.2 Recommendations .................................................................. 139
5.3 Limitations ........................................................................... 143
5.4 Future Research ..................................................................... 145
APPENDICES .......................................................................... 148
APPENDIX A: Complete Scenario List ........................................... 149
APPENDIX B: Survey Instrument ................................................ 152
APPENDIX C: Wilcoxon Rank Sum Test Results .............................. 162
APPENDIX D: Pre-Notification Letter ........................................... 181
APPENDIX E: Notification Letter ................................................ 182
APPENDIX F: Final Notification Letter ...................................................... 183
APPENDIX C: UCHRIS Assurance of Exempt Research Protocol ......... 184
BIBLIOGRAPHY ..................................................................... l 86

Table 2.1

Table 2.2

Table 2.3

Table 2.4

Table 2.5

Table 2.6

Table 3.1

Table 3.2

Table 3.3

Table 4.1

Table 4.2

Table 4.3

Table 4.4

Table 4.5

Table 4.6

Table 4.7

Table 4.8

Table 4.9

Table 4.10

LIST OF TABLES

Drivers of Security Initiatives ........................................

Supply Chain Security Best Practice I .............. ' ................
Supply Chain Security Best Practice 11 ..............................
Areas of Food Vulnerability ..........................................

MSU Food Supply Chain Security Competencies and
Capabilities ...............................................................

Importance of Supplier Choice Criteria Across Studies ..........
Pretest Steps ............................................................
Factors and Levels ...................................................

Factor and Factor Level Definitions Presented to Respondents.
Sample Characteristics ................................................
Early and Late Respondent Characteristics .........................
Research Questions and Hypotheses .................................
Hypotheses 1-4 Results: Utilities ....................................
Factor Importance Scores .............................................
Hypothesis 5 Results: Security by Price Interaction ..............

Hypothesis 5 Results: Assessment of Significant Differences
Between Utility Values ................................................

Security by Price Interaction: Market Preference Comparison.

Hypothesis 6 Results: Security by Delivery Reliability
Interaction ...............................................................

Hypothesis 6 Results: Assessment of Significant Differences
Between Utility Values ................................................

xi

17

20
71

A—

30

49

51

54

62

63

66

70

76

76

77

80

81

82

Table 4.11

Table 4.12

Table 4.13

Table 4.14

Table 4.15

Table 4.16

Table 4.17

Table 4.18

Table 4.19

Table 4.20

Table 4.21

Table 4.22

Table 4.23

Table 4.24

Table 4.25

Table 4.26

Security by Delivery Reliability Interaction: Market
Preference Comparison ................................................

Respondents’ Concern over Security Incidents ....................
Factor Importance Scores: Segmented Sample ....................

Hypothesis 7 Results: Security by Incident Concern
Interaction ...............................................................

Hypothesis 8 Results: Security by Delivery Reliability by
Incident Concern Interaction ..........................................

Hypothesis 8 Results: Assessment of Significant Differences
Between Utility Values: Concerned Respondents ................

Hypothesis 8 Results: Assessment of Significant Differences
Between Utility Values: Unconcemed Respondents .............

Security by Delivery Reliability Interaction: Market
Preference Comparison for Concemed Respondents ..............

Security by Delivery Reliability Interaction: Market
Preference Comparison for Unconcemed Respondents ...........

Hypothesis 9 Results: Security by Price by Incident Concern
Interaction ...............................................................

Hypothesis 9 Results: Assessment of Significant Differences
Between Utility Values: Concerned Respondents ................

Hypothesis 9 Results: Assessment of Significant Differences
Between Utility Values: Unconcemed Respondents .............

Security by Price by Incident Concern Interaction: Market
Preference Comparison for Concerned Respondents ..............

Security by Price by Incident Concern Interaction: Market
Preference Comparison for Unconcemed Respondents ...........

Hypothesis 10 Results: Security by Sourcing Location
Interaction ...............................................................

Hypothesis 10 Results: Assessment of Significant Differences
Between Utility Values: Unconcemed Respondents .............

xii

85

87

88

89

92

92

93

97

98

100

100

101

104

105

108

108

Table 4.27

Table 4.28

Table 4.29

Table 4.30

Table 4.31

Table 4.32

Table 4.33

Table 4.34

Table 4.35

Table 4.36

Table 4.37

Table C.1

Table C.2

Table C. 3

Table C.4

Table C. 5

Hypothesis 11 Results: Delivery Reliability by Sourcing
Location Interaction ...................................................

Hypothesis 11 Results: Assessment of Significant Differences
Between Utility Values ................................................

Hypothesis 11 Results: Security by Delivery Reliability by
Sourcing Location Interaction: Market Preference ...............

Security by Delivery Reliability Interaction for International
Sourcing: Market Preference Comparison I ........................

Security by Delivery Reliability Interaction for International
Sourcing: Market Preference Comparison 11 ......................

Hypothesis 12 Results: Price by Sourcing Location Interaction

Hypothesis 12 Results: Assessment of Significant Differences
Between Utility Values ................................................

Hypothesis 12 Results: Security by Price by Sourcing
Location Interaction ...................................................

Security by Price by Sourcing Location Interaction: Market
Preference Comparison I ..............................................

Security by Price by Sourcing Location Interaction: Market
Preference Comparison 11 .............................................

Summary of Results ...................................................
Hypothesis 5 Results: Security by Price Interaction ..............

Test for Significant Difference in Market Preference for Basic
and Advanced Security for all Levels of Price .....................

Hypothesis 6 Results: Security by Delivery Reliability
Interaction ................................................................

Test for Significant Difference in Market Preference for Basic
and Advanced Security for all Levels of Delivery Reliability...

Hypothesis 8 Results: Security by Delivery Reliability by
Incident Concern Interaction ..........................................

xiii

111

112

115

116

118

118

119

121

164

164

165

Table C.6

Table C.7

Table C.8

Table C.9

Table C.10

Table C.11

Table C.12

Table C. 13

Table C. 14

Table C. 15

Table C.16

Table C. 17

Test for Significant Difference in Market Preference for
Security and Delivery Reliability Between Concerned and
Unconcemed Respondents ............................................

Test for Significant Difference Between Market Preference of
Suppliers with Advanced Security v. Market Preference for
Suppliers with Basic Security for Concerned Respondents ......

Test for Significant Difference Between Market Preference of
Suppliers with Advanced Security v. Market Preference of
Suppliers with Basic Security for Unconcemed Respondents...

Hypothesis 9 Results: Security by Price by Incident Concern
Interaction ................................................................

Test for Significant Difference in Market Preference for
Security and Price for Concerned and Unconcemed
Respondents .............................................................

Test for Significant Difference Between Market Preference of
Suppliers with Advanced Security v. Market Preference of
suppliers with Basic Security for Concerned Respondents .......

Test for Significant Difference Between Market Preference of
Suppliers with Advanced Security v. Market Preference of
Suppliers with Basic Security for Unconcemed Respondents...

Hypothesis 11 Results: Security by Delivery Reliability by
Sourcing Location Interaction: Market Preference ...............

Test for Significant Difference in Market Preference for
Security and Delivery Reliability for Domestic and
International Sourcing .................................................

Test for Significant Difference Between Market Preference of
Suppliers with Advanced Security when Sourcing
Domestically v. Sourcing Internationally ...........................

Test for Significant Difference Between Market Preference of
Suppliers with Basic Security when Sourcing Internationally v.
Suppliers with Advanced Security when Sourcing
Internationally ...........................................................

Hypothesis 12 Results: Security by Price by Sourcing
Location Interaction ....................................................

xiv

166

167

168

169

170

171

172

173

174

175

176

177

Table C. 18

Table C.19

Table C. 20

Test for Significant Difference Between Market Preference for
Domestic and International Sourcing at all Levels of Security
and Price ................................................................. 178

Test for Significant Difference Between Market Preference of
Suppliers with Advanced Security when Sourcing
Domestically v. Sourcing Internationally ........................... 179

Test for Significant Difference Between Market Preference of
Suppliers with Basic Security when Sourcing Internationally v.
Suppliers with Advanced Security when Sourcing

Internationally ......................................................... l 80

XV

Figure 2.1
Figure 2.2
Figure 2.3
Figure 2.4
Figure 4.1
Figure 4.2
Figure 4.3
Figure 4.4

Figure 4.5

Figure 4.6

Figure 4.7

Figure 4.8

Figure 4.9

Figure 4.10

Figure 4.11

Figure 4.12

Figure 4.13

LIST OF FIGURES

The Conflict Between Business and Security Priorities ...........
Hoof to Plate Supply Chain ...........................................
Seed to Plate Supply Chain ...........................................
Hypothesized Model ...................................................
Hypothesized Model ...................................................
Security by Price Interaction: Utilities ..............................
Security by Price Interaction: Market Preference .................
Security by Delivery Reliability Interaction: Utilities. . . . . . . .

Security by Delivery Reliability Interaction: Market
Preference ...............................................................

Security by Incident Concern Interaction: Concerned
Respondent Utilities ...................................................

Security by Incident Concern Interaction: Unconcemed
Respondent Utilities ...................................................

Security by Incident Concern Interaction: Market Preference...

Security by Delivery Reliability by Incident Concern
Interaction: Utilities for Concerned Respondents .................

Security by Delivery Reliability by Incident Concern: Utilities
for Unconcemed Respondents .......................................

Security by Delivery Reliability by Incident Concern
Interaction: Market Preference .......................................

Security by Price by Incident Concern Interaction: Utilities for
Concerned Respondents ...............................................

Security by Price by Incident Concern Interaction: Utilities for
Unconcemed Respondents ............................................

xvi

25

28

29

65

78

79

83

84

89

90

90

94

95

96

102

102

Figure 4.14

Figure 4.15
Figure 4.16

Figure 4.17

Figure 4.18

Figure 4.19

Figure 4.20

Figure 5.1

Security by Price by Incident Concern Interaction: Market
Preference ...............................................................

Security by Sourcing Location Interaction: Utilities ..............
Security by Sourcing Location Interaction: Market Preference.

Delivery Reliability by Sourcing Location Interaction:
Utilities ..................................................................

Security by Delivery Reliability by Sourcing Location
Interaction: Market Preference .......................................

Price by Sourcing Location Interaction: Utilities ..................

Security by Price by Sourcing Location Interaction: Market
Preference ...............................................................

Security Implementation Matrix .....................................

xvii

103

109

109

113

114

119

120

135

CHAPTER 1
INTRODUCTION
This research investigates firm trade-offs related to supplier security in the food
industry. This research provides practitioners with empirical evidence to support supply
chain security change management initiatives. This research also informs suppliers of the
trade-offs their customers are willing to accept for increased security. The study further
provides insight regarding the importance of security as a supplier selection criterion.
First, this chapter provides background information supporting the need for security.
Second, the chapter introduces supply chains and supply chain security. Third, the
chapter discusses the importance of the food supply and food supply chain security.
Fourth, the chapter dissects the impact of security on suppliers’ price and delivery
reliability performance. Fifth, the chapter addresses objectives of the current research
and specific research questions. Finally, the chapter describes the scope, theoretical and

managerial contributions, and organization of the proposed research.

THE NEED FOR INCREASED SECURITY

The United States is increasingly aware of the threat posed by terrorists. Seldom
before has such a clear and present danger to the homeland been at the forefront of the
Nation’s collective consciousness. Sadly, it took the events of September 11"“, 2001 to
expose the vulnerability of the United States. Supply chain infrastructure and assets are
particularly vulnerable as targets and tools used to carry out terrorist acts. The bombing
of the World Trade Center in 1993 and the Murrah Federal Building on Oklahoma City in

1995 were accomplished through utilization of transportation assets to deliver weapons.

The supply chain infrastructure, and accompanying passengers, was the target of the Aum
Shinrikyo sarin gas attack on a Tokyo subway in 1995, the Spanish commuter train
bombing of 2004, and the London tube and bus attacks of 2005. These events, and those
of 9/ 1 1, heightened public and private awareness of the need to secure supply chain

infrastructure and assets.

Terrorist motivations

Terrorists hold many reasons for attacking the United States (Transportation
Security Agency, 2003). The United States now enjoys the benefits and pitfalls of its
status as the world’s only remaining super power. In this role, U.S. policies and culture
receive global media attention. Terrorists ofien use these policies, and their disdain for
US. culture, to justify their actions. For example, United States support of Israel is
diametrically opposed to some terrorist goals and they seek to punish the United States as
a form of retaliation. The culture of the United States is often thought to be sacrilegious
by many of these groups. This predominantly Christian culture, characterized by some as
placing worldly possessions over religious priorities, leads terrorists to view attacks
against the United States as part of a “jihad” (holy war). Finally, U.S. led coalitions have
participated in the Gulf Wars of the early 1990’s and 2000’s. The stationing of US.
troops in this region has led terrorists to seek retribution for the perceived occupation of

their holy land.

Impacts of terrorism

Few of these cultural and geopolitical terrorist motivations are likely to change
soon. Thus, the United States should be aware that a terrorist threat is imminent and
could severely impact the country. Wolfe (2001) posits three classes of impact resulting
from a terrorist attack. Primary impacts represent the actual damage, casualties, and
disruption that result. These are the primary motivations of terrorists seeking notoriety
for their cause. Direct secondary impacts include the cost of rescue and recovery efforts,
such as clearing damage and mobilizing resources. Both primary and direct secondary
impacts impose a heavy cost on the victimized nation. However, the primary economic
impact of a successful terrorist attack lies in the resulting indirect secondary impacts.
Indirect secondary impacts represent the countermeasures taken by the victim to prevent
further damage. A short-term example of an indirect secondary impact is the closing of
US. airspace after 9/11, which cost air carriers millions in lost revenue. In the long-term,
continuing security efforts cause inconvenience to passengers and may cause them to
utilize other modes of transportation. The total cost of post-9/ 11 security efforts to the
US. economy, also an indirect secondary impact, is anticipated to exceed $151 billion

annually (Russell and Sandalha, 2003).

SUPPLY CHAINS AND SUPPLY CHAIN SECURITY

Over the past thirty years, eighty percent of terrorist attacks against US. interests
have targeted business entities (Bremmer, 2001; Dobie et al., 2000). A major focus of
these attacks is the supply chain infrastructure and assets that can serve as a target or

weapon delivery system. Therefore, it is necessary for industry to protect their supply

chains from terrorist intrusion in order to prevent undue harm to the populace and protect

firm viability.

Consequences of supply chain security failure

Radjou (2003) provides several adverse consequences of supply chain security
failures. First, a terrorist attack levied upon a firm’s supply chain is likely to cause
widespread disruption in the delivery of goods to customers. This leads to a loss of
revenue derived from these deliveries. Second, a loss in brand equity could occur if
customers believe the attack was successful due to a systemic failure of the firm’s
security measures, or that the firm is likely to be a target in the future. Third, this loss of
Short-term and long-term revenue could lead to investor discontent with fimr
performance and subsequent sale of equity investments. A major supply chain disruption
has been shown to devalue stock price by as much as nine percent (Radjou, 2003).
Fourth, this disruption also stands to bring about regulatory scrutiny if the loss in sales is
large enough to prevent the firm from meeting financial predictions. Finally, a successful
terrorist attack against a firm could result in severe legal consequences and their

associated costs.

Vulnerability of the US. supply chain

Unfortunately, protecting the US. supply chain is difficult at best. Several factors
make it vulnerable to attack (Blalock, 2003). Large portions of the supply chain are
unguarded. Only the most important, or visible, parts of the supply chain are regularly

protected. Furthermore, the size of the supply chain infrastructure makes complete

protection nearly impossible. The domestic infrastructure includes roughly 47,000 miles
of interstate, 99,000 miles of class I railroad track, 26,000 miles of navigable waterways,
64,000 miles of oil pipeline, 5,200 airports, and 9,400 commercial waterway facilities
(DOT, 2005).

. The interconnectedness of this infrastructure, one of the greatest assets of the US.
supply chain, also renders it vulnerable. A major disruption in one mode stands to have
deleterious effects on the capabilities of another. There are also a large number of
stakeholders involved and the capability to secure a supply chain is only as strong as its
weakest link. These stakeholders include the government, customers, suppliers, service
providers (6. g. carriers and warehouses), and the human resources that enable efficient
and effective supply chain operations. Finally, there are Significant funding limitations
available to protect this infrastructure.

It is also important to consider the difficulty in securing supply chain assets,
particularly those used in international trade. Ninety percent of international trade, and
almost one-half of US. imports, are transported via cargo containers (United States
Customs and Border Protection, 2004). This figure represents almost nine million
containers unloaded annually in the United States. Ensuring that these containers carry
only appropriate cargo is a daunting task, not to mention the difficulty in securing the

thousands of containers, trailers, and railcars used for domestic shipments.

Supply chain security management
The serious repercussions of terrorist attack, and the difficulty of preventing such

an attack, have led to the study of supply chain security management. In order to

understand how to secure a supply chain, one should first understand how a supply chain
is defined. A supply chain is defined as, “The combination of organizations and service
providers that manage the raw material sourcing, manufacturing, and delivery of goods
from the source of the commodities to the ultimate users” (Closs and McGarrell, 2004 p.
8). Thus, supply chains consist of a network of organizations beginning with raw
material suppliers, concluding with the end consumer, encompassing manufacturers
charged with changing product form, and the service providers who transport and store
the products. Supply chains, and the management of the activities involved in their
efficient and effective operation, have come to represent a source of competitive
advantage for many firms (Bowersox, Closs, and Stank, 1999; Kent and Flint, 1997).
However, the circumstances inherent to the post-9/11 world have added significant
complexity and cost to supply chain management. This has led to the concept of supply
chain security management.

Closs and McGarrell (2004, p. 8) define supply chain security management as
“The application of policies, procedures, and technology to protect supply chain assets
(product, facilities, equipment, information, and personnel) from thefi, damage, or
terrorism, and to prevent the introduction of unauthorized contraband, people, or weapons
of mass destruction into the supply chain.” A number of observations can be drawn from
this definition. First, supply chain assets are defined as not only the equipment and
facilities used to carry out supply chain processes, but also the product, information, and
human resources required to operate the supply chain. Therefore, supply chain protection
does not stop with securing a facility through gates and locks, but also encompasses the

product and people involved in supply chain activities, as well as the internal and external

information flows between supply chain partners and functional silos. Second, supply
chain security is not Simply a matter of ensuring the safety of these assets, but also
preventing theft, damage, and unintended intrusions into their physical domains for the
purpose of preventing disruption of supply chain operations and maintaining acceptable

performance levels.

THE IMPORTANCE OF FOOD AND FOOD SUPPLY CHAIN SECURITY

Food supply chains are an important subset of the broader supply chain concept
due to their importance to the populace, the US. economy, and their vulnerability. A
safe food supply is a primary foundation upon which society is built. Agricultural
products, and their related industries, significantly contribute to the economic welfare of
the United States through jobs and exports (Rand Corporation, 2003). Furthermore, of
seven general areas of US. vulnerability to terrorism, five are related to the food supply
chain (Harl, 2002). This level of vulnerability, coupled with the lack of US. food supply
chain preparedness and potentially disastrous effects of such an attack (Rand
Corporation, 2003), has led to the United States Food and Drug Administration (2003) to
classify the potential for a terrorist attack on the US. food supply as a real and current
threat.

The importance and vulnerability of food supply chains has not gone unnoticed by
terrorists. The United States has recently uncovered Al Qaeda plans to attack crops,
livestock, and food processing operations (U SFDA, 2003). To the extent that the United
States has obtained credible evidence that the food supply is a potential target, it is

important that efforts be made to secure the food supply from agro-terrorism. The World

Health Organization (2002, p. 4) defines agro-terrorism as, “An act or threat of deliberate
contamination of food for human consumption with chemical, biological, or radio nuclear
agents for the purpose of causing injury or death to civilian populations and/or disrupting
social, economic, or political stability.” Drawing from Closs and McGarrell (2004), the
definition of food supply chain security to be used throughout this research is given as:
The application of policies, procedures, and technology to protect food supply chain

assets from theft, damage, or terrorism, and to prevent the introduction of unauthorized
contraband, people, or weapons of mass destruction into the food supply chain.

IMPACT OF SECURITY ON SUPPLIER EVALUATIONS

AS previously noted, one of the factors making the supply chain vulnerable to
attack is the limitation in the amount of funding available to pay for security. It is
estimated that the US. economy incurs an extra annual cost of $151 billion to fund these
initiatives (Russell and Sandalha, 2003). Motor and air carriers alone pay $2 billion for
locks, cameras, employee training, among other expenses (Wolfe, 2001). The inevitable
question arises: Who should pay for security? EyeforTransport (2004) conducted a
survey of supply chain professionals to answer this question. Four possible groups were
identified: shippers, consignees, transportation providers, and the government. In the
short-term, the answer seems to be all of the above, but fifty-five percent of the
respondents indicated that they are bearing the costs internally with little hope of
recovering them. The practice of bearing costs internally likely cannot continue forever.
In the long-term, the cost of security is likely to be passed down to the customer.

Cost is not the only issue. Supplier delivery reliability also stands to be affected

by increased security (Harmon, 2004; J orgensen 2003). Security inspections occurring

inside the firm, at border crossings, and points of embarkation and disembarkation, add
another layer of activities to the order cycle. With any added activity comes greater
process variability. This added order cycle variability decreases suppliers’ ability to meet
promised delivery dates.

It has been proposed that security may actually decrease costs and increase
delivery reliability. These improvements will likely come from the use of tracking
technologies, such as RFID and smart-seals, and process improvements (Rice and Spayd,
2005). Most firms have not progressed beyond implementation of physical security
measures (e. g. gates, guards, and cameras) and have not garnered the advantages that
may come from tracking technologies and process improvements (EyeforTransport,
2005). While physical security measures are valuable, they hold little promise for
lowering costs and improving delivery reliability (Wolfe, 2005).

As noted above, certain security measures could lead to improvements in supplier
performance. However, Since most firms have not progressed beyond the
implementation of physical security measures, the end result is that suppliers must now
shift their focus from delivering required service at the lowest total cost, to accomplishing
this task under new security initiatives. This is a difficult proposition. A supplier that
increases its security capability is faced with the unpleasant notion of increasing the
product price and decreasing the reliability of product delivery. As a result, customer
evaluations of a supplier on these criteria could be negatively impacted. Complicating
the problem, price and delivery reliability are two of the most important criteria used by
firms to evaluate suppliers (Lehman and O’Shaughnessy, 1974; Evans, 1982; Wilson,

1994). Suppliers are also faced with a lack of knowledge concerning how much price

and delivery reliability their customers might be willing to sacrifice for increased security
and under what conditions these customers might be more willing to make these trade-
offs.
RESEARCH OBJECTIVES

This research explores the extent to which food purchasing managers are willing
to trade-off price and delivery reliability for increased supplier security. These findings
provide practitioners empirical evidence to support security related change management
initiatives by increasing awareness of the trade-offs their customers are willing to accept
for increased security. This study also provides academics insight into the role of
security in relation to important supplier selection criteria. The specific objectives of this
research are:

A) To (re) establish how product quality, product price, delivery reliability, and
supplier security influence supplier preference.

B) To determine if purchasing firms are willing to trade-off price and delivery
reliability for increased supplier security.

C) To identify if 1) concern over past security related incidents and 2) international

sourcing increase the amount of price and delivery reliability purchasing firms are
willing to trade-off for increased supplier security.

RESEARCH QUESTIONS
These research objectives result in a series of specific research questions. The first
set of research questions examines purchasing firms’ willingness to sacrifice price and
delivery reliability for increased supplier security. The answers to research questions

posed in section “A” are obvious. Firms are unlikely to prefer suppliers with lower

10

quality, higher prices, lower delivery, and lower security ceritus paribus. However, these
research questions are necessary to set the stage for those posed in sections “B” and “”.C

A) To (re) establish how product quality, product price, delivery reliability, and
supplier security influence supplier preference.

1. Do purchasing firms prefer suppliers who deliver higher product quality
over those that delivery lower product quality?

2. Do purchasing firms prefer suppliers who charge lower prices over those
that charge higher prices?

3. Do purchasing firms prefer suppliers who provide higher delivery
reliability over those that provide lower delivery reliability?

4. Do purchasing firms prefer suppliers who provide greater security over
those that provide less security?

B) To determine if purchasing firms are willing to trade-off price and delivery
reliability for increased supplier security.

1. Are purchasing firms willing to pay a price premium for increased
supplier security?

2. Are purchasing firms willing to sacrifice delivery reliability for increased
supplier security?

The second set of research questions builds upon the findings of questions 8.1 and
B2. These questions examine two conditions that could affect purchasing firms’
willingness to sacrifice for increased supplier security.

C) To identify if 1) concern over past security related incidents and 2) international
sourcing increase the amount of price and delivery reliability purchasing firms are
willing to trade-off for increased supplier security.

1. Does concern over security related incidents that have occurred at the
purchasing firm influence respondents to place more value on increased

supplier security?

2. Do purchasing firms that source goods internationally place more value on
increased supplier security?

11

SCOPE OF RESEARCH

This research surveys food purchasing managers and the trade-offs they are willing to
make to receive increased supplier security. The food industry represents a particularly
important target population for many reasons. First, food is of great importance
important to society and the United States economy. Providing information to food
suppliers regarding the trade-offs their customers are willing to accept for increased
security could provide the impetus for those suppliers to begin a security initiative. This,
in turn, helps to protect society and the economy from attack. Second, perhaps more than
most industries, the food industry has a vested interest in ensuring their suppliers’ product
is prepared and processed safely. Consumers are particularly averse to consuming food
prepared by firms who have experienced a contamination incident and may engage in
litigious actions against companies involved in a contamination incident. It is therefore
in the purchasing firm’s best interest that their suppliers operate in a secure manner.
Third, food protection has recently received media attention, which makes it a timely
topic. Perhaps more importantly, this media attention may have increased the likelihood
of the food supply becoming a target.

Interviews are to be conducted with food industry representatives to identify how
much price and delivery reliability they are willing to trade-off for increased supplier
security. This information is used to create the scenarios presented to respondents in a
survey instrument. These surveys were sent to food industry purchasing managers. Each
respondent was then asked to evaluate a set of scenarios. These scenarios present
different levels of the factors under investigation, including basic and advanced supplier

security. A vignette describing a supplier employing basic security measures, and one

12

describing a supplier employing advanced security measures, were provided with the
survey. The basic and advanced security characteristics are drawn from Closs and
McGarrell (2004).

The sample of firms was drawn from the Institute for Supply Management
(http://www.napm.org) and from the American Purchasing Society
(http://www.american-purchasing.com). Choice-based conjoint analysis was used to
analyze the factors. Further discussion of methodological issues can be found in Chapter

Three.

CONTRIBUTIONS TO THEORY AND MANAGEMENT PRACTICE

This project represents one of the few empirical works investigating supply chain
security, and therefore stands to be of import to future academic research. Further, the
results provide academics insight into the importance of security as it relates to price and
delivery reliability, two important supplier selection criteria. The results are also useful
to practitioners to evaluate if their firm should invest in supply chain security measures.
It is often difficult for practitioners to justify investments in the absence of government
regulation or evidence supporting that the initiative is beneficial. This research provides
supporting, or dissenting, evidence related to a supplier’s ability to reap rewards from
security implementation. The results of the research are also of interest to public policy
makers. The results allow public policy makers to determine if regulation is necessary to
induce firms to increase security, or if there is a demand for increased supplier security.

In the latter case, the free market could substitute for burdensome regulation.

13

RESEARCH ORGANIZATION

The remainder of this research is organized into four chapters. Chapter Two
examines the relevant literature with respect to supply chain security in general, and
particularly protection of the food supply chain. Chapter Three provides an in-depth
discussion of the research methodology, reviews the research objectives, outlines data
collection methods, and explains the analytical methods applied. Chapter Four presents
the research results and interprets the findings of each research question. Finally,
Chapter Five explains the conclusions drawn from these results and the implications for
academia, industry, and government. The final chapter also includes limitations and

directions for future research.

14

CHAPTER 2
INTRODUCTION
Chapter Two synthesizes the supply chain security literature. First, the chapter
overviews supply chain security. Second, the chapter focuses on food and food supply
chain security. Third, the chapter reviews the criteria used by firms to choose suppliers.
Finally, the chapter integrates these literature streams and identifies the issues that are the

focus of the current research.

OVERVIEW OF SUPPLY CHAIN SECURITY

Supply chain security has become a priority in the post-9/11 world. Firms are
faced with the challenge of protecting complex supply chains from terrorist attack.
Terrorists are commonly thought of as part of an organized group, which may not always
be the case. Disgruntled employees or disenfranchised individuals are also a threat.
These employees often have access to sensitive areas in the firm and could sabotage or
contaminate product. Therefore, supply chain security is not only a matter of protecting
assets from external threats, but also from the enemy within. The term terrorist is used
throughout this research to imply any individual who would seek to do harm to others.
No distinction is made between terrorists that are a part of an organized group or those
that act alone.

Furthermore, supply chain security Should not be confused with risk. Risk is
context specific in that different industries define risk in different ways (Spekman and
Davis, 2004). An accountant may think of risk as the inability to detect irregularities in.

audit data, whereas a logistician might View risk as the possibility that inventory could

15

become obsolete. Johnson (2001) characterizes supply chain risk as consisting of two
facets: demand risk and supply risk. Demand risk occurs as a result of factors that make
demand unpredictable or might negatively skew demand. These factors may include
seasonality, volatility of fads, Short product life cycles, or the public’s perception of the
safety and ethics of the firm. Supply risk involves, “the probability of an incident
associated with inbound supply from individual supplier failures or the supply market
occurring, in which its outcomes result in the inability of the purchasing firm to meet
customer demand or cause threats to customers life and safety” (Zsidisian, 2003 p. 222).
Supply chain security management, as previously defined, serves as a way to mitigate
demand and supply risk. Security mitigates demand risk by helping to protect a fimr
from terrorist activity and thereby protect its brand image from harm in the eyes of the
public. Security mitigates supply risk in that it helps to protect the safety of a firm’s
customers and also helps to ensure supply to that firm by improving the quality of goods
delivered.

The previous discussion of risk also allows one to differentiate between the
concepts of security, protection, and defense. In this research, security is conceptualized
as a method of protecting the supply chain by defending against terrorist induced demand
and supply risk. Said differently, security protects supply chain assets by defending
against external threats. To put this in a more understandable context, think of a football
offensive lineman as supply chain security and the opposing team as potential terrorists.
In passing situations, the job of the offensive lineman is to protect the quarterback by

defending against the threat posed by the opposing team. Therefore, security protects by

16

defending. In this sense, security is a method of protecting “internal” assets by defending

against “external” threats.

PRESSURES LEADING TO SECURITY IMPLEMENTATION

The terrorist threat has increased pressure on firms to improve their supply chain
security capability. The threat of future terrorist acts is the primary driver of security
initiatives, but secondary drivers that occur as a result of this threat are important as well.
The Aberdeen Group (2004) and EyeforTransport (2004) conducted research into these
secondary drivers. Respondents were asked what drove their firms’ decision to enhance

security. The results are presented in Table 2.1:

TABLE 2.1 DRIVERS OF SECURITY INITIATIVES

 

Study Drivers

 

Aberdeen Group (2004) Brand damage from product recall or regulatory non-
compliance (i.e. protect brand equity)

 

Brand piracy, gray market activity, and product counterfeiting

 

Customer and trading partner requirements

 

Increased product safety and traceability concerns due to
outsourcing activity

 

EyeforTransport (2004) Customer requirements

 

 

Government pressure

 

 

The Aberdeen Research Group (2004) respondents reported their fimrs’ primarily
enhanced security to protect brand equity. Practitioners fear that contamination of their

product will damage customer perceptions of their brand. An example of this can be

17

 

found in the recent fast food contamination in California. In this case, a well-known fast
food franchise suffered a loss in business due to a human digit being found in their food.
This incident later turned out to be a hoax, but the company lost brand equity that
required expensive recovery efforts.

Related to brand equity is the issue of brand piracy, gray markets, and product
counterfeiting. Firms secure supply chain assets in order to reduce theft and provide
assurance to their customers of product origin. This issue has come to the forefront in
recent discussions over senior citizen’s ability to purchase pharmaceuticals from Canada.
The Food and Drug Administration (FDA) claims that unregulated pharmaceutical
imports stand a greater chance of being ineffective or harmful.

Customer security requirements are cited as a driver of security programs in both
studies. To protect themselves, firms have begun requiring their suppliers to increase
security. Working with suppliers and customers to ensure their supply chains are secure
is a primary tenet of the Customs-Trade Partnership Against Terrorism (C-TPAT). C-
TPAT is a voluntary, joint public-private initiative encouraging firms to secure their
supply chains, and those of their trading partners. The goal of C-TPAT is to certify
enough firms to create a critical mass of supply chain protection.

Government pressure is another significant driver. While C-TPAT is voluntary,
the government has also taken a regulatory stance. The Bioterrorism Act of 2002
requires that firms engaged in food processing be able to trace raw materials and output
one step up, and one step down, the supply chain (USFDA, 2003).

These drivers are prompting firms to be proactive in establishing their security

programs. Other firms are less prepared. Rogers et al. (2004) report that many managers

18

feel their firm will not be the target of a terrorist attack because they only sell certain
commodities. Other managers feel their security obligations end when they transfer
goods to a carrier. Both of these findings indicate that supply chain security is not a

priority for many firms.

THE STATE OF SUPPLY CHAIN SECURITY BEST PRACTICE

Despite the previously mentioned incentives for supply chain security, effective
security processes are still quite rare. Few academic works provide guidance into the
most efficient and effective means to secure a supply chain and even fewer are empirical.
The two primary academic works providing guiding recommendations are Closs and
McGarrell (2004) and Rice and Caniato (2003). A summary of their findings can be

found in Tables 2.2 and 2.3 below:

19

TABLE 2.2 SUPPLY CHAIN SECURITY BEST PRACTICE I

 

 

 

 

 

 

 

Relationships Internal Security Incident Security
Initiatives Management
Suppliers Personnel Planning
Government Information Mitigation
Carriers Facility(ies) Detection
Terminal/Port operators Inventory Response
Customers Transportation Recovery
Transportation tracking
and visibility

 

Receiving management

 

Storage management

 

Shipping management

 

Internal operations
management

 

Supply chain education

 

 

 

 

 

Source: Class and McGarrell (2004)

20

TABLE 2.3 SUPPLY CHAIN SECURITY BEST PRACTICE II

 

 

 

passwords

Physical Security Information Security Frflght Security
Background checks Hardware: firewalls, Inspections
dedicated networks,
etc.
Vulnerability checks Software: intrusion Compliance with US. government
by outside experts detection, anti-viruses, initiatives (C-TPAT, CS1, OSC,

etc...)

 

Access control,
badges

Audits of partners IS
security

Cargo seals

 

Gates, guards, camera
systems

Education and training
for IS security

Documented standards of care, use
of certified 3rd parties, defined chain
of custody

 

Industry initiatives to establish
standards of care among shippers
and carriers

 

 

 

 

Use of technology (GPS, RF 1D, e-
seals, biometrics, etc...)

 

Source: Rice and Caniato (2003)

Key to the discussion of supply chain security is the notion that a supply chain

protection program is only as strong as its weakest link. Sheffi (2001) posits

collaboration with external entities is necessary to ensure that security procedures are

communicated and followed. Closs and McGarrell (2004) posit five general stakeholders

with whom a firm should maintain relationships and communicate security information.

For example, firms should maintain collaboratory relationships with the government in

order to remain abreast of security regulations. The government also serves as a source

of threat information and provides guidance regarding methods to secure the supply chain

through certification programs such as C-TPAT.

21

 

Maintaining these relationships are necessary for firms to effectively implement
the disaster management process (labeled “incident security management” in Table 2.2).
The disaster management process consists of five stages and documents the actions that a
firm will undertake before, during, and afier a security incident (Helferich and Cook,
2003). Planning represents the firm’s efforts to formulate actions in anticipation of an
incident. Mitigation is the action taken prior to an incident that prevents or reduces
supply chain damage. Examples of mitigating actions include instituting parallel
production processes in case one is contaminated, or forming an emergency response
center to serve as a central point of contact. Detection is the firm’s ability to recognize an
incident. While incidents are sometimes easy to detect, such as those of 9/1 1, others are
not (National Academy of Sciences, 2004). Contamination of food products with a
biological agent may be mistaken for a batch of Spoiled product. The real challenge of
detection is to ascertain that an incident has occurred prior to it doing any harm. The
response stage begins as soon as the incident is detected and involves mobilizing
equipment to respond to an emergency, removing people from danger, providing for
those affected by the incident, and bringing affected services and systems back on-line.
Response represents the firm’s short-term actions following an incident. The long-term
efforts fall under recovery. Recovery often places the most strain on involved parties.
Long-term effects of morbidity and mortality can be devastating to families, financial
institutions, firms involved in the attack, and the government. The long-term effects of
9/11 on the airline industry, and the US. Government’s effort to assist them through

expensive loan guarantees, serve as an example.

22

The actual actions that must be taken to secure the supply chain, and entities that
must be secured, are given under internal security initiatives in Table 2.2. The Closs and
McGarrell (2004) model presented in Table 2.2 is more prescriptive in this respect when
taken at face value. Steps to assess supply chain security are provided as an appendix in
their work. This appendix also delineates characteristics of firms who have achieved
basic and advanced levels of security implementation. The Rice and Caniato (2003)
model presented in Table 2.3 provides more in the way of actual security measures, but is
not as thorough as the Closs and McGarrell (2004) appendix. Detailing the actual
methods of securing the supply chain is beyond the scope of this work. Many resources,
both private and public, provide suggestions (e. g. www.dhs. gov). Identifying what
actually constitutes best practice is a matter of ongoing research. Such research is being
conducted by Michigan State University, where investigations into food supply chain
security best practice are ongoing (http://www.fpd.umn.edu). The preliminary results of

this investigation will be discussed subsequently.

COST AND SERVICE IMPLICATIONS OF SECURITY IMPLEMENTATION
The current research is grounded in the notion that firms who endeavor to secure
their supply chain incur a cost and service penalty. For example, Closs and McGarrell
(2004) advocate securing the receiving department through cargo inspections. While this
secures the firm’s receiving function, it also adds cost. Receiving personnel must be
dedicated to (at least part-time) performing the inspections. Further, the time it takes to

perform the inspection adds greater variability to the receiving activity. This adds greater

23

variability to the firm’s entire process. This added variability slows the process and
makes it more difficult to predict delivery dates, ultimately decreasing service levels.

AS stated in Chapter One, the US. economy will incur $151 billion annually in
additional costs to secure the supply chain (Russell and Saldanha, 2003). From a macro-
level, these expenditures reallocate public and private resources away from investments
that would improve supply chain efficiency and effectiveness, to those that ensure safety
and security (Phillips, 2001). This leads to higher long-term costs for shippers.

From a more micro-perspective, the cost and service penalties incurred by US.
businesses are the result of conflicts between security and business goals. Sheffi (2001)
posits the following ways in which this occurs (see Figure 2.1). Note that the security
goals and business goals presented in Figure 2.1 are not necessarily ends of a continuum
or mutually exclusive. This figure is meant to simply convey that trade-offs may be

necessary in order to implement supply chain security measures.

24

FIGURE 2.1 THE CONFLICT BETWEEN BUSINESS AND SECURITY
PRIORITIES

Business Goals ———* CONFLICT WITH ———'> Security Goals

 

 

 

 

Repeatability Unpredictability

 

 

 

 

 

 

Firms try to make tasks repeatable in order to perform them more efficiently.
Security requires that some tasks be unpredictable (e.g. changes in HAZMAT
routes and passwords)

 

 

‘

 

Lowest Bidder Known Supplier

 

 

 

 

 

 

Firms will choose a known supplier over the lowest bidder to enhance security.
This will force them to incur the opportunity cost of not selecting a supplier that
may have lower prices, new ideas, or innovative processes.

 

 

 

Centralization Dispersion

 

 

 

 

 

 

Firms centralize activities to gain economies of scale. Security demands activities
be dispersed so they are at less risk.

 

 

 

Delivering Value ' Managing Risk

 

 

 

 

 

 

It is hard to justify investments in security on the value they deliver to customers,
employees, or Shareholders, and will be hard to justify in the foreseeable future.

 

 

Collaboration Secrecy

 

 

 

 

 

 

 

Collaboration is necessary to decrease supply chain costs. Security requires more
caution in Sharing information.

 

 

 

Efficiency Redundancy

 

 

 

 

 

 

Firms usually try to reduce redundancy and increase efficiency, but security
requires redundancy.

 

 

Direct Shareholder Government
Value Cooperation

 

 

 

 

 

 

 

Firms are conditioned to put a priority on shareholder value, but security may
require cooperation with external entities (e. g. government) at the expense of
shareholder value.

Source: Sheffi (2001)

25

There are a number of other mechanisms by which security investments can
increase costs. For example, the cost of security to warehouses is estimated to be $1 to
$2 per square foot of storage space (Warehouse Education Research Council, 2004).

This includes the cost of installing cameras and fences, training personnel, and dedicating
labor to the security function. Motor carriers must incur the cost of locks for trailers,
installation of these locks, added labor, extra administration, key distribution, and backup
systems. These costs could reach $150 per trailer (Kahaner, 2003). For motor carriers to
become C-TPAT compliant, they must also utilize remote camera systems, electronic
gates, and security guards at storage facilities. The cost of screening and training
personnel alone could reach $50 per driver (Cullen and Kilcarr, 2004).

There exist cost and service implications for international trade as well. One facet
of the US. Government’s plan to increase security is to increase container inspections at
US. and foreign ports. The Container Security Initiative (CS1) seeks to extend the zone
of security surrounding the United States to the foreign ports where cargo is inspected
before being loaded onto a vessel. The cost of these inspections is home by the importer
and can cost $125 - $250 if the inspection is conducted on-site. If conducted off-site, the
price increases to $600 - $900 (Jorgensen, 2003). These inspections also cause delays,
which adds greater variability to the process and can potentially cause the container to
miss its sailing date, with little or no recourse available for the shipper (Harmon, 2004).

Several observations can be drawn from the supply chain security literature. First,
firms are motivated by many factors to improve their security capability. These include a
desire to protect their brand equity and to meet customer demands. Second, relatively

little has been published with regard to the most effective means to secure the supply

26

chain. Finally, firms that increase their security capability stand to incur greater costs and
decrease their service capability. Sheffi (2001) implies that cost and service penalties are
an inherent part of security initiatives, but does not examine if a market exists for these
initiatives. Suppliers cannot be expected to engage in security initiatives if they stand to
lose business because of these penalties. Purchasing firms should be willing to trade-off
price and reliability of delivery if they expect to obtain higher security from their
suppliers. This will be explored further subsequently. The next section will describe the

role of food and food supply chain security in the context of the previous discussion.

FOOD AND FOOD SUPPLY CHAIN SECURITY

As with the broader topic of supply chain security, food supply chain security has
received little in the way of empirical research. What has been published primarily deals
with the vulnerabilities of food, how it may be attacked or used as a weapon, and various
government recommendations pertaining to securing the food supply. This section will
begin with a general overview of food and where it is vulnerable. Following this
discussion, the Michigan State University food supply chain security best practice
research is provided.

Food is defined by the United States Food and Drug Administration (2004) as
articles used for food or drink for man or other animals, chewing gum, or articles used for
components of other foods. As described in Chapter 1, food is a likely target for
terrorism due to its importance to society, vulnerability, and terrorist awareness of these.
Meriwether and Goodell, (2004a, p. 1) define food security as, “The prevention or

elimination of deliberate contamination of food [by method of]. . .chemical, physical,

27

nuclear, or radioactive [agents].” The definition of food supply chain security used in the
current research is adapted from Closs and McGarrell (2004) and given as:

The application of policies, procedures, and technology to protect food supply chain
assets from theft, damage, or terrorism, and to prevent the introduction of unauthorized
contraband, people, or weapons of mass destruction into the food chain.

Different types of food require different amounts of processing (World Health
Organization, 2002). This compounds the difficulty of protecting the food supply
because the production system and steps vulnerable to attack will be different for each
commodity (World Health Organization, 2002). Examples of the supply chains for

livestock (hoof to plate) and grain (seed to plate) are provided in Figures 2.2 and 2.3

(Cupp et al. 2004):

FIGURE 2.2 HOOF TO PLATE SUPPLY CHAIN

 

Farm

 

 

 

Livestock Poultry Eggs

Sale Barn
I
Feed Lot
I r
Slaughter House

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Processing Facility r Rendering

I
Distributor/Warehouse Edible

I

Wholesaler r Retailer *7

 

 

A

 

V

 

 

 

 

 

 

 

 

 

 

Consumer

 

 

 

 

 

 

 

 

Source: Cupp et al. (2004)

28

 

FIGURE 2.3 SEED TO PLATE SUPPLY CHAIN

 

Farm
I

Coop/Buyer Elevator
F

 

 

 

 

 

Storage Elevator
T

 

 

 

 

 

 

Mill/Processing Facility Exporters/Whole Grain
I

 

 

 

 

Milled/Processed

Y

 

 

 

 

 

 

 

Human Foodstuffs 7 Animal Feed ExportS

 

 

 

 

T
Wholesaler

r

Retailer

I

Consumers
Source: Cupp et al. (2004)

 

 

 

 

 

 

 

 

 

 

 

 

FOOD VULNERABILITY
To secure the food supply chain one must examine where it is most vulnerable.

The literature provides us some guidance in that respect (see Table 2.4):

29

TABLE 2.4 AREAS OF FOOD VULNERABILITY

 

Author

Vulnerability

 

Bledsoe and Rasco (2002)

Targets include corporations, universities,
and government research facilities

 

Bruemmer (2003)

Perishable foods are a likely target as they
require little processing and their lower
shelf life means they are less likely to be
recalled in time

 

Chalk (2001)

Small food processors

 

Cremin (2001)

Small to medium sized firms

 

Products that don’t require cooking or
extensive preparation

 

Products that will elicit an emotional
response (e. g. baby food)

 

Gips (2003)

Any area that is highly accessible from
multiple modes (e.g. air, road, and rail)

 

Bulk milk tanks are very vulnerable to
contamination b/c they combine the milk of
many cows and one infected cow could
contaminate the whole tank

 

Nganje et al. (2003)

High volume grain elevators

 

Strozniack (2004)

Average, medium sized firms are

 

Wheelis et al. (2002)

Farms/ranches next to international borders

 

 

World Health Organization (2002)

Slaughterhouses

 

Bulk foods

 

The most readily accessible food processes

 

Foods that are most vulnerable to
undetected tampering

 

Foods that are the most widely
disseminated

 

 

The least supervised food production areas
or processes

 

Synthesizing the information in Table 2.4 uncovers several common themes.

First, accessibility is an issue. Accessibility is a factor because it makes securing a

location more difficult. The more avenues through which a location is accessible, the

more vulnerable that location. For example, a food manufacturer accessible by three

30

 

modes of transportation would be more vulnerable than one only accessible by two. A
location accessible by fewer modes is easier to control.

The second issue contributing to vulnerability is consolidation. Consolidation is a
concern for a number of reasons. First, livestock and grain are regularly stored in large
quantities. This allows a terrorist to contaminate large quantities of product in a single
attack. Second, Should a terrorist contaminate a single shipment pre-consolidation, this
contamination stands to contaminate a large quantity of product post—consolidation. Bulk
milk is provided as an example by Gips (2003). The milk from many cows is combined
into a bulk milk tank during shipment. Contaminated milk from one cow could then
contaminate the entire shipment. This could cause morbidity or mortality in a large
number of consumers with little terrorist effort.

Third, the level of security protecting a location contributes to its vulnerability.
The least supervised areas, firms, or processes, are the most vulnerable. This explains
several authors’ emphasis on the vulnerability of small to medium sized firms (see Table

2.4). These firms are perceived to be less prepared and guarded against intrusion.

FOOD SUPPLY CHAIN SECURITY BEST PRACTICE

Michigan State University, in conjunction with the Department of Homeland
Security, is currently engaged in assessing best practice in protecting food from the
aforementioned vulnerabilities. Through this research, investigators have found there to
be ten competencies, encompassed by four general themes, which make up food supply
chain security best practice. These themes and competencies closely resemble the

security best practices described above in Closs and McGarrell (2004).

31

TABLE 2.5 MSU FOOD SUPPLY CHAIN SECURITY COMPETENCIES AND
CAPABILITIES

Strategy
1. Process Strategy — Executive commitment to fostering an enterprise-wide

environment conducive to security and protection.

Operations

1. Process Management — Procedures and actions taken to ensure the safety of each
activity involved in the product flow into, through, and out of a facility.

2. Infrastructure Management — The presence of hardware and personnel responsible
for ensuring the physical domain of a space is not breached.

3. Metrics and Measurement -— The continuous development, use, testing, and
redefinition of guidelines measuring security related procedures, plans, and
capabilities.

Collaboration
1. Communication Management — The security related intra-firm relationships and
exchange of information.

2. Relationship Management — The security related inter-firm relationships and
exchange of information.

3. Public Interface Management — The security related relationships and exchange of
information with the government and public.

4. Service Provider Collaboration Management — The security related relationships
and exchange of information with 3rd party service providers.

Technology
1. Management Technology — The presence, use, and ability of information systems

to provide accurate, timely, and reliable security information internally, across the
supply chain, to the government, and to the public.

2. Process Technology - The presence, use, and ability of information systems to
track movement of products and monitor processes internally and across the

supply chain.

Source: MSU Food Supply Chain Security Research Team (research in progress)

32

 

The literature relating to food supply chain security reveals several themes. First,
food is a likely target for terrorists. Second, food supply chains are vulnerable for
reasons including accessibility, consolidation, and lack of preparedness. Finally, the
Michigan State University research team is uncovering what constitutes food supply
chain security best practice. No works are available that examine the trade-offs food
manufacturers might be willing to make in order to receive security from their suppliers.
Furthermore, no literature exists detailing the attitudes of food manufacturers toward

security.

SUPPLIER SELECTION CRITERIA

Suppliers are a crucial part of firm success (Porter, 1985). Supplier selection is
important because it commits resources while Simultaneously impacting activities such as
inventory management, production planning and control, cash flow requirements, and
product quality (Choi and Hartley, 1996). There are two steps involved in selecting
suppliers (Braglia and Petroni, 2000). First, suppliers are evaluated to ascertain if they
meet certain conditions, or choice criteria, set forth a priori. Second, suppliers that meet
or exceed these choice criteria are selected. Choice criteria are defined as “The factor(s)
[a buyer] uses to evaluate competitive offerings” (Lehmann and O’Shaughnessy, 1974, p.
36). Suppliers can be evaluated using many different criteria. The four most important
are product price, delivery reliability, product quality, and service after-the-sale (Braglia
and Petroni, 2000). Price is the economic outlay necessary for the purchase of the
product. Quality is defined as the extent to which the supplier’s product conforms to

buyer specifications. Delivery reliability is the number of times the supplier’s product is

33

available for use by a promised date. Service after-the-sale is defined as the support
afforded by the supplier, to the buyer, to resolve issues pertaining to a past product
purchase. Rankings of these criteria from prominent supplier selection studies are
subsequently provided in Table 2.6. It should be noted that these works drew from
different samples and are meant to only provide a generalized view of the importance of
these supplier selection criteria.

TABLE 2.6 IMPORTANCE OF SUPPLIER CHOICE CRITERIA ACROSS
STUDIES

 

 

 

 

 

 

 

 

 

Authors Importance Rank of Selection Attribute
Price QualityI Delivery Reliability Service

Lehmann and 2 3 1 4
O’Shaughnessy (1974)

Evans (1982) 2 3 1 4
Lehmann and 2 1 4 3
O’Shaughnessy (1982)

Wilson (1993) 3 l 4 2
Min (1994) 5 l 2 12
Choi and Hartley (1996) 6 l 1 7

 

 

 

 

 

 

This research is only concerned with the trade-off of price and delivery reliability
as they relate to supplier security. There is little theoretical justification to support that an
inverse relationship exists between supplier security and product quality, or supplier
security and service after-the-sale. Security initiatives may actually increase quality
through reductions in lost, damaged, or destroyed product. Therefore, quality is included
as a control variable in this study. No trade-off can be theoretically justified between
service after-the-sale and security. Service after-the-sale is important, but not currently
under investigation.

Several conclusions about product price and delivery reliability can be drawn

from Table 2.6. First, delivery reliability seems to have progressed from being very

34

important as a supplier choice criterion, to less important, to very important. This is
likely due to the growing importance of J ust-In-Time (J IT) inventory policies in the early
to mid 1990’s. Second, price has waned in its importance since the 1970’s. This
phenomenon is most likely due to the growing emphasis placed on product quality and
collaboration with suppliers as opposed to adversarial relationships (Bowersox et al.,
2002). Taking the average of the above supplier choice criteria rankings reveals that
delivery reliability is ranked second only to product quality and price follows closely
behind in third place.

There are many reasons delivery reliability is highly ranked. First, delivery
reliability is a very visible supplier attribute. When delivery is unreliable, it can cause
severe economic consequences by shutting down a plant or delaying production
(Lehmann and O’Shaughnessy, 1974). Second, delivery reliability affects the level of
safety stock a firm must maintain. Money Spent on safety stock could be reallocated
elsewhere if the supplier were able to deliver product reliably.

Price is important because it provides the purchasing agent a way to justify
selecting a supplier should a problem with them occur (Lehmann and O’Shaughnessy,
1974). Further, price bears a direct relationship with the firm’s cost of goods sold, which
influences profit margin and the price the firm charges for their goods.

To tie in the previous discussion with the current research, it should be noted that
supplier security was not examined in any of the surveyed choice criteria literature. This
is most likely due to the timeliness of security as a research topic. There does exist
support that security is important to firms when choosing a transportation service

provider. EyeforTransport (2004) surveyed supply chain executives and found that

35

ninety-four percent of respondents think security is a “very important” carrier attribute.
EyeforTransport (2004) did not examine the importance of security when choosing a
product supplier or if the respondents were willing to trade-off price and delivery

reliability for increased security from their transportation or product suppliers.

SYNTHESIS

A survey of the literature above yields a number of lessons. First, the study of
supply chain security, and particularly food supply chain security, is still in its infancy.
While notable contributions have been made (Closs and McGarrell, 2004), ample
opportunity for future research exists. Second, it is clear that security efforts stand to
increase costs and decrease delivery reliability suppliers provide to their customers. This
seems to be especially relevant to international shipments. Third, there is a lack of
knowledge pertaining to the value firms place on their suppliers’ security competency.
Most of the literature focuses on a single firm. The supplier-customer dyad has not been
investigated. No study has investigated the importance of security to the supplier
selection decision. Finally, evidence exists that firms feel pressure to increase security,
but it is unclear how many are actually doing something about it. If firms are taking a
myopic View toward their own vulnerability, do they take this same view toward their
suppliers’ vulnerability? Even if they do recognize their own vulnerability, and that of
their suppliers, is security important enough that firms are willing to trade-off product

price and delivery reliability in order to achieve a more secure supply chain?

36

STATEMENT OF PROBLEM
The synthesis of the literature review suggests that there remains a need for
supply chain security research examining whether buyers are willing to make trade—offs
that are necessary for suppliers to secure their operations. Further, the literature suggests
that there is a need for any sound empirical research pertaining to supply chain security.
The research objectives presented in Chapter One suggest a series of research questions.
These research questions are presented below along with their accompanying hypotheses

and theoretical justifications.

RESEARCH QUESTIONS AND HYPOTHESES

The literature review demonstrates the relevancy of the research questions posed
in Chapter One. The following research questions and hypotheses have drawn from the
literature to pose refutable, yet theoretically grounded, statements to be tested in this
study. The relationships described in the following hypotheses are depicted in Figure 2.4.

The first four research questions pertain to the relationship between product
quality, product price, delivery reliability, and supplier security as they pertain to supplier
preference. Drawing from Lehmann and O’Shaughnessy (1974; 1982), Evans (1982),
Wilson (1993), Min (1994), and Choi and Hartley (1996) it is proposed that higher levels
of product quality and delivery reliability, and lower levels of product price, are
positively associated with supplier preference. These are rather obvious hypotheses, but
are needed to set the stage for future research questions/hypotheses. The Aberdeen
Group (2004) and EyeforTransport (2004) both report that security is important to firms.

Therefore, higher levels of supplier security are proposed to be preferred.

37

 

 

doumooa wEoBom

 

 

 

 

85265
598m no.5 5850

 

 

 

s Because museum 2 an

£A£ em
mafia am

 

 

 

 

 

 

. 8:86.35 Saddam

E U

ram

 

REE

unaccoun—
Essam boas:

 

 

+ fizz/g

L
S
e

 

 

+ N—m# .

 

59

0322:

 

bisoom com—amsm

 

 

 

 

 

 

A; /\

 

a

Ea

03205
moth 838m

 

 

rum

 

 

 

sum—GO: GHNEHEHOE EN gwmm

£

90%

0382:
£30 6:85

 

 

+5

38

 

— I _ __ .——-—-— . .-y—.—.g-.—:‘.."-‘--.& Z-_ -. '.-' . ....- ._... ‘ ~

1. Do purchasing firms prefer suppliers who deliver higher product
quality over those that delivery lower product quality?

H1: Higher product quality is preferred over lower product quality.

2. Do purchasing firms prefer suppliers who charge lower prices over
those that charge higher prices?

H2: Lower product price is preferred over higher product price.

3. Do purchasing firms prefer suppliers who provide higher delivery
reliability over those that provide lower delivery reliability?

H3: Higher delivery reliability is preferred over lower delivery reliability.

4. Do purchasing firms prefer suppliers who provide greater security
over those that provide less security?

H4: Higher supplier security competence is preferred over lower supplier
security competence.

This chapter illustrated several reasons why security implementation could cause
suppliers to raise their prices. These include the cost of gates, guards, locks, seals, and
labor. Research Question Five concerns whether firms are willing to sacrifice price for
their suppliers to increase security.

5. Are purchasing firms willing to pay a price premium for supplier
security?

H5: Product price has a greater influence on supplier preference than
supplier security competence.

Widespread implementation of supply chain security has not occurred due to
many managers’ myopic view about their firm’s vulnerability (Rogers et al., 2004).
Many firms are thinking about security, but may not have implemented a comprehensive
program (Rogers etal., 2004). Further, EyeforTransport (2004) finds that suppliers are

not currently recouping the cost of their security initiatives through price increases passed

39

on to their customers. This leads one to believe that they do not believe that their
customers are willing to pay for security at this time. Eventually these costs will be
passed down if free-market demands for security increase, or if the government mandates
security. For now, it appears that firms are not willing to pay more for their suppliers to
operate in a secure manner, but no conclusive evidence to that effect has been found in
the literature. Therefore, H5 posits that firms are not willing to pay a price premium to
their suppliers in order to receive increased security from them.

This chapter described how suppliers’ capability to provide reliable delivery
service could be adversely affected by their security initiatives. Research Question Six
concerns whether firms are willing to sacrifice delivery reliability for their suppliers to
operate in a secure manner.

6. Are purchasing firms willing to sacrifice delivery reliability for
increased supplier security?

H6: Delivery reliability has a greater influence on supplier preference than
supplier security competence.

It is not believed that firms are currently willing to sacrifice delivery reliability in
return for increased supplier security. The supplier choice criteria literature posits that
delivery reliability is more important than price. Evidence exists that firms are not
willing to sacrifice price for their suppliers to increase security (EyeforTransport, 2004).
Delivery reliability is more important than price (Lehmann and O’Shaughness, 1974;
Evans, 1982; Choi and Hartley, 1996; Min, 1994) because price increases can be passed
on to customers whereas low levels of delivery reliability can lead to the idling of
expensive production equipment. Logically then, suppliers would not be willing to

sacrifice delivery reliability for their suppliers to increase security. Therefore, H6 posits

40

that firms are not willing to sacrifice delivery reliability in order to receive increased
security from their suppliers.

The previous research questions operated under the assumption that no potentially
confounding influences moderated the amount of price or delivery reliability firms are
willing to trade-off in return for increased supplier security. Research questions Seven
and Eight present two potential moderators to these relationships.

7. Does concern over security related incidents that have occurred at the
purchasing firm influence respondents to place more value on
supplier security?

H7: An increase in concern over security incidents experienced by the

purchasing firm positively affects preference for suppliers with higher
security competence.

Hg: An increase in concern over security incidents experienced by the
purchasing firm positively affects preference for suppliers with lower
delivery reliability and higher security.

H9: An increase in concern over security incidents experienced by the
purchasing firm positively affects preference for suppliers with higher
prices and higher security.

Research Question Seven concerns whether respondent concern over their firms’
past security incidents are more willing to trade-off price and delivery reliability for
increased supplier security. Rogers et al. (2004) Show that many managers are unwilling
to accept that their supply chain may be the target of a terrorist attack. This myopic, “it
could never happen to me” view of security is likely to change should they experience a
security event. Preliminary interviews with industry executives conducted for the
Michigan State University food supply chain security best practice research have
provided support for H7, H3, and H9. One executive stated that security was not a

concern to his firm until an employee contaminated a vat of product with smokeless

tobacco. This ruined the product and forced the company to realize that the employee

41

could have placed a more serious contaminant into the product causing morbidity or
mortality in their customers. After this incident, the firm began a security program and
also began auditing their suppliers’ security measures. The notion of a security incident
providing food manufacturers incentive to improve their security has also been cited by
others (Hulme 2005). Hulme (2005) reports the case of a food processor who had not
considered security a security event occurred. This incident caused the processor to
implement security measures to protect against possible intrusion. Therefore, H7, Hg, and
H9 posit that firms who have experienced a security incident place more value on supplier
security.

8. Do purchasing firms that source goods internationally place more
value on increased supplier security?

H10: International sourcing by the purchasing firm positively affects
preference for suppliers with higher security competence.

H11: International sourcing by the purchasing firm positively affects
preference for suppliers with lower delivery reliability and higher security.

H12: International sourcing by the purchasing firm positively affects
preference for suppliers with higher prices and higher security.

International sourcing presents added security difficulties for firms. While the
product is being produced (i.e. harvested, mined, manufactured, etc...) overseas, the firm
is unable to monitor their supplier and protect against theft, contamination, or insertion of
unauthorized counterfeit cargo. The Aberdeen Group (2004) identified several reasons
firms’ initiate security programs (see Table 2.1). Two of these drivers are increased
product and traceability concerns due to outsourcing activity, and brand piracy, gray
market activity, and product counterfeiting. Due to purchasing firm’s decreased ability to
monitor their international suppliers and shipments it is likely that the purchasing fimr

will be more apt to sacrifice in order to protect themselves from the concerns raised by

42

The Aberdeen Group (2004) respondents. Sheffi (2001) also suggests that security
concerns will cause buyers to move from purchasing product from the lowest bidder to
purchasing product from a known supplier. In turn, they stand to incur the opportunity
cost of utilizing a supplier lower cost or higher delivery reliability. H10, H1 1, and H12
therefore posit that to assure firms of product origin, they will place more value on
supplier security.
CONCLUSION

Chapter Two presents a literature review related to the current study. This review
examined previous research pertaining to supply chain security. Chapter Two also
reviewed two generalized food supply chains and discussed the reasons food is
vulnerable to terrorist attack. This Chapter further examined supplier selection criteria as
they relate to the current research. Following a synthesis of the literature, the research
problem was formalized. Research questions and hypotheses were subsequently

provided. Chapter Three describes the methods used to test the hypotheses.

43

CHAPTER 3
INTRODUCTION

Chapter Three describes the research design and methods used to test the
hypotheses. First, this chapter begins with a review of the research purpose and
objectives. Second, the chapter outlines the research methodology including elaboration
on the unit of analysis, sampling procedure, data collection and implementation,
instrumentation, and the data analysis procedures. Finally, the chapter concludes by
bringing the research proposal to a close and directs attention toward results and

implications.

RESEARCH PURPOSE AND OBJECTIVES
The purpose of this research is to determine if firms are willing to sacrifice price and
delivery reliability in return for supplier security and the conditions under which firms
are more willing to sacrifice price and delivery reliability. Specifically, the research
objectives are:

A) To (re) establish how product quality, product price, delivery reliability, and
supplier security influence supplier preference.

B) To determine if purchasing firms are willing to trade-off price and delivery
reliability for increased supplier security.

C) To identify if 1) concern over past security related incidents and 2)

international sourcing increase the amount of price and delivery reliability
purchasing firms are willing to trade-off for increased supplier security.

44

RESEARCH METHODOLOGY
This section describes the specific research methodology utilized in this work.
First, this section describes the relevant unit of analysis. Second, this section defines the
sampling procedure to be used. Third, this section outlines the proposed data collection
and instrumentation. Finally, this section closes with the proposed data analysis

procedures.

UNIT OF ANALYSIS

The unit of analysis in this research is the supplier selection decision. Supplier
selection is important because it commits resources while simultaneously impacting
activities such as inventory management, production planning and control, cash flow
requirements, and product quality (Choi and Hartley, 1996). Food purchasing managers
were presented with five criteria defining a potential supplier and then asked to rate the
likelihood that they would purchase from that supplier. Respondents were also be asked
to answer a question designed to assess their concern over security incidents that have

occurred at their firm.

SAMPLING PROCEDURE

The sampling procedure utilized in this research consists of six steps (Churchill
and Iacobucci, 2002). First, the target population is defined. In this step the researcher
must define the target population about which one wishes to make an inference (e.g.
firms, households, etc. . .). The current research assesses the willingness of food

manufacturer purchasing managers to trade-off price and reliability of delivery for

45

supplier security. Therefore, the target population in this research is purchasing
managers employed by domestic firms in the food industry.

Second, the sampling frame must be identified. A sampling frame is a listing of
the elements from which the actual sample will be drawn (e. g. phone book, industry
group membership list) and should be consistent with, and representative of, the
phenomenon under study. The sampling frame for this research consists of the Institute
of Supply Management and American Purchasing Society membership lists. Prior to
selection, each industry group was evaluated for its applicability to the current research.

Third, a sampling procedure must be selected. Churchill and Iacobucci (2002)
recommend probability samples. Probability samples enable the researcher to ensure that
each population element has a known, non—zero chance of being included in the sample.
This allows the results of the investigation to be generalized across the larger population.
A random probability sample drawn from the sample frame was used.

Fourth, the sample size needed for the research must be determined. In order to
determine the sample size needed, one must take into account the statistical technique to
be used and the anticipated power level necessary to detect significant differences (Hair
et al. 1998). In this case, choice based conjoint analysis is utilized. Louviere and
Woodworth (1983) recommend thirty or more respondents for choice based conjoint
analysis. Given the number of factors and interactions proposed in the current research,
Ome (2006) recommends a minimum sample size of fifty (N = 50).

Fifth, one must select the sample elements. Sample elements are the key
informants who will serve as the respondents in the research. The key informant in this

research will be the purchasing managers in domestic food manufacturers.

46

Finally, data is collected from these sample elements. The method for data

collection and analysis is described in the next section.

DATA COLLECTION AND IMPLEMENTATION

This study collects data by surveying food purchasing managers and their
willingness to trade-off price and delivery reliability for greater supplier security.
Conjoint analysis is used to determine the value placed on each of these attributes.
Without providing a full elaboration of the advantages and disadvantages of surveys
relative to experimental designs, this section briefly identifies the primary reasons for
using a survey method to collect data. In addition, this section details the data collection
method. This section also describes the survey instrument to be used and methods of data
analysis.

The choice of survey versus experiment is one of trading-off external versus
internal validity. Through sampling procedures, surveys achieve a high degree of
external validity, which allows the results to be generalized to a larger population.
Experiments allow the researcher a high degree of control in data collection and therefore
allow any external influences to be removed. Experiments therefore provide a high
degree of accuracy in their results, but these results may not be accurate in less controlled
environments (Dunn, Seaker, and Waller, 1994).

The following criteria make surveys the chosen method for this research. First,
this research requires a sample size equal to fifty. Due to the geographical dispersion of
respondents, travel costs of administering experiments across this distance would be

prohibitive. Second, there is a need for respondent anonymity when researching

47

potentially sensitive topics such as supply chain security. Surveys allow the researcher to
afford anonymity to respondents in order to increase their willingness to participate.

As recommended by Dunn, Seaker, and Waller (1994), this research combines
both qualitative and quantitative methods. Research began by conducting a series of
interviews with purchasing managers. These interviews consisted of a series of structured
and open ended questions regarding the value they place on supplier security and how
much they are willing to sacrifice price and delivery reliability in order to receive higher
levels of supplier security. This information was used to form the relevant levels of
price, delivery reliability, and security.

A survey instrument was then developed to assess the relationships under
investigation. Due to the nature of the research questions, the instrument must be
designed to allow respondents to evaluate the utility of price, delivery reliability, and
security. The multivariate methodology best suited to this type of research question is
conjoint analysis. Conjoint analysis is a multivariate technique used specifically to
understand how respondents develop preferences for products or services (Hair et a1.
1998). It is based on the premise that respondents evaluate the value of a product or
service by combining the separate amounts of utility, or part-worths, provided by each
attribute. In conjoint analysis, the researcher constructs a set of hypothetical products or
services by combining different levels of each attribute. The researcher then asks the
respondent to perform the task of choosing among a set of products defined by these
attribute levels. The influence of each attribute, and each level of each attribute, on the
respondent’s utility function can then be determined from the respondent’s overall

ratings.

48

Data collection began with a pretest of the survey instrument as recommended by
Churchill (1979). The following steps are recommended when performing a pretest
(Dillman, 2000).

TABLE 3.1 PRETEST STEPS

 

Stage 1: Review by knowledgeable colleagues and analysts

 

0 Have I included all the necessary questions?

 

0 Can I eliminate some of the questions?

 

- Did I use categories that will allow me to compare responses to census data or
results of other surveys?

 

o What are the merits of modernizing categories versus keeping categories as they
have been used in past studies?

 

Stage 2: Interviews to evaluate cognitive and motivational qualities

 

- Are all of the words understood?

 

Are all of the questions interpreted similarly by all respondents?

 

Do all of the questions have an answer that can be marked by every respondent?

 

IS each regondent likely to read and answer each question?

 

Does the mailing package (envelope, cover letter, and questionnaire) create a
positive impression?

 

Stage 3: A small pilot study

 

0 Have I constructed the response categories for scalar questions so people
distribute themselves across categories rather than being concentrated in only
one or two of them?

 

0 Do any items from which I hope to build a scale correlate in a way that will
allow me to build scales?

 

What kind of response rate is the survey likely to obtain?

 

Are some questions generating a high non-response rate?

 

Do some variables correlate so highly that for all practical purposes I can
eliminate one or more of them?

 

o Is useful information being obtained from open-ended questions?

 

o Are entire pages or sections of the questionnaire being skipped?

 

 

o What regonse rate can I expect?

 

 

Source: Dillman (2000)

49

Levels of price, delivery reliability, and security were then be modified. The survey

instrument was then sent out to a larger sample of respondents.

INSTRUMENT

As recommended by Hair et al. (1998) a vignette was presented at the beginning
of the survey instrument describing the levels of quality, price, delivery reliability,
security, and supplier location. These represent the factors in the conjoint analysis. Hair
et al. (1998) define a factor as, a “variable the researcher manipulates that represents a
specific attribute. . .the factors (independent variables) are non-metric (and are)
represented by two or more values (or levels), which are also specified by the researcher”

(p. 390). The factors and levels examined in this research are presented below:

50

in,

t._‘\

 

__.__ _...

 

 

 

 

 

 

 

 

 

 

TABLE 3.2 FACTORS AND LEVELS
Level 1 Level 2 Level 3
Equal to or Moderately Si gnificantly
Product Quality below industry higher than higher than
average industry average industry average
Equal to or o . 3% or more
Product Price below industry L2 A) higher than higher than
industry average
average average
Equal to or 0 3% or more
Delivery Reliability above industry . 1'2 A) below below industry
Industry average
average average
Supplier Security Basic Advanced
Competence
Location of Supplier Domestic Middle East
Concern Over Security Concerned Unconcemed

Incidents

 

 

 

 

This creates a 3 x 3 x 3 x 2 x 2 matrix of possible scenarios evaluated by

respondents, and yields a total of one-hundred eight scenarios, or stimuli. The various

stimuli combinations are presented in Appendix A. Note that respondent concern is not a

supplier characteristic and is therefore not presented in conjoint scenarios. Therefore,

concern is not represented in the 3 x 3 x 3 x 2 x 2 matrix of possible scenarios, but is

included in Figure 3.2 as a reference. Table 3.3 provides the definitions of each of these

factors, and their corresponding levels. These definitions were presented to respondents

at the beginning of the survey. If respondents needed to refresh their memory of how

each factor and factor level was defined, they had the option of retrieving these

definitions from a “pop-up” window while completing the conjoint scenarios.

51

 

The reader Should note that Level One for quality, price, and delivery reliability
are designed such that this level is equal to or better than industry average. Another way
of presenting this information would have been to add a fourth level to each of these
factors that delineated “equal to industry average” and “better than industry average”.
The choice was made to combine these two levels to conform with the purpose of this
research and to reduce sample size requirements. One purpose of this research is to
assess how much price and delivery reliability purchasing managers are willing to trade-
off for supplier security. In that sense, the “worse” values of price and delivery reliability
are of more interest than the “better” values. Thus, there are two levels of “worse” values
and one combined level of “better” values. With respect to sample size requirements, the
greater the number of levels present in any one factor, the greater the sample size that is
required to accurately estimate part-worth values (Ome 2006). Thus, adding more levels
would have increased the sample size required for this research. The determination was
made that it was better to have fewer levels, reduce the sample size needed, and
simultaneously focus on the levels that were most pertinent.

There are three approaches to presenting the stimuli in conjoint analysis (Hair et
al. 1998). The first approach is the trade-off method. The trade-off method compares
attributes two at a time by asking respondents to rank order all combinations of levels in
terms of preference. The second approach is the full-profile method. The full-profile
method presents each stimulus separately and allows the respondent to view a “full-
profile” of all product (service) attributes under examination. The respondent then ranks
or rates each full-profile in terms of preference or intention to purchase. The third

approach is termed the pairwise comparison method. Pairwise comparison combines the

52

trade-off method and the full-profile method. The respondent is asked to compare two
full-profiles and choose/rate which one is most preferred. The choice of presentation
method depends on a number of factors, but the most commonly used method is the full-
profile approach (Hair et al. 1998).

The full-profile approach has the following advantages (Green and Srinivasan,
1978). First, it requires the respondent to make fewer evaluations, reducing the chance of
respondent fatigue. Second, the full-profile approach allows interaction effects to be
modeled. Third, the full-profile approach gives a more realistic stimuli description by
simultaneously defining the levels of each attribute. This more accurately depicts the

simultaneous trade-offs encountered when a respondent makes a purchasing decision.

53

TABLE 3.3 FACTOR AND FACTOR LEVEL DEFINITIONS PRESENTED TO

RESPONDENTS

 

Factor
Factor Level

Definition

 

Product Quality

Product quality is defined as how well a supplier's
product conforms to your firrn's pre-set specifications.

 

Equal to or below
industry average

The supplier's product quality is equal to or below the
industry average.

 

Moderately higher
than industry average

The supplier's product quality is moderately better
than the industry average.

 

Significantly higher
than industry average

The supplier's product quality is significantly higher
than the industry average.

 

Product Price

Product price is defined as the monetary amount a
supplier charges your firm for their product.

 

Equal to or below
industry average

The supplier's price is equal to or below the industry
average.

 

 

Above industry The supplier's price is moderately higher than the
average industry average (1-2% above industry average).
Significantly above The supplier's price is significantly higher than the

industry average

industry average (3% or more above industry average).

 

Delivery Reliability

Delivery reliability is defined as the percentage of time
that a supplier delivers product at the promised
delivery time.

 

Equal to or above
industry average

The supplier's delivery reliability is equal to or above
the industry average.

 

 

 

Below industry The supplier's delivery reliability is moderately worse

average than the industry average (1-2% below industry
average).

Significantly below The supplier's delivery reliability is significantly

industry average

 

worse than the industry average (3% or more below
industry average).

 

54

 

TABLE 3.3 CONT

 

Supplier Security Competence

Security is defined as the application of policies,
procedures, and technology to protect food supply
chain assets from theft, damage, or terrorism, and to
prevent the introduction of unauthorized contraband,
people, or weapons of mass destruction into the food
supply chain.

 

Basic

Suppliers with a basic security competence only utilize
limited measures to protect their product, processes,
and facilities from internal (e. g. disgruntled employee)
and external (e. g. terrorist) food contamination threats.
A supplier with a basic security competence typically
does_not employ any, or at least very few, of the
following secruity measures: lock their doors, employ
guards, have fences around their facilities, perform
background checks on employees, secure inventory,
and inspects product for possible contamination before
it is shipped to your firm.

 

Advanced

Suppliers with an advanced security competence make
every effort to protect their product, processes, and
facilities from internal (e. g. disgruntled employee) and
external (e. g. terrorist) food contamination threats. A
supplier with an advanced security competence
typically does employ most, or the majority, of the
following actions: lock their doors, has fences and
gates protecting their facilities, employ guards
protecting entrances, perform thorough background
checks on all employees and contractors, constantly
tracks inventory in a secure environment to make sure
it is not contaminated, and thoroughly inspects product
for possible contamination before it is shipped to your
firm.

 

Location of Supplier

Supplier location is defined as the location where the
supplier manufactures the food product that is then
shipped to you via the most appropriate mode of
transportation.

 

Domestic

Located in Iowa and employs only US. residents.

 

 

Middle East

 

Located in Saudi Arabia and employs only Saudi
residents.

 

55

 

Fourth, the full-profile approach has been shown to be superior to the other two in terms
of predictive validity, the extent to which the actual choices of the respondent are
predicted by conjoint analysis (Segal, 1982). Finally, the full-profile approach has been
shown to be more reliable in terms of test-retest correspondence (Segal, 1982).

The primary disadvantage of the full-profile approach is that the respondent’s task
is more complicated in that he/she must evaluate a level of each factor simultaneously
(Green and Srinivasan, 1978). It is recommended that no more than six factors (Green
and Srinivasan, 1978), or a total of twenty stimuli (Hair et al. 1998), be examined using
this presentation method to prevent information overload and respondent fatigue.

The current study examines five factors with one-hundred eight total stimuli.

This is greater than the twenty stimuli (Hair et al. 1998) guideline. Therefore, respondent
fatigue is a concern. Sawtooth Software’s choice-based conjoint analysis package was
used (http://www.sawtoothsofiwarecom) to alleviate this problem. Choice-based
conjoint analysis allows a large number of factors to be included in a survey by randomly
selecting scenarios to be included. This random selection of scenarios prevents
respondent fatigue. The program then calculates the part-worths respondents attach to

each factor by their responses to each scenario.

DATA ANALYSIS

Choice-based conjoint analysis was used for data analysis. Choice-based conjoint
allows the simultaneous modeling of main and interaction effects (Hair et al., 1998). The
outcome of this analysis will be the part-worths respondents attach to each factor (Hair et

al. 1998). As one part-worth is higher than another, the independent variable to which

56

that part-worth relates is deemed to be more important in determining supplier preference
than those with lower part-worths. For example, if the part-worth for price is higher than
the part-worth for supplier security then price is more important in determining supplier
preference than is supplier security. The same logic could be applied to reliability of
delivery and security. Alternatively, if the part-worth of security is higher than either
price or reliability of delivery, then security is more important than one (or both) of those

supplier attributes.

CONCLUSION

Chapter Three provided an overview of the research design and method of this
study. This chapter first reviewed the research purpose and objectives. Second, the
specific research method for this research was examined, including the unit of analysis,
sampling procedure, data collection and implementation framework, and a description of
the survey instrument to be employed. Finally, the data analysis section described the
proposed methods of examining the data and assessing goodness of fit. This research
continues by discussing the results in Chapter Four, and the implications of the research

in Chapter Five.

57

CHAPTER 4: ANALYSIS OF DATA
INTRODUCTION

Chapter Four details the research results. First, the preliminary analysis section
describes the survey mailing procedure, sample Size obtained, and sample characteristics.
Second, the hypothesis testing section presents each research question, corresponding
hypotheses, and relevant results. The Chapter concludes with a summary of the research
findings.

While reading through the following results, one should keep in mind the
definitions of basic and advanced security used by respondents when completing the
conjoint scenarios. Suppliers with a basic security competence only utilize limited
measures to protect their product, processes, and facilities from internal (e. g. disgruntled
employee) and external (e. g. terrorist) food contamination threats. A supplier with a
basic security competence typically does not employ any, or at least very few, of the
following security measures: I) look their doors, 2) employ guards, 3) have fences
around their facilities, 4) perform background checks on employees, 5) secure inventory,
and 6) inspects product for possible contamination before it is shipped. Suppliers with an
advanced security competence make every effort to protect their product, processes, and
facilities from internal (e. g. disgruntled employee) and external (e. g. terrorist) food
contamination threats. A supplier with an advanced security competence typically does
employ most, or the majority, of the following actions: I) lock their doors, 2) has fences
and gates protecting their facilities, 3) employ guards protecting entrances, 4) perform
thorough background checks on all employees and contractors, 5) constantly tracks

inventory in a secure environment to make sure it is not contaminated, and 6) thoroughly

58

inspects product for possible contamination before it is shipped. In sum, in this research
supplier security competence is conceptualized to encompass the extent to which
suppliers: 1) look doors, 2) use fences and gates to protect facilities, 3) employ guards to
protect facility entrances, 4) perform background checks on all employees and
contractors, 5) track inventory, and 6) inspect product for possible contamination before

Shipment occurs.

PRELIMINARY ANALYSIS
This section examines the mailing procedures utilized, sample Size obtained, and
sample characteristics. Sample characteristics include both respondent and firm level

data.

MAILING PROCEDURES AND FINAL SAMPLE SIZE

Three waves of mailings were utilized to obtain responses. In the first wave, a
postcard was mailed informing respondents they had been selected to participate and
detailing the goals of the research. This postcard was mailed in late March 2006.

The second mailing was sent four days after the first. This letter reminded
respondents they had been selected and provided them with the URL address and
password to access the on-line survey. Respondents were offered the opportunity to
receive an executive summary of the results. Sixty-four respondents completed the
survey after this mailing.

Another request for participation letter was mailed four weeks later in April 2006.

An additional forty-three completed surveys were returned for a final sample of one-

hundred seven responses (N = 107). A sample Size of fifty (N = 50) is needed in order to

59

accurately estimate the conjoint part-worths (Ome 2006). A sample Size of one-hundred

seven exceeds this criterion.

CHARACTERISTICS OF SAMPLE

Respondents were drawn from the membership lists of two professional
purchasing organizations. All respondents indicated to these organizations that their
firms primarily operated in the food industry. Respondent characteristics are detailed in
Table 4.1. Respondents most commonly indicated that they held the title of “manager”
(38.31 percent). Examination of responses to the “other” title category shows that these
respondents were employed in various purchasing related roles, such as a buyer (34.57
percent). The majority of the respondents classified their position as purchasing (79.43
percent), had spent between five and fourteen years in the industry (37.38 percent), two
to nine years in their position (68.61 percent), and five to nine years with their employer
(24.29 percent).

Respondents’ firms were for-profit, food manufacturers with most of their
employees in the United States. Roughly one-half of the respondents indicated that their
divisions encompassed more than two-hundred fifty employees. The other half employed
less than two-hundred fifty. This allows results to be adequately generalized across firms
of different sizes.

To address the issue of non-response bias (Armstrong and Overton 1977) the
sample was segmented into early and late respondents. Demographic data for early and
late respondents were compared using chi-square difference tests. Results are presented
in Table 4.2. Early and late respondents are indifferent with respect to each

demographic criterion except the number of employees in their division (note: profit

60

orientation was not examined due to the presence of only one “non-profit” respondent).
Early respondents tended to have fewer employees in their division than did late
respondents. However, there was no Si gnificant difference in the groups with respect to
number of employees in the United States. Number of employees in the United States is
a more aggregate measure of firm Size. Therefore, early and late respondents appear to be
possess similar demographic characteristics. Utility preference patterns for the two
groups were visually compared (note: CBC does not allow one to assess significant
differences across the utility values of different samples). Early and late respondents’
utility preference patterns were similar for each factor-level under investigation.

Therefore, data analysis proceeds treating early and late respondents as a single sample.

61

TABLE 4.1

N = 107
Title

Position

Time in Position

Time with

in Division

lo in United States

of Market Area

of F irm’s Activities

Profit Orientation

SAMPLE CHARACTERISTICS

Vice President
Director

Other

Other

0-1

2-4

5-9

10-14

15-19

20 or more
0-1

2-4

5-9

10-14

15-19

20 or more
0-1

2-4

5-9

10-14

15-19

20 or more
Less than 50
50 — 250
250-1 000

l 000 — 5 000
More than 5 000
Less than 50
50 — 250
250-1 000

l 000 — 5 000
More than 5 000
Local

National
Global
Local

National
Global

Merchandiser/Retailer

Warehouse/ Distributor
Service Provider

For-Profit

Not-For-Profit

62

Count

8
21
41
37
85
14

8

6
12
20
20
13
36
17
34
33

7

7

9
14
21
26
14

9
23
27
27
30
ll
12

5
11
25
25
41

2
10
37
58

2

4
38
63
91
ll

3

2

1

%of

 

.11.

~lr5

~\.~

~1-~

Fig-1 A-

. .\

si

\

:
I

TABLE 4.2 EARLY AND LATE RESPONDENT CHARACTERISTICS

N=107 (Early Respondents=64; Late Respondents=43) Early Late
ts ts

Title Vice President 6 2
Director 15 6
20 21
Other 23 14
Position Purc ' 49 36
1 . . . 10
Other 5
Time in 0-1 2
2-4 8
5-9 12
10-14 12
15-19 7
20 or more 8
Time in Position 0-1 8
2-4 20
5-9 19
10-14 5
15-19 4
20 or more 8
Time with 0-1 6
2-4 14
5-9 17
10-14 9
15-19 5
20 or more 13
in Division Less than 50 18
50 — 250 21
250-1 000 11
l 000 -— 5 000 6
More than 5 000 8
10 in United States Less than 50 4
50 — 250 7
250-1 000 15
l 000 — 5 000 17
More than 5 000 21

of Market Area Local 1
' 7

I—Ih—i
A

~

~

H

N
\le—OOOO-bt-‘bmoaoohMKOQOOHWN-D-AQAONOOOOAAW

National 25

NH

Global 31
Local 1
' 2
National 23
Global 38
of Firm’s Activities ' 53
Merchandiser/Retailer 9
Warehouse/Distributor 2
Service Provider 0

* = significant difference between early and late respondents at .05 level

Nu—
LIILIIN~

 

63

HYPOTHESIS TESTS

The model proposed in this research is presented in Figure 4.1. Corresponding
research questions and hypotheses are presented in Table 4.3. Choice-Based Conjoint
(CBC) produces two types of results. First, utilities represent respondents’ preference for
each factor level as it relates to supplier preference. Utilities are zero-centered for each
factor with the magnitude of the utility indicating the degree of preference. CBC utilizes
Multinomial Logit Analysis (logit) to compute utilities. Similar to multiple regression,
logit seeks “weights” for factor levels. These weights are analogous to “part-worth
utilities” in conjoint analysis. Logit utilities represent a measure of relative desirability,
or worth. The higher the utility, the more desirable the factor level. Levels with a large,
positive utility have a positive impact on supplier preference. Levels with a negative
utility do not necessarily denote that the level was undesirable to respondents, but that the
level was less desirable than others. Utilities are estimated from “counts”, the percentage
of times a particular factor level was chosen. Therefore, utilities provide direct insight
into the mindset of respondents. Utilities are computed for each factor separately

utilizing a kind of dummy coding called “effects coding”.

64

 

 

 

doumooq wamocsom

 

 

$5305
bisoom ~25 Hoodoo

 

 

 

geese 88m 2 an

m& A Na ”mm
mm A é um

 

 

 

 

 

 

cococomocm Hosmmsm

Ede

68285
$323 8.5.5

 

 

L
L e e
U

+ 2m#

 

 

+ N—m#

awe.

888a:
558m commasm

 

 

 

 

 

 

 

 

rmm

FE

8865
norm 8:on

 

 

 

 

AHA—O2 GHNENEHOE

£

+5

20%

8855

58 secs

 

 

H6 H.339?—

65

TABLE 4.3

RESEARCH QUESTIONS AND HYPOTHESES

 

Research Question

Corresponding Hypotheses

 

1. Do purchasing firms prefer suppliers who
deliver higher product quality over those that
delivery lower product quality?

HI: Higher product quality is preferred over lower
product quality.

 

2. Do purchasing firms prefer suppliers who
charge lower prices over those that charge
higher prices?

H2: Lower product price is preferred over higher
product price.

 

3. Do purchasing firms prefer suppliers who
provide higher delivery reliability over those
that provide lower delivery reliability?

H3: Higher delivery reliability is preferred over
lower delivery reliability.

 

4. Do purchasing firms prefer suppliers who
provide greater security over those that provide
less security?

H4: Higher supplier security competence is
preferred over lower supplier security competence.

 

5. Are purchasing firms willing to pay a price
premium for increased supplier security?

H5: Product price has a greater influence on
supplier preference than supplier security
competence.

 

6. Are purchasing firms willing to sacrifice
delivery reliability for increased supplier
security?

H6: Delivery reliability has a greater influence on
supplier preference than supplier security
competence.

 

7. Does concern over security related incidents
that have occrured at the purchasing firm
influence respondents to place more value on
increased supplier security?

H7: An increase in concern over security incidents
experienced by the purchasing firm positively
affects preference for suppliers with higher security
competence.

 

Hg: An increase in concern over secru‘ity incidents
experienced by the purchasing firm positively
affects preference for suppliers with lower delivery
reliability and higher security.

 

H9: An increase in concern over security incidents
experienced by the purchasing firm positively
affects preference for suppliers with higher prices
and higher security.

 

8. Do purchasing firms that source goods
internationally place more value on increased
supplier security?

Hm: International sourcing by the purchasing firm
positively affects preference for suppliers with
higher security competence.

 

H. .: International sourcing by the purchasing firm
positively affects preference for suppliers with
lower delivery reliability and higher security.

 

 

 

H”: International sourcing by the purchasing firm
positively affects preference for suppliers with
higher prices and higher security.

 

66

 

In effects coding, the utilities for each factor sum to zero. Therefore, utility figures are
interval data with an arbitrary origin, or zero point, which prevents one from comparing
the relative importance of two factors by simply comparing the utility for one level of one
factor versus the utility of one level for a second factor [c.g. the utility for basic security
= ( - 0.400) and the utility for delivery reliability three percent or more below industry
average = ( - 1.049), but Since an arbitrary zero is set for each of these factors, one cannot
say that basic security is preferred over poor delivery reliability Simply because basic
security is “less negative” than poor delivery reliability]. However, factor levels can be
compared when an interaction is present. Significant differences between factor levels in
interactions can also be assessed and are presented in the following results.

The ability of a logit utility run to assess respondent preferences is determined by
a significant )8 value. All utility runs reported in this chapter have highly significant )6
values (p< 0.01). Thus, respondent choices are significantly affected by the factors under
investigation and data analysis can proceed.

The second type of results is derived from market Simulations. Through market
Simulations, one has the ability to estimate the effects of various combinations of levels
on the market preference of a product. In the context of this research, market simulation
gives the percentage of market preference garnered by hypothetical suppliers
characterized by different factor levels. Market simulation is based on logit utilities, but
differs from logit in that it uses algorithms to group respondents with similar preferences
and estimate part-worths within these segments. This information is then used to

determine preference for a group of levels that have been combined to simulate a

67

hypothetical supplier. These group preferences are then used to calculate the market
preference of each hypothetical supplier.

There are several reasons for using market Simulations. First, market simulations
transform raw utility data into a managerially useful and appealing model for predicting
the market preference of product mix combinations. Second, simulations can capture
idiosyncratic respondent preferences, not discemable through utilities, that may have a
significant impact on overall preference. Third, simulations reveal the extent to which
respondents are willing to substitute one factor for another. This is particularly valuable
given the goals of this research. Fourth, simulations reveal three-way interaction effects
between attributes, whereas CBC only provides utility values for two-way interactions.
Again, this capability is particularly valuable given the present research questions.

In each market Simulation scenario in this research, all other potential supplier
attributes other than those under evaluation are held constant at their most desirable level
as determined by the individual factor level utilities (e. g. in the case of price vs. security:
delivery reliability is equal to or above industry average, supplier location is United
States, and quality is significantly higher than industry average). This allows the
researcher to focus on the security by price interaction (in this case) without other
confounding influences. As a caveat, market Simulations are only meant to provide
supporting or dissenting evidence for the hypothesized relationship between variables
under investigation. Market simulation results presented in this chapter are not meant to
capture the influence of other supplier characteristics and will not necessarily hold true if
these other characteristics, a greater number of suppliers, or other factors such as level

and effectiveness of advertising, product availability, and newness of a product or

68

supplier were taken into account. Market simulation results will also not hold true in
every industry. It is beyond the scope of this research to include every possible supplier
characteristic, market factor, or industry specific influence. Therefore, the term “market
share” is replaced with “market preference” in the preceding discussion and following
results. As another caveat, efforts were made in this research to determine significant
differences between market preferences of interest through use of the Wilcoxon rank-sum
test (Lehmann 1975), however, due to the limited number of observations few significant
differences were found. Without use of further statistical techniques, such as
bootstrapping, caution should be taken when interpreting these insignificant findings (see
Appendix C). This limitation does not severely hamper the overall goal of this research
in as much as market simulations are but one of several evaluation methods used.

The choice of results to interpret depends on the research question under
investigation. Utilities are sufficient for determining preference for one level over
another in a single factor or two-way interactions in the same logit run. Utilities cannot
be used to compare preference patterns across different factors (where an interaction is
not examined) and logit runs. When three-way factor interactions, substitutability, and
demand elasticity are examined, market simulation provides information over and above
what can be obtained via utilities.

It should be noted that utility preferences and market simulation data may not
always agree. It is possible that respondents may prefer one factor level over another, but
other information is included in the simulation to determine supplier market preference.
Therefore, utilities provide insight into managerial thought processes, but market

simulations provide further insight into choices groups of managers might make. H1

69

through H4 examine simple preferences of one factor level over another. Therefore, only

the utilities for each level are examined.

H1 through H4 examine respondent preferences for three levels of supplier product

quality, price, delivery reliability, and two levels of supplier security. Utility figures for

each factor level are presented in Table 4.4.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE 4.4 HYPOTHESES 1-4 RESULTS: UTILITIES
Hypothesis Factor Level Utilities
Significantly Higher than Industry Average 0.528
H1 Quality Moderately Higher than Industry Average 0.378
Equal to or Below Industry Average (0.906)
Equal to or Below Industry Average 0.763
H2 Price l-2% Higher than Industry Average 0.148
3% or more Higher than Industry Average (0.911)
Delive Equal to or Above Industry Average 0.986
H 3 Reliabilliil 1-2% Below Industry Average 0.064
y 3% or more Below Industry Average (1.049)
. Advanced 0.400
H4 secumy Basic (0.400)

 

 

 

 

RQl: Do purchasing firms prefer suppliers who deliver higher product quality
over those that delivery lower product quality?

H1: Higher product quality is preferred over lower product quality.

H1 proposes that respondents prefer suppliers who deliver higher levels of product quality

over those that deliver lower levels of product quality. H1 is supported. Comparing

higher quality levels to lower quality levels, utilities indicate that respondents

70

 

significantly prefer suppliers with higher quality product over those with lower quality
(p<0.05) for each level of quality.l
H2 examines respondent preferences for three price levels.
RQ2: Do purchasing firms prefer suppliers who charge lower prices over those
that charge higher prices?

H2: Lower product price is preferred over higher product price.

H2 proposes that respondents prefer suppliers who charge a lower price for their product
over those that charge a higher price for their product. H2 is supported. Comparing
higher price levels to lower price levels, utilities indicate that respondents significantly
prefer suppliers with lower prices over suppliers with higher prices (p<0.05) for each
price level.
H3 examines respondent preferences for three levels of delivery reliability.
RQ3: Do purchasing firms prefer suppliers who provide higher delivery
reliability over those that provide lower delivery reliability?
H3: Higher delivery reliability is preferred over lower delivery
reliability.
H 3 proposes that respondents prefer suppliers who provide higher levels of delivery
reliability over those that provide lower levels of delivery reliability. H3 is supported.
Comparing higher levels of delivery reliability with lower levels of delivery reliability,
utilities indicate that respondents significantly prefer suppliers with higher delivery
reliability over suppliers with lower delivery reliability (p<0.05) for each delivery

reliability level.

 

I Given the basic nature of Hypotheses 1-4, specific z-values indicating significant differences are not
presented in tabular form.

71

H4 examines respondent preferences for two levels of supplier security.

RQ4: Do purchasing firms prefer suppliers who provide greater security over

those that possess less security?

H4: Higher supplier security competence is preferred over lower
supplier security competence.
H4 proposes that respondents prefer suppliers with advanced security competence over
those with basic security competence. H4 is supported. Comparing basic security with
advanced security, utilities indicate that respondents significantly prefer suppliers with
advanced security over suppliers with basic security (p<0.05).

The results of H1 through H4 are intuitive and require little explanation.
Respondents derive more utility from “better” levels of each factor. H; through H4 were
necessary to set the stage for subsequent hypotheses. H5 and H6 explore factor
interactions to determine the relationship of supplier security to price and delivery
reliability respectively.

H5 and H6 are discussed subsequently.

RQS: Are purchasing firms willing to pay a price premium for increased supplier

security?

H5: Product price has a greater influence on supplier preference than
supplier security competence.
H5 pr0poses that the price charged by a supplier has a greater influence on supplier
preference than the security capability of the supplier. In order to assess H5, it is

necessary to examine two criteria: 1) the importance of price and security in determining

72

supplier preference and 2) preferences for various price/security supplier service
offerings.

First, the importance (utility) of price must be compared with the importance
(utility) of security. In order to accomplish this, it is necessary to determine the relative
importance of price and security via factor importance scores (see Table 4.5). Factor
importance scores denote the amount of total possible utility in a model accounted for by
individual factors (e. g. price and security). In order to calculate factor importance scores,
one must first ascertain the total amount of utility respondents can derive from all factors
under investigation.

The total utility respondents can derive from all factors under investigation is
equal to the sum of the ranges of utility values for each factor. The range of utility values
for a single factor is equal to the highest utility value in that factor minus the lowest
utility value in that factor. Column one (from left to right) in Table 4.5 presents the five
factors, and corresponding factor levels, under investigation. Column two presents the
utility values for each factor level. Column three presents the range between the highest
utility and lowest utility for each factor. The greater the range between the most
preferred level of a factor and the least preferred level of a factor, the more that factor
contributes to overall utility. This is due to the fact that respondents who place a great
deal of importance on low price (for example) choose the lowest price option most
frequently. Frequently choosing the lowest price level results in a very high utility value
for that level and a very low utility value for the most expensive level of price. If
respondents do not take price into account when making their choice, they choose a high

price level in one scenario, and a low price level in another, with little consistency. This

73

leads to a smaller range of utility values between price levels and indicates respondents
made their decisions based on factors other than price. Taking the range of all five
factors (e. g. security, price, delivery reliability, supplier location, and quality) and
summing these ranges yields the total amount of utility that can be derived in a given
model. This total utility value is presented at the bottom of column three in Table 4.5.
Dividing the range of each individual factor by this total yields the percentage of total
utility accounted for by each factor (i.e. the factor importance score). Higher factor
importance scores indicate that a factor accounts for a greater percentage of the total
utility in a given model. Factor importance scores are presented in column four of Table
4.5. Presently, if a factor has a higher factor importance score, it accounts for a greater
percentage of utility in the model and therefore is more important to respondents in their
mental models of supplier preference. H5 is partially supported if price has a higher
factor importance score than security.

Second, if respondents place more value on price than security, then they would
not be willing to pay a higher price in order to utilize a supplier with advanced security.
In order to determine if this is true, the researcher must examine the two-way interaction
of price and security. This is accomplished by examining 1) utilities and 2) market
simulation results (note: utility magnitudes can be examined in H5 because they are
estimated via a two-way interaction in a single logit run). If respondents are unwilling to
trade-off price for advanced security, the following must hold true: 1) utilities must
indicate respondents always prefer basic security over advanced security at every price
level and 2) the market preference for suppliers with basic security and a lower price

level must be greater than the market share for suppliers with advanced security and a

74

higher price level. In order to ease the comparison of market share figures for basic
security/lower price (or delivery reliability) suppliers and advanced security/higher price
(or delivery reliability) suppliers, several tables are included in this Chapter that resemble
Table 4.8. These tables each contain three comparisons (from top to bottom) comparing
the market preference of basic security/lower price (delivery reliability) suppliers with
advanced security/higher price (delivery reliability) suppliers. These comparisons are
followed by an “implication” in bold that details the meaning of the directional
relationship depicted above it. Utilities and market simulation results are presented in
Table 4.6.

Each of these three criteria (factor importance scores and utilities/market
simulations) contributes to the overall support, or lack thereof, of H5. If results of the
three criteria are congruent with each other and the hypothesized relationship, then full
support is given to H5. If one result is in disagreement, less support is afforded to H5. If
the three results are congruent with each other, but incongruent with the hypothesized

relationship, then no support is found for H5.

75

TABLE 4.5 FACTOR IMPORTANCE SCORES

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Utility Range for Each Factor Importance
Factor Utilities Factor Score (”/o)
Security 0.800 10. 64
Advanced 0.398
Basic (0.398)
Price 1.674 22 .27
Equal to or Below 0.763
Industry Average
1-2% Higher than 0.148
Industry ArLrage
3% or more Higher than (0.911)
Industry Average
Delivery Reliability 2.036 27.09
Equal to or Above 0.986
Industry Average
l-2% Below Industry 0.064
Average
3% or more Below (1.049)
Industry Average
Location 1.571 20.91
United States 0.786
Saudi Arabia (0.786
Quality 1.435 19.09
Significantly Higher 0.528
than Industry Average
Moderately Higher than 0.378
Industry Average
Equal to or Below (0.906)
Industry Average
Sum 7.515

 

TABLE 4.6 HYPOTHESIS 5 RESULTS

: SECURITY BY PRICE

 

 

 

 

 

 

 

INTERACTION
Security Market Preference
Level Price Level Utilities %
Basic Equal to or below 0.083 27.06
Basic l-2% higher (0.076) 6.82
Basic 3% or more higher (0.007) 2.03
Advanced Equal to or below (0.083) 37.1 1
Advanced 1-2% higher 0.076 22.50
Advanced 3% or more higher 0.007 4.48

 

 

 

 

 

 

76

 

TABLE 4.7 HYPOTHESIS 5 RESULTS: ASSESSNIENT OF SIGNIFICANT
DIFFERENCES BETWEEN UTILITY VALUES

Level Adv Adv
Equal 1- or Equal 1-2% 3% or
Security to or higher more to or higher more
Level Price Level below below

to or

1-2% 2.1
3% or more 1.1
to or
1- 0.10 l
3% or more . 1.14 0.85
at p <

    

Factor importance scores reveal that price contributes more to supplier preference
(score = 22.27 percent) than does security (score = 10.64 percent). Utilities for the two—
way interaction between security and price indicate that respondents prefer basic security
when price is equal to or below industry average, advanced security when price is one to
two percent higher than industry average, and advanced security when price is three
percent or more higher than industry average. Respondents significantly prefer basic
security and equal to or below prices to advanced security and equal to or below prices
and advanced security and one to two percent higher prices over basic security and one to

two percent higher prices. Figure 4.2 depicts these relationships graphically.

77

FIGURE 4.2 SECURITY BY PRICE INTERACTION: UTILITIES

l
|
l
l
l
l
l
l
l

 

 

 

More Preferred

 

in Basic Security

 

 

 

 

 

 

 

 

 

 

Less Preferred

|
|

Market simulation results graphically depicted in Figure 4.3 reveal that respondents
would be more likely to choose a supplier with advanced security over one with basic
security at each price level. Figure 4.3 also indicates that the market preference
percentage of a supplier with advanced security declines linearly as price increases. The
market preference percentage of a supplier with basic security drops more sharply as
price increases from equal to or below average to one to two percent higher and then
levels out as price increases further. This indicates that preference for suppliers with

advanced security is somewhat less elastic with respect to price.

78

I. .
l

l _- Advanced Security '1

l

FIGURE 4.3 SECURITY BY PRICE INTERACTION: MARKET
PREFERENCE

 

 

 

 

 

 

40.00%
30.00% i - A
3+“ " 555%;qu
20.00% — s- y . .
l - 4:}— - Advanced Secunty .
10.00%
0.00% .
Equal to or below 1-2% higher 3% or more higher

 

 

 

Table 4.8 compares the market preference percentages of suppliers with lower
security and lower price to those with higher security and higher price in order to
determine if respondents are willing to trade-off price for security. Results indicate
suppliers with basic security and lower prices garner a greater percentage of market
preference than suppliers with advanced security and higher prices. This indicates
respondents are more likely to choose a supplier with basic security and lower prices than

a supplier with advanced security and higher prices.

79

TABLE 4.8 SECURITY BY PRICE INTERACTION: MARKET
PREFERENCE COMPARISON2

 

 

 

 

 

 

 

Security Market Directional Market Security
Level Price Level Preference Relationship Preference Price Level Level
. Equal to or 0 o 1-2% .
Basrc below 27.06 A) > 22.50 A) higher Advanced
Implication

 

 

 

Respondents chose a supplier with basic security/equal to or below prices more frequently than one

 

with advanced security/l-2% higher prices.

 

 

 

 

 

 

 

F“ , ,_ _-.v. T31}. .-;-:.-.:. ”SRTZ; ...... , . . .4
Security Market Directional Market Security
Level Price Level Preference Relationship Preference Price Level Level
' 0
Basic Equal 1° °‘ 27.06% > 4.48% 3 /° 9’ “’0” Advanced
below hrgher
Implication

 

 

 

Respondents chose a supplier with basic security/equal to or below prices more frequently than one

 

with advanced security/3% or more higher prices.

' h-wA-n . A -...,.- ~,. .Ali‘A‘J) 'V""V"ql.‘cul .tgpr- ----- . ‘ v. < I'mv‘v ‘
.... .. ,,.....a‘.. 3' ........ --. .-I . .._ ...,._..,.,.. . .~ ,. _‘_,, ,,,.,. . .
3f... ............ ~~, , ... ,..,,,..(. ‘, _, a... .t- e» .. .. . -‘

 

 

 

 

 

 

 

Security Market Directional Market Security
Level Price Level Preference Relationship Preference Price Level Level
. 1-2% 0 0 3% or more
Basrc higher 6.82 /o > 4.48 A) higher Advanced
Implication

 

 

 

Respondents chose a supplier with basic security/l-2% higher prices more frequently than one with
advanced security/3% or more higher prices.

 

 

 

In summary, H5 posits that price has a greater influence on supplier preference
than security. Factor importance scores support H5 and indicate price influences supplier
preference more than security. Utilities do not support H5 and indicate that respondents
prefer a supplier that charges one percent or more above industry average if that supplier
employs advanced security measures. Market simulation results support H5 and indicate
that, while respondents are more likely to choose a supplier with advanced security over

one with basic security at the same price level, they are also more likely to choose a

 

2 Note that Table 4.7, as well as other sirrrilar tables in this Chapter, contain three comparisons (from top to
bottom) comparing the market preference percentage of suppliers with basic security and “better”

price/delivery reliability levels to suppliers with advanced security and “worse” price/delivery reliability
levels.

80

supplier with lower prices and lower security over one with higher prices and higher
security.

Therefore, H5 is partially supported. Respondents place more value on price than
security and are not willing to trade-off price for security, but the utility interactions are
interesting in that they Show some preference for higher levels of security at higher
prices. This may indicate that two groups of respondents are represented: Those that
focus on price and those that focus on other supplier attributes (e. g. security).

Ho examines the relationship between security and delivery reliability.

RQ6: Are purchasing firms willing to sacrifice delivery reliability for increased

supplier security?

H6: Delivery reliability has a greater influence on supplier preference
than supplier security competence.
H6 proposes that delivery reliability has a greater influence on supplier preference than
security. H6 is evaluated in the same manner as H5. Factor importance scores presented
in Table 4.4 reveal that respondents place more importance on delivery reliability (score
= 27.09 percent) than security (score = 10.64 percent). Utility and market simulation
results are presented in Table 4.9.

TABLE 4.9 HYPOTHESIS 6 RESULTS: SECURITY BY DELIVERY
RELIABILITY INTERACTION

 

 

 

 

 

 

 

 

 

 

Security Market Preference
Level Delivery Reliability Level Utilities %
Basic Equal to or Above 0.054 30.28
Basic l-2% Below (0.098) 5.11
Basic 3% or more Below 0.044 1.65
Advanced Equal to or Above (0.054) 43.36
Advanced l-2% Below 0.098 16.90
Advanced 3% or more Below (0.044) 2.70

 

 

 

81

TABLE 4.10 HYPOTHESIS 6 RESULTS: ASSESSMENT OF SIGNIFICANT
DIFFERENCES BETWEEN UTILITY VALUES

 

 

 

 

 

 

 

 

 

 

Security Level Basic Basic Basic Adv Adv Adv

Security Delivery Reliability Equal to 1-2% 3% or Equal 1-2% 3% or

Level Level or above below more to or below more
below above below

Basic Equal to or above __4 if:

Basic l-2% below 203* ,5

Basic 3% or more below 0.12 1.72 f 17:13.: :7;

Adv Equal to or above 1.47 0.59 1.20 y; j‘f‘jii

Adv 1-2% below 0.59 2.57* 0.65 203* 7

Adv 3% or more below 1.20 0.65 0.99 0.12 1.72

 

 

 

 

 

 

 

 

* = Significant difference at p < 0.05

Utilities for the two-way interaction between security and delivery reliability

indicate respondents prefer basic security at the highest and lowest levels of delivery

reliability and advanced security when delivery reliability is one to two percent below

industry average. Respondents Si grrificantly prefer advanced security and one to two

percent below average delivery reliability over basic security and one to two percent

below average delivery reliability. Figure 4.4 depicts this relationship graphically.

82

 

FIGURE 4.4 SECURITY BY DELIVERY RELIABILITY INTERACTION:
UTILITIES

 

 

 

13 I
0 1
1 r: 1
2 .
2 .
o.
9 i
O 1
s 1 . - . .-
1 13 BaSIC Secunty
IAdvanced Security
1
'13 l
2 .
L
.2 1
n- 7. _
(0
:3; Equal to or above 1-2°/o below 3% or more below

Market simulation results graphically depicted in Figure 4.5 reveal that respondents are
more likely to choose a supplier with advanced security at each price level. Figure 4.5
also reveals similar declines in market share for suppliers with basic and advanced
security as delivery reliability decreases to one to two percent below average. As
delivery reliability decreases to three percent or more below average, the market
preference percentage declines more for a supplier employing advanced security, but the
market preference percentage for suppliers employing basic security did not have as far

to fall.

83

FIGURE 4.5 SECURITY BY DELIVERY RELIABILITY INTERACTION:
MARKET PREFERENCE

50. 00%

l 40.00%

 

' 30.00% - * 1' ILL-.. .- — "’T’
| + Basrc Security 1‘
1 — p— - Advanced Security ‘1

l 20.00%

l 10.00% 1

 

 

 

0.00%

 

1
Equal to or above 1-2% below 3% or more below 1
1

Table 4.11 compares market preference percentages for suppliers with higher
delivery reliability and basic security versus those with lower delivery reliability and
advanced security. Results indicate suppliers with basic security and higher delivery
reliability garner more market preference than suppliers with advanced security and
lower delivery reliability. This indicates respondents are more likely to choose a supplier
with basic security and higher delivery reliability than a supplier with advanced security

and lower delivery reliability.

84

TABLE 4.11 SECURITY BY DELIVERY RELIABILITY INTERACTION:
MARKET PREFERENCE COMPARISON

 

 

 

 

 

 

 

Delivery Delivery
Security Reliability Market Directional Market Reliability Security
Level Level Preference Relationship Preference Level Level
. Equal to or o 0 1-2%
Basrc Above 30.28 /o > 16.90 /o Below Advanced
Implication

 

 

 

Respondents chose a supplier with basic security/equal to or above delivery reliability more

 

.-. r....,'.,..

‘ -'"~"-~u-Jr

 

 

 

 

 

 

 

* “Berna” ' ' ‘ """"“"“""""‘"“S.1:v.;y“ i ”7‘
Security Reliability Market Directional Market Reliability Security
Level Level Preference Relationship Preference Level Level
. Equal to or o 0 3% or more
Basrc Above 30.28 /0 > 2.70 /o Below Advanced
Implication

 

 

 

Respondents chose a supplier with basic security/equal to or above delivery reliability more
frequently than one with advanced security/3% or more below average delivery reliability.

 

..........................
......

I I ' ~‘-.‘l‘t' '. .‘

--a;-Ana_...._

'.'. n.
‘A ._i_

 

\ 1 I 7 -‘n - .' '.' Va 1'. . . ’
..‘..“ . .‘.I‘.".. .,,'..',,.. . A: .

Delivery ’

 

 

 

 

 

 

Security Reliability Market Directional Market Reliability Security
Level Level Preference Relationship Preference Level Level
. 1-2% 0 0 3% or more
Basrc Below 5.1 l /o > 2.70 /0 Below Advanced
Implication

 

 

 

 

Respondents chose a supplier with basic security/l-2% below delivery reliability more frequently
than one with advanced security/3% or more below average delivery reliability.

 

In surrunary, H6 posits that delivery reliability has a greater influence on supplier

preference than security. Factor importance scores support H6 and indicate delivery

 

reliability influences supplier preference more than security. Utilities do not support Ho
and indicate that respondents prefer a supplier with one to two percent below industry
average delivery reliability if that supplier employs advanced security measures. Market
Simulation results support H6 and indicate that, while respondents are more likely to
choose a supplier with advanced security over one with basic security at each level of

delivery reliability, they are also more likely to choose a supplier with higher delivery

85

reliability and basic security over one with lower delivery reliability and advanced
security.

H6 is partially supported. While factor importance scores and market simulation
results support the assertion that delivery reliability is more important than security,
utilities indicate that respondents prefer one to two percent below industry average
delivery reliability if the supplier employs advanced security measures. Utility results
may be a function of consolidation in the food industry, which leads to the presence of
few suppliers for many goods. In a situation where few suppliers exist for a given
commodity, purchasing firms may be forced to accept lower levels of delivery reliability,
but would place a higher priority on the security competence of these suppliers due to
their strategic importance. Should one of these strategic sources of materials experience
a security incident, the viability of the purchasing firm would be threatened.

H7 through H9 examine respondents’ concern over security incidents that have
occurred at their firm as it affects preference for suppliers with advanced security, lower
delivery reliability, and higher prices. H7 examines the role of respondent concern as it
relates to preference for suppliers with advanced security. These results are presented
subsequently.

RQ7 : Does concern over security related incidents that have occurred at the firm

influence respondents to place more value on increased supplier security?

H7: An increase in concern over security incidents experienced by the
purchasing firm positively affects preference for suppliers with higher
security competence.

H7 proposes that respondents who are concerned about the security incidents that have

occurred at their firm have greater preference for suppliers with advanced security. In

86

order to evaluate H7 it was necessary to segment the respondents into those who were
concerned about security incidents and those who were not. Respondents were asked to
indicate their level of concern regarding their firm’s security incidents. Table 4.12

provides the question, response options, and response frequencies and percentages.

TABLE 4.12 RESPONDENTS’ CONCERN OVER SECURITY INCIDENTS

 

 

 

 

 

 

 

 

 

 

“How concerned are you about the intentional contamination/security
incidents that have occurred at your firm?”
Number who Percent who

Response Option answered this option answered this option
Very Unconcemed 4 3.73
Unconcemed 9 8.41
Neutral 21 19.62
Concerned 37 34.57
Very Concerned 23 21.49
Not Applicable 13 12.14

Total 107 100

 

 

 

 

Respondents were then segmented into two groups: First, concerned respondents
are those who indicated they were “Concerned” or “Very Concerned” about their fimr’s
security incidents (N = 60). Second, unconcerned respondents are those who indicated
they were “Very Unconcemed”, “Unconcemed”, “Neutral”, or that the question was “Not
Applicable”, which would conceivably indicate that they do not believe a security
incident had occurred at their firm (N = 47).

H7 is evaluated by comparing factor importance scores, utilities, and market
simulations for concerned and unconcerned respondents. In order for H7 to be fully
supported, 1) factor importance scores must indicate that security is more important for
concerned respondents than unconcerned respondents, 2) utilities must indicate

concerned respondents prefer suppliers with advanced security and unconcerned

87

respondents do not, and 3) market simulation results must indicate that concerned
respondents are more likely to choose a supplier with advanced security than are
unconcerned respondents. For the market simulation method, two separate simulations
were performed: one for concerned respondents and one for unconcerned respondents.
Both simulations held all other factors at their “best” levels in order to examine the
effects of security. Table 4.13 provides factor importance scores and Table 4.14 provides
utility and market share results.

TABLE 4.13 FACTOR IMPORTANCE SCORES: SEGMENTED SAMPLE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Concerned Respondents Unconcemed Respondents
Factor Factor
Importance Importance
Factor Utilities Range Score (%) Utilities Range Score (%)
Security 0.981 12.79 0.602 8.16
Basic 0.491 (0.301)
Advanced (0.491) 0.30]
Price 1.514 19.73 1.856 25.15
Equal to or Below 0.676 0.860
Industry Average
1-2% Higher than 0.163 0.137
Industry Average
3% or more Higher (0.838) (0.996)
than Industry
Avegage
Delivery 2.208 28.78 1.878 25.44
Reliability
Equal to or Above 1.060 0.91 1
Industry Average
l-2% Below 0.087 0.055
Industry Average
3% or more Below (1.148) (0.966)
Industry Avergge
Location 1.591 20.74 1.535 20.79
United States 0.796 0.767
Saudi Arabia (0.796) (0.767)
Quality 1.379 17.97 1.511 20.47
Significantly 0.499 (0.943)
Higher than
Industry Average
Moderately Higher 0.380 0.375
than Industry
Average
Equal to or Below (0.879) 0.568
Industry Average
Sum 7.673 Sum 7.382

 

 

 

 

 

 

 

88

 

TABLE 4.14 HYPOTHESIS 7 RESULTS: SECURITY BY INCIDENT
CONCERN INTERACTION

0. 1 0.301 92
1 .80 1 45.

 

Factor importance scores presented in Table 4.13 indicate that concerned
respondents place more importance on security (score = 12.79 percent) than do
unconcerned respondents (score = 8.16 percent). The magnitude of utilities for
concerned and unconcerned respondents cannot be compared because they were derived
from separate logit runs. However, Figures 4.6 and 4.7 illustrate that both groups prefer
advanced security over basic security and this difference is significant for both concerned
and unconcerned respondents (p < 0.05).

FIGURE 4.6 SECURITY BY INCIDENT CONCERN INTERACTION:
CONCERNED RESPONDENT UTILITIES

 

 

More Preferred

l

 

 

 

 

 

 

 

 

 

Basic Advanced

‘ Less Preferred

89

FIGURE 4.7 SECURITY BY INCIDENT CONCERN INTERACTION:
UNCONCERNED RESPONDENT UTILITIES

 

More Preferred 1

I Unconcemed

 

 

 

Less Preferred

Basic Advanced

Market simulation indicates concerned respondents choose a supplier with higher security
more often (market preference = 60.20 percent) than their unconcerned counterparts
(market preference = 54.92 percent). This relationship is graphically depicted in Figure
4.8.

FIGURE 4.8 SECURITY BY INCIDENT CONCERN INTERACTION:
MARKET PREFERENCE

‘ 80.00%

 

 

1
60.00%
40 000/ +5” Concemiedfli‘
' a — -C1- — Unconcemed}
20.00%
0.00%

 

 

 

Basic Advanced

90

In summary, H7 posits that respondents who are concerned about the security
incidents that have occurred at their firm have greater preference for suppliers with
advanced security. Factor importance scores support H7 and indicate that security
influences supplier preference more for concerned respondents (score = 12.79 percent)
than unconcerned respondents (score = 8.16 percent). Utility figures do not support H7
and indicate both groups prefer advanced security over basic security. Market simulation
results support H7 and indicate that concerned respondents are more likely to choose a
supplier with advanced security than unconcerned respondents.

H7 is partially supported. Concerned respondents place more importance on
security and choose a supplier with advanced security more often than unconcerned
respondents. Both groups prefer suppliers with advanced security.

H3 examines the influence of concern on respondents’ preference for suppliers
with lower delivery reliability and higher security.

H8: An increase in concern over security incidents experienced by the

purchasing firm positively affects preference for suppliers with lower

delivery reliability and higher security.
Hg posits that concerned respondents are more willing to sacrifice delivery reliability for
security than unconcerned respondents. H3 is evaluated similarly to H7, except that a
three-way interaction of security, delivery reliability, and respondent concern over
security incidents is being evaluated. For H3 to be fully supported, 1) factor importance
scores must indicate that delivery reliability is less important for concerned respondents
than unconcerned respondents, 2) utilities must indicate that concerned respondents

prefer higher levels of security, coupled with lower levels of delivery reliability, and

91

unconcerned respondents do not, and 3) market simulation results must Show that

concerned respondents are more likely to choose suppliers with higher security, and

lower delivery reliability, than unconcerned respondents. Table 4.15 presents utility and

market simulation results for concerned and unconcerned respondents.

TABLE 4.15 HYPOTHESIS 8 RESULTS: SECURITY BY DELIVERY
RELIABILITY BY INCIDENT CONCERN INTERACTION

 

 

 

 

 

 

 

 

 

 

Concerned Unconcemed
Respondents Respondents
Market Market
Security Delivery Reliability Preference Preference

Level Level Utilities % Utilities %
Basic Equal to or Above 0.098 28.31 0.023 32.1 1
Basic 1-2% Below (0.017) 5.00 (0.195) 4.90
Basic 3% or more Below (0.082) 0.65 0.172 3.38
Advanced Equal to or Above (0.098) 47.18 (0.023) 38.71
Advanced 1-2% Below 0.017 15.64 0.195 18.97
Advanced 3% or more Below 0.082 3.23 (0.172) 1.94

 

 

 

 

 

TABLE 4.16 HYPOTHESIS 8 RESULTS: ASSESSMENT OF SIGNIFICANT
DIFFERENCES BETWEEN UTILITY VALUES: CONCERNED RESPONDENTS

 

 

 

 

 

 

 

 

 

 

Security Level Basic Basic Basic Adv Adv Adv

Security Delivery Reliability Equal to 1-2% 3% or Equal 1-2% 3% or

Level Level or above below more to or below more
below above below

Basic Equal to or above ,,,,,,,, ..

Basic 1-2% below 1.13 ,4

Basic 3% or more below 1.59 0.57 ' 7' 3

Adv Equal to or above 1.96* 0.80 0.15 *

Adv l-2% below 0.80 0.32 0.86 1.13 ' _

Adv 3% or more below 0.15 0.86 1.31 1.59 0.57 7’

 

 

 

 

 

 

 

 

* = Significant difference at p < 0.05

92

 

 

TABLE 4.17 HYPOTHESIS 8 RESULTS: ASSESSMENT OF SIGNIFICANT
DIFFERENCES BETWEEN UTILITY VALUES: UNCONCERNED
RESPONDENTS

to 1-2% or 1-2% 3% or
Level or above below more to or below more
below above below

to or above
1- 1
or more 1.22
to or 1 1.5
1— 1.51 3 1 1
3% or more 1.60 0.19 2.61 1.22 296*
at p < 0.05

    

Factor importance scores presented in Table 4.13 reveal that concerned
respondents actually place more importance on delivery reliability (score = 28.78
percent) than do unconcerned respondents (score = 25.44 percent). Utility preference
patterns indicate that the two groups are generally the same, with one exception:
concerned respondents preferred a supplier with advanced security and three percent or
more below industry average delivery reliability, whereas unconcerned respondents did
not prefer this option. Examining significant differences between utility values for
concerned respondents reveals that they significantly prefer basic security and equal to or
above delivery reliability over advanced security and equal to or above delivery
reliability. Unconcemed respondents significantly prefer advanced security and one to
two percent below delivery reliability over basic and one to two percent below average
delivery reliability and basic security and three percent or more below average delivery

reliability over advanced and three percent or more below average delivery reliability.

93

Figures 4.9 and 4.10 depict this relationship graphically for concerned and unconcerned
respondents respectively.

FIGURE 4.9 SECURITY BY DELIVERY RELIABILITY BY INCIDENT
CONCERN INTERACTION: UTILITIES FOR CONCERNED RESPONDENTS

 

 

 

 

 

B
I:
I‘D , 7
1 E 1
D.
9 1
1 ° -7 ‘
5 E] Concerned/Basic ‘1
1 I Conerned/Advapeedl1
l .
1
'O
E
1 .2
2
0.
w 1
g Equal to or above 1-2% below 3 % or more below 1

7 ”7A-..- 2. , , #7,. 1

94

FIGURE 4.10 SECURITY BY DELIVERY RELIABILITY BY INCIDENT
CONCERN: UTILITIES FOR UNCONCERNED RESPONDENTS

 

 

 

 

 

 

‘5

E

a)

E

Q.

2

O

E [:1 Unoncerned/Basic

I Unconerned/Advanced

‘U

E

.2

2

D.

8

3 Equal to or above 1-2% below 3 % or more below ‘

Figure 4.1 1 graphically depicts market simulation results of the three-way
interaction between delivery reliability, incident concern, and security. When a supplier
has delivery reliability equal to or above average and advanced security, concerned
respondents are more likely to choose this supplier (market preference = 47.18 percent)
than unconcerned respondents (market preference = 38.17 percent). However, when a
supplier’s delivery reliability drops to one to two percent below industry average,
unconcerned respondents are more likely to choose a supplier with advanced security
(market preference = 18.97 percent) than concerned respondents (market preference =
15.64 percent). At three percent or more below industry average delivery reliability,
unconcerned respondents are more likely to choose a supplier with basic security (market
preference = 3.38 percent) and concerned respondents are more likely to choose
advanced (market preference = 3.23 percent).

These results reflect the finding that concerned respondents derive more utility

from delivery reliability than unconcerned respondents, as evidenced by the factor

95

importance scores in Table 4.13. While concerned respondents also place more value on
security than unconcerned respondents, Figure 4.11 illustrates that they do not value
security to the extent that they are more likely than unconcerned respondents to choose a
supplier with advanced security and one to two percent below average delivery
reliability.

FIGURE 4.11 SECURITY BY DELIVERY RELIABILITY BY INCIDENT
CONCERN INTERACTION: MARKET PREFERENCE

 

 

 

 

 

 

 

50.00%
40.00%
, 30 000/ f—o—T" concerned/Bag?
' ° 7 - -D- — Concerned/Advanced
20 000/ .— -A - — Unconcemed/Basic
. o « _
. i f 7’? i:_UD_¢°"Pef”e.d’.iA£V?”°e<l ,
10.00%
0.00% ,
Equal to or 1-2% below 3 % or more
above below

Market simulations also indicate that both groups of respondents are unlikely to trade-off
delivery reliability for security. Tables 4.18 and 4.19 depict these comparisons for

concerned and unconcerned respondents respectively.

96

 

TABLE 4.18 SECURITY BY DELIVERY RELIABILITY INTERACTION:
MARKET PREFERENCE COMPARISON FOR CONCERNED RESPONDENTS

 

Concerned Respondents

 

 

 

 

 

 

 

Delivery Delivery
Security Reliability Market Directional Market Reliability Security
Level Level Preference Relationship Preference Level Level
. Equal to or o 0 1-2%
Basrc Above 28.31 /o > 15 .64 /o Below Advanced
Implication

 

 

 

Concerned respondents chose a supplier with basic security/equal to or above delivery reliability

 

more frequently than one with advanced security/14% below average delivery reliability.

lath

‘1 roam-epis-w'. yq. . tl“

""‘-,‘,fl' .,.V~-

 

Delivery

 

 

 

 

 

 

bass-y ’ A" '
Security Reliability Market Directional Market Reliability Security
Level Level Preference Relationship Preference Level Level
. Equal to or o o 3% or more
Basrc Above 28.31 /o > 3.23 /o Below Advanced
Implication

 

 

 

Concerned respondents chose a supplier with basic security/equal to or above delivery reliability

more frequently than one with advanced security/3% or more belo

........

1‘1. mgr—err
. . ,,.
, .

w average delivery reliability.

:7 Trig-.- gm, .4; .

 

 

 

 

 

 

 

Delivery Delivery
Security Reliability Market Directional Market Reliability Security
Level Level Preference Relationship Preference Level Level
Basic l-2% Below 5.00% > 3.23% 3% or more Advanced
Below
Implication

 

 

 

Concerned respondents chose a supplier with basic security/l-2% below delivery reliability more
frequently than one with advanced security/3% below average delivery reliability.

 

97

 

TABLE 4.19 SECURITY BY DELIVERY RELIABILITY INTERACTION:
MARKET PREFERENCE COMPARISON FOR UNCONCERNED

 

 

 

 

 

 

 

 

RESPONDENTS
Unconcemed Respondents
Delivery Delivery
Security Reliability Market Directional Market Reliability Security
Level Level Preference Relationship Preference Level Level
. Equal to or 1-2%
Basrc Above 32.1 1% > 18.97% Below Advanced
Irmlication

 

 

 

Unconcemed respondents chose a supplier with basic security/equal to or above delivery reliability

 

 

 

 

 

 

 

 

Umorefrequently than onetrv‘vit'h advanced security/L296 below average delivery reliability.
Delivery " " "bangs“ ‘ " '
Security Reliability Market Directional Market Reliability Security
Level Level Preference Relationship Preference Level Level
. Equal to or o o 3% or more
Basrc Above 32.1 1 /o > 1.94 A: Below Advanced
Implication

 

 

 

Unconcemed respondents chose a supplier with basic security/equal to or above delivery reliability

 

.v
-..4....w

more frequently than one with advanced security/3% or more below average delivery reliability.

. vvx‘l",dwv'r‘x‘ltl .w —H— ' my m 7 - "TW-x‘a"v'-r-V\—' .-..,._mm‘-1xa--v:~~~ *-

 

 

 

 

 

 

 

Delivery Delivery
Security Reliability Market Directional Market Reliability Security
Level Level Preference Relationship Preference Level Level
0
Basic 12% Below 4.90% > 1.94% 3 ”’8‘” ”‘0‘" Advanced
elow
Implication

 

 

 

 

Unconcemed respondents chose a supplier with basic security/1-2% below delivery reliability more
frequently than one with advanced security/3% below average delivery reliability.

 

Tables 4.18 and 4.19 indicate suppliers with basic security and higher delivery reliability
garner more market preference than suppliers with advanced security and lower delivery
reliability for both groups. This indicates concerned and unconcerned respondents are
more likely to choose a supplier with basic security and higher delivery reliability over a
supplier with advanced security and lower delivery reliability.

In summary, H3 posits that concerned respondents have greater preference for
suppliers with lower delivery reliability and higher security. Factor importance scores do
not support H3 and indicate concerned respondents placed more importance on delivery

reliability than did unconcerned respondents. Utility figures support Hg and indicate that

98

"‘r‘1-1._, .,Jg.'vv.

 

concerned respondents prefer a supplier with advanced security and three percent or more
below industry average delivery reliability while unconcerned respondents did not (note:
Significant differences between concerned and unconcerned respondents utility values
cannot be assessed because they were derived from separate lo git runs. Only directional
preferences can be assessed). Market simulation results do not support Hg and indicate
that concerned respondents are less likely to choose a supplier with lower delivery
reliability than unconcerned respondents.

Hg is not supported. Two out of three evaluation methods do not support the
hypothesized relationship. Factor importance scores indicate that concerned respondents
place more importance on delivery reliability than unconcerned respondents. Utilities
indicate concerned respondents prefer a supplier with advanced security and three percent
or more below industry average delivery reliability, whereas unconcerned respondents do
not. Market simulations indicate that both groups are unwilling to trade-off delivery
reliability for security.

H9 examines the influence of concern on respondents’ preference for suppliers
with higher prices and higher security.

H9: An increase in concern over security incidents experienced by the
purchasing firm positively affects preference for suppliers with higher
prices and higher security.
H9 proposes that concerned respondents are more willing to pay a higher price to utilize
suppliers with advanced security than unconcerned respondents. H918 evaluated Similarly

to Hg. Utilities and market simulation results are presented in Table 4.20.

99

TABLE 4.20 HYPOTHESIS 9 RESULTS: SECURITY BY PRICE BY
INCIDENT CONCERN INTERACTION

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Concerned Unconcemed
Respondents Respondents
Market Market

Security Preference Preference

Level Price Level Utilities % Utilities %
Basic Epual to or below 0.063 23.29 0.091 31.36
Basic l-2% higher (0.061) 7.52 (0.083) 6.54
Basic 3% or more higher (0.002) 2.38 (0.008) 1.79
Advanced Equal to or below @063) 37.42 (0.091) 36.70
Advanced 1-2% higher 0.061 24.13 0.083 20.04
Advanced 3% or more higher 0.002 5.27 0.008 3.57

TABLE 4.21 HYPOTHESIS 9 RESULTS: ASSESSMENT OF SIGNIFICANT

DIFFERENCES BETWEEN UTILITY VALUES: CONCERNED

RESPONDENTS

 

 

 

 

 

 

 

 

 

 

Security Level Basic Basic Basic Adv Adv Adv

Security Price Level Equal to 1-2% 3% or Equal 1-2% 3% or

Level or higher more to or higher more
below higher below higher

Basic Equal to or below 5

Basic l-2% higher 1.24 i'f'f‘WW

Basic 3% or more higher 0.61 0.56 M...

Adv Equal to or below 1.26 0.02 0.58 2'77". .

Adv 12% higher 0.02 1.21 0.59 1.24 ‘fTiL‘TfJJZT'jTT:

Adv 3% or more higher 0.58 0.59 0.03 0.61 0.56 "‘

 

 

 

 

 

 

 

100

 

TABLE 4.22 HYPOTHESIS 9 RESULTS: ASSESSMENT OF SIGNIFICANT
DIFFERENCES BETWEEN UTILITY VALUES: UNCONCERNED

 

 

 

 

 

 

 

 

 

 

RESPONDENTS
Security Level Basic Basic Basic Adv Adv Adv

Security Price Level Equal to 1-2% 3% or Equal 1-2% 3% or

Level or higher more to or higher more
below higher below higher

Basic Equal to or below E

Basic 12% higher 1.57 ;”;"‘ffif;jgl....;.’.'.‘..i..:

Basic 3% or more higher 0.81 0.62 iflji'ft'if‘t'1jiif

Adv Equal to or below 1.64 0.07 0.68 """" .Lf;T;lf‘,;j_

Adv l-2% higher 0.07 1.49 0.74 1.57 Diff""it;

Adv 3% or more higher 0.68 0.74 0.12 0.81 0.62 .W- j ' '

 

 

 

 

 

 

 

 

Factor importance scores presented in Table 4.13 reveal that concerned

respondents place less importance on price (score = 19.73 percent) than do unconcerned

respondents (score = 25.15 percent). Utilities indicate that both groups Share the same

preference patterns. Figures 4.12 and 4.13 graphically depict the utility preference

patterns for concerned and unconcerned respondents respectively. Again, the magnitude

of utility figures between concerned and unconcerned respondents cannot be compared.

101

 

FIGURE 4.12 SECURITY BY PRICE BY INCIDENT CONCERN
INTERACTION: UTILITIES FOR CONCERNED RESPONDENTS

 

 

 

 

 

 

'0
t-Z
8
2
o.
2
O
E D Concerned/Basic ‘
l Conerned/Advanced 1
'U
8
.2
1 2
n.
1" 1
E Equal to or below 1-2% higher 3% or more higher 1

FIGURE 4.13 SECURITY BY PRICE BY INCIDENT CONCERN
INTERACTION: UTILITIES FOR UNCONCERNED RESPONDENTS

 

More Preferred

13 banshee/ease 7 7
lvUnconerned/Adyaniced 1

 

 

 

 

 

 

‘ Less Prefemed

Equal to or below

1-2°/o higher

3% or more higher

Figure 4.14 graphically depicts market simulation results for the three—way
interaction between, security, price, and incident concern. When a supplier possesses

advanced security and charges a price equal to or below industry average, they will

102

garner marginally more market preference from concerned respondents (market
preference = 37.42 percent) than unconcerned respondents (market preference = 36.70
percent). Additionally, unconcerned respondents are more likely to choose a supplier
with basic security (market preference = 31.36 percent) than concerned respondents
(market preference = 23.29 percent). However, when price increases to one to two
percent above industry average, the gap between the groups’ likelihood of choosing a
supplier with advanced security widens. Concerned respondents are more likely to
choose a supplier with advanced security and one to two percent higher prices (market
preference = 24.13 percent) than are unconcerned respondents (market preference =
20.04 percent). At three percent or more higher price, the gap between the groups’
likelihood of choosing advanced security narrows somewhat, but concerned respondents
are still more likely to choose a supplier with advanced security (market preference =
5.27 percent) than are unconcerned respondents (market preference = 3.57 percent).

FIGURE 4.14 SECURITY BY PRICE BY INCIDENT CONCERN
INTERACTION: MARKET PREFERENCE

 

 

 

 

 

 

 

 

1' 40.00%

‘ -‘~:\
‘ 30.00% ~-—-——A~.; "7:\ — — , _ __ E_E_ _ 1
1 ‘° '\. : \D 3 + Concerned/Basic .
‘ ° 2- -D- - Concerned/Advanced ll
1 20.00% -—~-~ ‘ . 1'
‘n-An- Unconcemed/Basrc .
1 5 T f 5 2:99.92? E'EESVAQYE‘QC? Edi ‘
10.00% ~ - ~~~ 1
i 1
1
0.00% 1 . ‘
Equal to or 1-2% higher 3% or more ‘

below higher

103

 

Market simulations also reveal that concerned respondents were marginally more
willing to choose a supplier with advanced security and one to two percent higher prices
over one with basic security and equal to or below prices, which indicates a greater
willingness to trade-off price for security. Unconcemed respondents were not as willing
to trade-off price for security. Tables 4.23 and 4.24 depict these relationships for
concerned and unconcerned respondents respectively.

TABLE 4.23 SECURITY BY PRICE BY INCIDENT CONCERN

INTERACTION: MARKET PREFERENCE COMPARISON FOR CONCERNED
RESPONDENTS

 

Concerned Respondents

 

 

 

 

 

 

 

Security Market Directional Market Security
Level Price Level Preference Relationship Preference Price Level Level
. Equal to or o o 1-2%
Basrc below 23.29 /o < 24.13 A) higher Advanced
Implication

 

 

 

Concerned respondents chose a supplier with basic security/equal to or below prices less frequently

 

............... .——r

than one with advanced security/1-2% higher prices.

a~-m-.AW.‘.;.—'-‘._ twig“, m- 7 . m~.um , f. .-. , , .

 

Security Market Directional Market

 

 

 

 

 

 

 

Security N 3
Level Price Level Preference Relationship Preference Price Level Level
. Equal to or o 0 3% or more
Basrc below 23.29 /0 > 5.27 /0 higher Advanced
Implication

 

 

Concerned respondents chose a supplier with basic security/equal to or below prices more frequently
than one with advanced (security/3% or more higher prices.

 

49-4.. L--.) l.rf‘.t-v~vu-|g -r. .....~.-.:..«. .7. r 3.5.”: u ,5. ,- . ..~ 1 41.“- 9: '

""""""""" . *--"*1\h~\tfihx.x'~.‘m":’ -‘Lwao v 4..jt .4-...-ta.. - ‘ y '-'

 

Security Market Directional Market

 

 

 

 

 

 

‘ Security
Level Price Level Preference Relationship Preference Price Level Level
. 1-2% 0 0 3% or more
Ba81c higher 7.52 /o > 5.27 /o higher Advanced
Implication

 

 

 

Concerned respondents chose a supplier with basic security/l-2% higher prices more frequently than
one with advanced security/3% or more higher prices.

 

 

104

TABLE 4.24 SECURITY BY PRICE BY INCIDENT CONCERN
INTERACTION: MARKET PREFERENCE COMPARISON FOR
UNCONCERNED RESPONDENTS

 

Unconcerned Respondents

 

 

 

 

 

 

 

Security Market Directional Market Security
Level Price Level Preference Relationship Preference Price Level Level
. Equal to or 0 o 1-2%
Basrc below 31.36 /0 > 20.04 /o higher Advanced
Implication

 

 

 

Unconcemed respondents chose a supplier with basic security/equal to or below prices more

 

frequentlythanone with advanced security/14% higher prices. -

 

 

 

 

 

 

 

“Security Mzket i Directional Market 7 “J U M if ruSecurityfl
Level Price Level Preference RelationshiL Preference Price Level Level
. Equal to or 0 o 3% or more
Ba81c below 31.36 /o > 3.57 /o higher Advanced
Implication

 

 

 

Unconcemed respondents chose a supplier with basic security/equal to or below prices more

 

j—v——
a

frequently than one with advanced security/3% or more higher prices.

 

 

 

 

 

 

 

"Security A i H 3 Market Directional ‘ Market A i r i . -. Security;
Level Price Level Preference Relationship Preference Price Level Level
_ o 0
Basic bl. 2 /" 6.54% > 3.57% 3 A’ 9’ "‘0“ Advanced
lgher hlgher
Implication

 

 

 

 

Unconcemed respondents chose a supplier with basic security/1-2% higher prices more frequently
than one with advanced security/3% or more higher prices.

 

Table 4.23 indicates that concerned respondents are more likely to choose a
supplier with advanced security and one to two percent higher prices over one with basic
security and equal to or below prices. Concerned respondents are also more likely to
choose a supplier with basic security and equal to or below average prices over a supplier
with advanced security and three percent or more higher prices. Further, concerned
respondents are more likely to choose a supplier with basic security and one to two
percent higher prices over one with advanced security and three percent or more higher
prices. Unconcemed respondents are always more likely to choose a supplier with basic

security and lower prices over one with advanced security and higher prices. These

105

 

results indicate concerned respondents are more willing to trade-off price for security
than unconcerned respondents.

In summary, H9 states that concerned respondents have greater preference for
suppliers with higher prices and higher security. Factor importance scores support H9 and
indicate concerned respondents place less importance on price than unconcerned
respondents. Utility results do not support H9 and indicate that preference patterns for
both groups were the same, implying that concerned respondents and unconcerned
respondents have similar preferences for all levels of security and price. Market
simulations support H9 and indicate that concerned respondents are more likely to choose
a supplier with one to two percent higher price and advanced security, whereas
unconcerned respondents were not as likely to sacrifice price for security.

H9 is partially supported. Concerned respondents place less importance on price.
Concerned respondents are more likely to trade-off price for security. Utility preferences
are the same.

H10 through H12 examine the influence of international sourcing on respondents’
preference for suppliers with advanced security, lower delivery reliability, and higher
prices. H10 examines the influence of international sourcing as it affects preference for
suppliers with advanced security. These results are presented subsequently.

RQ8: Do purchasing firms that source goods internationally place more value on

increased supplier security?

H10: International sourcing by the purchasing firm positively affects
preference for suppliers with higher security competence.

106

Returning to the full sample (i.e. not segmenting respondents by concern over
security incidents), H10 proposes that international sourcing positively affects preference
for suppliers with higher security competence. H10 is evaluated by examining utilities
and market simulation results comparing basic and advanced security across the two
supplier locations. For H10 to be fully supported, the following must be true: 1) utility
results must indicate respondents derive more utility from advanced security, as opposed
to basic security, when they source internationally and 2) market simulation results must
indicate that respondents are more likely to choose a supplier with advanced security
when they source goods internationally than when they source goods domestically.
Utility figures were derived from the two-way interaction of security and supplier
location. Market simulation results were derived by performing two separate
simulations. One simulation examined security and domestic sourcing while holding
price, delivery reliability, and quality at their “best” levels. The second simulation
examined security and international sourcing while holding price, delivery reliability, and
quality at their “best” levels (note: two separate market simulations were run — one for
domestic and one for international. This was necessary because respondents’ strong
preference for domestic sourcing skewed the results and masked the effect of security).

Table 4.19 provides utility and market simulation results related to H10.

107

TABLE 4.25 HYPOTHESIS 10 RESULTS: SECURITY BY SOURCING
LOCATION INTERACTION

Level Utilities Utilities

1 . 0.102 .27
l 73

 

TABLE 4.26 HYPOTHESIS 10 RESULTS: ASSESSMENT OF SIGNIFICANT
DIFFERENCES BETWEEN UTILITY VALUES: UNCONCERNED
RESPONDENTS

    

at p < 0.05

Utilities indicate that when firms source domestically, they significantly prefer
suppliers with basic security over advanced security. When firms source internationally,
they significantly prefer suppliers with advanced security over basic security. Figure

4.15 depicts these results graphically.

108

FIGURE 4.15
UTILITIES

SECURITY BY SOURCING LOCATION INTERACTION:

 

More Preferred

 

 

 

 

Less Preferred

  

Advanced

Cl Domestic i l
I International.
1

 

Market simulations reveal that respondents are more likely to choose a supplier with

advanced security when they source internationally than they are when they source

domestically. These results are depicted graphically in Figure 4.16.

FIGURE 4.16 SECURITY BY SOURCING LOCATION INTERACTION:
MARKET PREFERENCE

 

80.00%

 

70.00%
60.00%
50.00%
40.00%
30.00%
20.00%
10.00%

0.00%

 

l ‘—-—0— Domestic

 

 

 

Basic Adwnced

- -Cl~ — international

 

 

109

In sum, H10 states that international sourcing positively affects preference for
suppliers with advanced security. Utility figures support H10 and indicate greater
preference for advanced security when goods are sourced internationally. Market
simulations support H10 and indicate that respondents choose a supplier with advanced
security when they source goods internationally more often than when they source
domestically.

H10 is supported. Utilities indicate respondents prefer advanced security when
they source internationally and basic when they source domestically. Market simulations
indicate respondents are more likely to choose a supplier with advanced security when
they source internationally than when they source domestically. The two methods of
evaluation are congruent with each other and the hypothesized relationship.

H“ examines the influence of international sourcing on respondents’ preference
for suppliers with lower delivery reliability and higher security.

H11: International sourcing by the purchasing firm positively affects

preference for suppliers with lower delivery reliability and higher

security.
H11 proposes that firms sourcing internationally prefer suppliers with lower delivery
reliability and higher security. H” requires the examination of a two-way interaction
between delivery reliability and supplier location and a three-way interaction between
security, delivery reliability, and supplier location. For H” to be supported, the following
must hold true: 1) when comparing utility figures for domestic vs. international supplier
locations, respondents must prefer lower delivery reliability for international sourcing

and higher delivery reliability for domestic sourcing (note: utilities are not examined for

110

three-way interactions due to CBC’s inability to estimate these results), and 2) market
simulations must reveal that suppliers who source goods internationally are more likely to
choose a supplier with higher security and lower delivery reliability than their domestic
counterparts (note: two separate market simulations were run —- one for domestic and one
for international. This was necessary because respondents’ strong preference for
domestic sourcing skewed the results and masked the effect of security). Utility and
market simulation results are presented in Tables 4.27 and 4.29 respectively.

TABLE 4.27 HYPOTHESIS 11 RESULTS: DELIVERY RELIABILITY BY
SOURCING LOCATION INTERACTION

 

 

 

 

 

 

 

 

 

Domestic International

Sourcing Sourcing
Delivery Reliability Level Utilities Utilities
Equal to or Above 0.049 (0.049)
1-2% Below (0.128) 0.128
3% or more Below 0.079 (0.079)

 

111

TABLE 4.28 HYPOTHESIS 11 RESULTS: ASSESSMENT OF SIGNIFICANT
DIFFERENCES BETWEEN UTILITY VALUES

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Delivery Equal Equal 1-2% 1-2% 3% or 3% or
Reliability to or to or below below more more
Level above above below below
Te ":3 E
.2 .5 .2 .5 .2 .2
a ’5 a ’5 E is}:
Delivery Sourcing E E E E) 2‘5, E
Reliability Level Location Q ’5‘ Q 5 Q '5
Egal to or above Domestic
Equal to or above International 1.21
1-2% below Domestic 2.17* 0.97
1-2% below International 0.97 2.17* 3.09*
3% or more below Domestic 0.33 1.37 2.19* 0.51 ,
3% or more below International 1.37 0.33 0.51 I 2.19* 1.51

 

* = Significant difference at p < 0.05

TABLE 4.29 HYPOTHESIS 11 RESULTS: SECURITY BY DELIVERY

RELIABILITY BY SOURCING LOCATION INTERACTION: MARKET
PREFERENCE

Level Market Preference % Market Preference %
to or 20.52

1- 5.11 3.29
or more 1.65 1.13
to or 52.33

1- 16.90 19.55
or more 70 3.1.

 

Utilities in Table 4.27 and 4.28 reveal that respondents significantly prefer one to
two percent lower delivery reliability when they source goods internationally versus
when they source domestically. Respondents derive more utility from the other two

levels when they source domestically. These results are depicted graphically in Figure

4.17.

112

 

FIGURE 4.17 DELIVERY RELIABILITY BY SOURCING LOCATION
INTERACTION: UTILITIES

 

 

 

.5 1
a: I
I- l
l 5 ‘
,2 l
1 '3 l
L-
n. 1
1 o i
3 1
E l a * ~
5 1:1 Domestic
1 .
.1 I lntematlonal .1
1 ,
1 l
1
1: l i
0 l
h .
h 1
.2 .
2 l
o. LL
In
E Equal to or above 1-2% below 3% or more below

Market simulation results graphically depicted in Figure 4.18 reveal that when delivery
reliability is equal to or above industry average, both international and domestic suppliers
with advanced security will garner more market preference than suppliers with basic
security. This relationship holds for all three levels of delivery reliability. However, the
market preference percentage gap between advanced security and basic security for
international suppliers is larger than the gap between advanced security and basic security
for domestic suppliers. This indicates that security is more important to respondents
when they source internationally. Note again that two separate market simulations were
run — one for domestic and one for international — in order to focus on the effects of
security on supplier preference. Unfortunately, this prevents one from stating that an
international supplier with advanced security will garner more market share than a
domestic supplier with advanced security as Figure 4.18 seems to imply. Given
respondents’ strong preference for domestic suppliers (see utilities in Table 4.5) this is

not likely to be realistic.

113

FIGURE 4.18 SECURITY BY DELIVERY RELIABILITY BY SOURCING
LOCATION INTERACTION: MARKET PREFERENCE

 

 

 

 

 

60.00%
1
50.00% - Ax, e e ~~ ,, , a , W e I--- -~ ~
0 7 , _ .--_._.__‘__-__--__ _.
40'00 /° i—o— Domestic/Basic .
| ' 'l
30.00% 7 _ i——<>——Domest1c/Advanced ‘1
------A lntemationaVBasic .
l .
' 20.00% Lief;J"tE'Da}i°“a'/Adva”ced L
10.00% ’
0.00% .
Equal to or 1-2% below 3 % or more
above below

Market simulation results also reveal that respondents are more likely to choose
an international supplier with basic security and higher delivery reliability over an
international supplier with advanced security and lower delivery reliability. These results
are depicted in Table 4.30. However, comparing the market preference percentages for
domestic and international suppliers with similar attributes reveals that respondents place
more importance on security when they source internationally. These results are depicted

in Table 4.31.

114

TABLE 4.30 SECURITY BY DELIVERY RELIABILITY INTERACTION FOR
INTERNATIONAL SOURCING: MARKET PREFERENCE COMPARISON I

 

International Sourcing

 

 

 

 

 

 

 

Delivery Delivery
Security Reliability Market Directional Market Reliability Security
Level Level Preference Relationship Preference Level Level
. Equal to or o 1-2%
Bas1c . Above 20.52 /o > 19.55% Below Advanced
Implication

 

 

 

Respondents who source goods internationally chose a supplier with basic/equal to or above delivery

 

'5"

. 2 ~
‘_4uv-~ .-‘¢v+h '__...1>

A...- 1c-v ~u‘j -- t: ‘v

reliability more frequently than one with advanced/1-2% below average delivery reliability.

.7 .-

 

Delivery

Delivery

 

 

 

 

 

 

Security Reliability Market Directional Market Reliability Security
Level Level Preference RelationshiL Preference Level Level
. Equal to or o o 3% or more
Basrc Above 20.52 A) > 3.19 A) Below Advanced
Implication

 

 

 

Respondents who source goods internationally chose a supplier with basic/equal to or above delivery

ith advanced/3% or more

 

tints Y‘n1- - ' 1.3.3.9121; .1

uh:

reliability more frequently than one w

ugvvm h-.-—' 0“: sinrar-'

tS-‘I’AVJVV (A A‘J’; In I“:

he}. ._.‘-(- N

Ir- h‘l am}. 1.,

..A-rl ‘

V M. ‘1' 'rkufl' ‘1"‘ LLJI ‘

below averagdelivery reliability.

 

 

 

 

 

 

 

" " 021125; ‘ '" W ._ ' "' 1521128}? " ‘ ' i "
Security Reliability Market Directional Market Reliability Security
Level Level Preference Relationship Preference Level Level
. 0 o 0 3% or more
Basrc 1-2 /o Below 3.29 /o > 3.19 /o Advanced
Below
Implication

 

 

 

 

Respondents who source goods internationally chose a supplier with basic/l -2% below average
delivery reliability more frequently than one with advanced/3% or more below average delivery

reliability.

 

115

main.

 

TABLE 4.31

SECURITY BY DELIVERY RELIABILITY BY SOURCING
LOCATION INTERACTION: MARKET PREFERENCE COMPARISON II

 

Domestic Sourcing

International Sourcing

 

 

 

 

 

 

 

 

 

Delivery Delivery
Security Reliability Market Directional Market Reliability Security
Level Level Preference Relationship Preference Level Level
Equal to or o 0 Equal to or
Advanced Above 43.36 /o < 52.33 A: Above Advanced
Implication

 

Respondents who source internationally chose a supplier with advanced security/equal to or above

 

average delivery reliability more frequently than respondents who source domestically.
he: 32"»; '52“. 75,: :f.w:-n ‘‘‘‘‘‘ r 1 ‘ ‘

 

 

 

 

 

 

 

 

 

.. Deliver; ...... ,.,__.. .. -. -. Delivery .
Security Reliability Market Directional Market Reliability Security
Level Level Preference Relationship Preference Level Level

_ 0
Advanced 1-2% Below 16.90% < 19.55% 13 2 A Advanced
elow
Implication

 

Respondents who source internationally chose a supplier with advanced security/1-2% below average

 

gr... . .7}. - . . r1..- ..-. «v

\‘1 . ---- t
v . ,.

35'. . 35' ‘0'.“

3,-5.1'rhg *4 .‘afifs'i)_‘.fi‘3‘t“£'

$3“fo xt‘u'fd 1“} ',P."Y5'~I.; '1 '3“

delivery reliability more frequently than respondents who source domestically.

 

 

 

 

 

 

 

 

 

H ' Delivery. ‘ . i i A Delivery H T i i i L
Security Reliability Market Directional Market Reliability Security
Level Level Preference Relationship Preference Level Level
0 0 0
Advanced 3 A: or more 2.70% < 3.19 /o 3 /o or more Advanced
Below Below
Implication

 

 

Respondents who source internationally chose a supplier with advanced security/3% or more below
aveLage delivery reliability more frequently than respondents who source domestically.

 

Results in Table 4.30 indicate that firms who source internationally are more

likely to choose a supplier with basic security and higher delivery reliability over one

with advanced security and lower delivery reliability. However, Table 4.31 indicates that

security is more important to firms when they source internationally than it is when they

source domestically. If a firm purchases product internationally, they are more likely to

utilize a supplier with advanced security than if they source domestically. This indicates

that international sourcing has a positive effect on firm’s preference for suppliers with

greater security.

116

 

In sum, H” states that firms who source internationally are willing to accept lower
delivery reliability from their suppliers in return for higher security (note: respondents
may have no choice but to accept lower delivery reliability when they source
internationally. The difference between expectations and actual preferences will be
discussed in the limitations section of Chapter Five). Utilities support H11 and indicate
that respondents are willing to accept one to two percent lower delivery reliability from
their international suppliers. Market simulation results partially support H11. On one
hand, when sourcing internationally, respondents are more likely to choose a supplier
with higher delivery reliability and lower security (see Table 4.30). On the other hand,
assuming delivery reliability levels are the same, respondents are more likely to choose a
supplier with advanced security when they source internationally than they are when
sourcing domestically (see Table 4.31).

H11 is partially supported. Utilities indicate preference for one to two percent
lower delivery reliability when sourcing internationally. Market simulation indicates that
respondents are more likely to choose a supplier with basic security and higher delivery
reliability when they source internationally. They are also more likely to choose a
supplier with advanced security when they source internationally versus one with
advanced security when they source domestically (assuming delivery reliability levels for
domestic and international suppliers are equal).

H12 examines the influence of international sourcing on respondents’ preference
for suppliers with higher prices and higher security.

H12: International sourcing by the purchasing firm positively affects
preference for suppliers with higher prices and higher security.

117

H12 proposes that firms who source internationally prefer suppliers with higher prices and
higher security. H12 requires the examination of a two-way interaction between price and
supplier location and a three-way interaction between security, price, and supplier
location. H12 is evaluated in the same way as H“. Utility and market simulation results
are presented in Tables 4.32 and 4.34 respectively.

TABLE 4.32 HYPOTHESIS 12 RESULTS: PRICE BY SOURCING
LOCATION INTERACTION

to or

1-
3% or more

 

TABLE 4.33 HYPOTHESIS 12 RESULTS: ASSESSMENT OF SIGNIFICANT
DIFFERENCES BETWEEN UTILITY VALUES

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Price Level Equal Equal 1-2% 1-2% 3% or 3% or
to or to or above above more more
below below above above

Tu Tc '5
.2 .5 .2 .5 .2 .5
8 w 8 N 8 2
Sourcing 5 E E E) E .3}.
Price Level Location Q E Q ’E‘ Q 5
Equal to or below Domestic
Equal to or below International 0.77
1-2% above Domestic 0.78 0.04
1-2% above International 0.04 0.78 0.80
3% or more above Domestic 0.31 0.38 0.40 0.33
3% or more above International 0.38 0.31 0.33 I 0.40 0.06 . ’5'

 

 

118

TABLE 4.34 HYPOTHESIS 12 RESULTS: SECURITY BY PRICE BY
SOURCING LOCATION INTERACTION

Market
Price Level %
to or

6.82
or more 2.03

to or 37.1 1
-2% 22.50
or more 4.

 

Utilities indicate that respondents derive more utility from one to two percent
higher prices when they source goods internationally but the difference is not significant.
Respondents derive more utility from the other two levels when they source goods
domestically but, again, the difference is not significant. These results are presented
graphically in Figure 4.19.

FIGURE 4.19 PRICE BY SOURCING LOCATION INTERACTION:
UTILITIES

 

    

 

 

 

U

0

i: ,i, L I L, L I L

.92

9.’

n. L

2

O

E .f .
CI Domestic
l Internationali

‘D

a:

t

.9.’ ,

2

0. Wm . . , p . . .

‘3

3 Equal to or below 1-2% higher 3% or more higher

Market simulation results presented in Figure 4.20 reveal that at prices equal to or below

industry average, respondents are more likely to choose advanced security for both

119

domestic and international suppliers. This relationship holds for all three levels of price.
However, the gap between market preferences for international suppliers with advanced
security and international suppliers with basic security is larger than the gap between
market preferences for domestic suppliers with advanced security and domestic suppliers
with basic security. Again, this underscores the importance respondents place on security

when they source internationally.

FIGURE 4.20 SECURITY BY PRICE BY SOURCING LOCATION
INTERACTION: MARKET PREFERENCE

‘ 50.00%

 

40.00%

 

 

 

 

 

——<>— Domestic/Advanced
o l...‘... lntemational/Basic
. 20.002 _ ‘ é:--A-:-_histnationaVAdYeDEed,
l 10.00% i
0.00%

 

 

Equal to or below 1-2% higher 3% or more higher

 

 

Market simulations also indicate that respondents are more likely to choose a supplier
with advanced security and one to two percent higher prices over one with basic security
and equal to or below prices when they source internationally. Further, respondents are
more likely to choose a supplier with advanced security and three percent or more higher
prices than one with basic security and one to two percent higher prices when they source

internationally. These results are presented in Table 4.35.

120

E 30.00% - \ —+— Domestic/Basie ' '1

l

l

Comparing market preferences for similar domestic and international suppliers

reveals that respondents are more likely to choose advanced security when they source

internationally as opposed to when they source domestically. These results are presented

in Table 4.36.

TABLE 4.35 SECURITY BY PRICE BY SOURCING LOCATION
INTERACTION: MARKET PREFERENCE COMPARISON I

 

International Sourcing

 

 

 

 

 

 

 

 

 

Security Market Directional Market Security
Level Price Level Preference Relationship Preference Price Level Level
. Equal to or o o 1-2%
Basrc Below 18.33 /o < 25.94 /0 Higher Advanced
Implication

 

Respondents who source goods internationally chose a supplier with advanced/ 1-2 °/o higher prices

 

..¢.2‘.I.‘, .".._.-..s.-o¢ o‘i. 'J J '"

pnq-".fi. erratic: a 4 -I'-’

r 1L1} ' 4".43-c

-v"‘-—- m_‘r---‘

.Vn‘m....‘

T I 7.:
. I I-

‘1

 

Market

 

 

 

 

 

 

 

 

Security Market“ Directional i J . Security,
Level Price Level Preference Relationship Preference Price Level Level
0
Basic Equal to or 18.33% > 5.28% 3 /° or more Advanced
Below Higher
Implication

 

Respondents who source goods internationally chose a supplier with basic/equal to or below average

equently than‘one with advanced/3% orfimporehigher prices.

 

prices more fr

12‘1“; i..fll-\ ' - a v e n ‘. .. "

-— uni-o" n—n‘

-Imw mml'm*‘-

| r ‘1 L‘lmu’ "'7‘

'1 a; HIS A iL'l‘

 

 

 

 

 

 

 

 

 

Security ii IIIIIII Market Directiona Market!” - V Security“ I
Level Price Level Preference Relationship Preference Price Level Level
. . 3% or more
_ o o o
Basrc 1 2 /o Higher 4.39 /o < 5.28 A: Higher Advanced
Implication

 

 

more frequently than a supplier with basic/1-2% higher prices.

Respondents who source goods internationally chose a supplier with advanced/3% higher prices

 

121

 

TABLE 4.36 SECURITY BY PRICE BY SOURCING LOCATION
INTERACTION: MARKET PREFERENCE COMPARISON II

 

 

 

 

 

 

 

 

 

Domestic Sourcing International Sourcing
Security Market Directional Market Security
Level Price Level Preference Relationship Preference Price Level Level
Advanced Equal t° 0’ 37.11% < 44.74% Equal t° 0‘ Advanced
Below Below
Implication

 

 

 

Respondents who source internationally are more likely to choose a supplier with advanced
securrty/eppal to or below average prices than respondents who source domestrcallv

 

..... 'th’ 'c'Iasl1),’:V-V'v.ovow—,‘.'i"--.-

-------- -.-.'.-2 v": 5.1-"urn a" wave '5'“ ‘ ~.'. -: VT 19'; .-,e.r. .. :r w. -. .m- v xxx 'u . .-. 5.». 'r ~5h-13‘.'.'.-.',.'

.n- ‘
x_--'

 

 

Security Market Directional Market Security i
Level Price Level Preference Relationship Preference Price Level Level
0 _ 0
Advanced 1-2% Higher 22'5” A’ < 25.94% P; :11: Advanced

 

 

 

 

 

 

 

Implication

 

Respondents who source internationally are more likely to choose a supplier with advanced
security/14% h gher prices than respondents who source domestically

 

..... 71.7.... ,,.. ..,.,‘F . »
.......... M_.{H_-,-J....s',-_rl_ .2 u: “5"" --o~.‘~“,‘I','..'_.I-x'l~"\' -.

. . r ‘. . i ~~~~~ ". . , . ,.
D’s: wwwwwwwww -- ""r- ' “Err-f n. b 'M‘Ivm u “M 1m ww~1~nv.~. ~ox;g-—-.vv . “4'5 .,.,.-4,

 

 

 

 

 

 

 

Security Market Directional ' iMiarket Security, A,
Level Price Level Preference Relationship Preference Price Level Level
0 0 0 0
Advanced 3 /o or more 4.48 /o < 5.28 /o 3 A) or more Advanced
Higher Higher
Implication

 

 

 

Respondents who source internationally are more likely to choose a supplier with advanced
security/3% or more higher prices than respondents who source domestically.

 

 

Results in Table 4.35 indicate that when firms choose between intemational
suppliers, they are more likely to choose a supplier with advanced security and one to two
percent higher prices over a supplier with basic security and equal to or below prices.
Further, Table 4.35 reveals that when firms choose between international suppliers, they
are more likely to choose a supplier with advanced security and three percent or more
higher prices over one with basic security and one to two percent higher prices.

Table 4.36 compares similar domestic and international suppliers. Results
indicate that security is more important to firms when sourcing internationally than it is
when they source domestically. For example, an international supplier with advanced

security and equal to or below prices is predicted to garner more market share than a

122

domestic supplier with advanced security and equal to or below prices. This indicates
that international sourcing has a positive effect on firm’s preference for suppliers with
greater security.

In sum, H12 states that firms who source internationally are willing to accept
higher prices from their suppliers in return for higher security. Utilities do not support
H12 and indicate that respondents do not derive significantly more utility from
international suppliers with higher prices. Market simulations support H12 and indicate
that when respondents source internationally, they are more likely to 1) choose suppliers
with advanced security and one to two percent higher prices over suppliers with basic
security and equal to or below average prices, and 2) choose suppliers with advanced and
three percent or more higher prices over suppliers with basic security and one to two
percent higher prices. Market simulation results also indicate that respondents are always
more likely to choose a supplier with advanced security when they source internationally
as compared to when they source domestically.

H12 is partially supported. Respondents do not derive more utility from suppliers
with higher prices when they source internationally. Market simulations indicate
respondents are more likely to choose a supplier with advanced security and higher prices
over one with basic security and lower prices when they source internationally. Market
simulations also indicate respondents are more likely to choose a supplier with advanced
security when they source internationally than they are when they source domestically.
This concludes the hypothesis evaluation section. The next section of the chapter

summarizes the information presented heretofore.

123

SUMMARY

Chapter Four presented the research results. The Chapter began by detailing the
mailing procedures, sample size, and sample characteristics. A review of the sample
characteristics found that the respondents were from the targeted population and that no
difference exists between the utility patterns of early and late respondents.

The Chapter reviews the hypotheses tests through CBC and presents each
research question and corresponding hypotheses in order. Conclusions were drawn from
the results to provide support, or lack thereof, for each hypothesis. Table 4.37
summarizes the hypotheses and results presented in the chapter.

Examining the results of H1 through H4, respondents prefer “better” levels of each factor
over “worse” levels. These results are intuitive, but necessary to set the stage for H5
through H12.

H5 and H6 examine respondent preferences for security as it relates to price and
delivery reliability. Generally, respondents preferred suppliers with basic security, lower
prices, and higher delivery reliability over suppliers with advanced security, higher
prices, and lower delivery reliability. Interestingly, utility figures show that at certain
levels of price and delivery reliability respondents preferred higher levels of security
indicating that respondents may be willing to trade off these supplier attributes for greater

security.

124

TABLE 4.37 SUMMARY OF RESULTS

 

 

 

 

 

 

 

 

 

 

 

 

 

purchasing firm positively affects
preference for suppliers with
higher prices and higher security.

 

 

 

Hypotheses Supported/Not Supported Rationale (+ = supported; - = not
supported; +/- = mixed results)

HI: Higher product quality is Supported Utilities: +
preferred over lower product
quality.
H2: Lower product price is Supported Utilities: +
preferred over higher product
price.
H3: Higher delivery reliability is Supported Utilities: +
preferred over lower delivery
reliability.
H4: Higher supplier security Supported Utilities: +
competence is preferred over
lower supplier security
cogetence.
H5: Product price has a greater Partially Supported Factor Importance: +
influence on supplier preference Utilities: -
than supplier security Market Simulation: +
competence.
H6: Delivery reliability has a Partially Supported Factor Importance: +
greater influence on supplier Utilities: -
preference than supplier security Market Simulation: +
competence.
H7: An increase in concern over Partially Supported Factor Importance: +
security incidents experienced by Utilities: -
the purchasing firm positively Market Simulation: +
affects preference for suppliers
with higher security competence
and higher security.
H3: An increase in concern over Not Supported Factor Importance: -
security incidents experienced by Utilities: -/+
the purchasing firm positively Market Simulation: -
affects preference for suppliers
with lower delivery reliability and

rh_igher security.
H9: An increase in concern over Partially Supported Factor Importance: +
security incidents experienced by Utilities: —
the purchasing firm positively Market Simulation: +
affects preference for suppliers
with higher prices.
H10: International sourcing by the Supported Utilities: +
purchasing firm positively affects Market Simulation: +
preference for suppliers with

flgher security competence.
HI 1: International sourcing by the Partially Supported Utilities: +
purchasing firm positively affects Market Simulation: -/+
preference for suppliers with
lower delivery reliability and

_lu;gher security.
Hm: International sourcing by the Partially Supported Utilities: -

Market Simulation: +

 

125

 

H7 through He examined the role of respondents’ concern over security incidents
that have occurred at their firm as it relates to preference for suppliers with higher
security competence, higher prices, and lower delivery reliability. Findings indicate that
concerned respondents 1) prefer suppliers with advanced security, 2) place less
importance on price and are more likely to pay a higher price for advanced security, and
3) place more importance on delivery reliability and are not likely to choose a supplier
with low delivery reliability and advanced security.

H10 through H12 examined the role of international sourcing as it relates to
preference for suppliers with advanced security, lower delivery reliability, and higher
prices. Findings indicate that international sourcing positively affects preference for
suppliers with advanced security and higher prices, but the results for suppliers with
lower delivery reliability were not as conclusive.

The data partially or fully support all hypotheses with the exception of Hg.
Results indicate that food purchasing managers are not willing to trade-off price and
delivery reliability for their suppliers to operate in a secure manner unless they source
internationally or are concerned about security incidents at their firm. When sourcing
internationally, respondents place more value on supplier security. Respondents who are
concerned about security incidents that have occurred at their firm also place more value
on supplier security. Results indicate that international sourcing and respondent concern
increase the value respondents place on security to the extent that they are willing to pay
one to two percent higher prices in return for increased supplier security. However,
international sourcing and respondent concern do not increase the value respondents

place on security to the extent that they are willing to trade-off delivery reliability for

126

security. Chapter Five discusses the theoretical and managerial implications of these

findings as well as the contributions of this study and directions for future research.

127

CHAPTER 5: IMPLICATIONS AND FUTURE RESEARCH

INTRODUCTION
Chapter Five presents the implications and limitations of this research as well as
directions for future research. First, managerial and public policy implications are
discussed. Second, theoretical implications are presented. Next, recommendations are
outlined. This research concludes with limitations and directions for future

investigations.

IMPLICATIONS

The premise of this research is that suppliers seeking to improve their security
capabilities are likely to incur increased cost that will be passed down to their customers
if there is a demand for security and suffer a decrease in their ability to deliver products
to customers in a reliable manner. Chapter Four detailed the results of this research
wherein purchasing managers employed in the food industry were surveyed to assess
their willingness to accept price increases, and delivery reliability decreases, from their
suppliers in return for security. Chapter Four also detailed the role of concern over
security incidents and international sourcing as they affect respondents’ willingness to
sacrifice price and delivery reliability for supplier security.

Results indicate that food purchasing managers are not willing to trade-off price
and delivery reliability for their suppliers to operate in a secure manner unless they
source internationally or are concerned about security incidents at their firm. When

sourcing internationally, respondents place more value on supplier security. Respondents

128

who are concerned about previous security incidents that have occurred at their firm also
place more value on supplier security. Results indicate that international sourcing and
respondent concern increase the value respondents place on security to the extent that
they are willing to pay one to two percent higher prices in return for increased supplier
security. However, international sourcing and respondent concern do not increase the
value respondents place on security to the extent that they are willing to trade-off delivery
reliability for security.

Two levels of supplier security competence were evaluated by respondents: basic
and advanced. Suppliers with a basic security competence were conceptualized to only
utilize limited measures to protect their product, processes, and facilities from internal
(e. g. disgruntled employee) and external (6. g. terrorist) food contamination threats. A
supplier with a basic security competence typically does not employ any, or at least very
few, of the following security measures: 1) lock their doors, 2) employ guards, 3) have
fences around their facilities, 4) perform background checks on employees, 5) secure
inventory, and 6) inspects product for possible contamination before it is shipped.
Suppliers with an advanced security competence were conceptualized to make every
effort to protect their product, processes, and facilities from internal (e. g. disgruntled
employee) and external (e.g. terrorist) food contamination threats. A supplier with an
advanced security competence typically does employ most, or the majority, of the
following actions: I) lock their doors, 2) has fences and gates protecting their facilities,
3) employ guards protecting entrances, 4) perform thorough background checks on all
employees and contractors, 5) constantly tracks inventory in a secure environment to

make sure it is not contaminated, and 6) thoroughly inspects product for possible

129

contamination before it is shipped. In sum, in this research supplier security competence
is conceptualized to encompass the extent to which suppliers: I) lock doors, 2) use fences
and gates to protect facilities, 3) employ guards to protect facility entrances, 4) perform
background checks on all employees and contractors, 5) track inventory, and 6) inspect
product for possible contamination before shipment occurs.

In order to increase the managerial relevance of these results, it is necessary to
engage in a final round of qualitative research wherein results were presented to industry
representatives. During these interviews, industry representatives were asked to interpret
results presented in Chapter Four. The information gathered during these interviews is
used to supplement the researcher’s views in the following discussion. Where information
from these interviews is included, the source (i.e. practitioner interviews) is cited. If
“practitioner interviews” are not cited in the following discussion, the views expressed

are solely those of the researcher.

MANAGERIAL IMPLICATIONS

With regard to price, findings indicate that 1) price plays a greater role in
determining supplier preference than security and 2) respondents are not willing to pay a
higher price for advanced security unless moderating conditions apply. Interestingly,
findings also indicate that security may lead to a competitive advantage in terms of
market share if suppliers match the price of their competitors who operate at a lower
security level. However, if a supplier is located internationally, or if a supplier’s

customers are concerned about security incidents that have occurred at their firm, results

130

indicate that customers are willing to pay one to two percent above industry average to
utilize a supplier with advanced security.

These findings uncover several managerial implications. First, an international
supplier appears to be more able to derive a competitive advantage from security than a
domestic supplier. F inns who source goods internationally face a certain agency
dilemma wherein they are less able to monitor suppliers and ensure product is produced
in a secure manner. Results indicate that international suppliers employing advanced
security measures can capitalize on this information asymmetry and charge one to two
percent higher prices for their goods.

Second, suppliers who have invested in security should segment their customer
base by those who are concerned about security incidents that have occurred at their firm
and those who are not. Previous research has shown that suppliers are not currently
passing on the cost of security to their customers. Results indicate that suppliers can
charge customers that are concerned about security incidents one to two percent higher
prices and gain a competitive advantage over suppliers that do not employ advanced
security measures. However, in order for these suppliers to have the ability to charge a
higher price for their security capabilities they must educate their customers on the
importance of security and the measures taken to secure their operations. When
customers are accepting bids from suppliers, these suppliers should highlight the
importance of security in their bid package and further highlight their efforts to ensure
physical security (e. g. use of locked doors, fences, gates, and guards to protect facility

entrances), perform background checks on all personnel who have access to facilities,

131

track inventory to ensure it does not deviate from a pre-specified route, and thoroughly
inspect product for possible contamination before shipment occurs.

With regard to delivery reliability, results indicate there is a positive relationship
between delivery reliability and security. In no case were respondents willing to sacrifice
delivery reliability for supplier security. International sourcing decreases the amount of
delivery reliability respondents were willing to accept for advanced security, but
concerned respondents were actually less willing to sacrifice delivery reliability than
unconcerned respondents. This indicates that domestic and international suppliers must
find ways to hold their delivery reliability constant while simultaneously increasing
security if they hope to gain a competitive advantage. This can be accomplished by
holding excess inventory, but with excess inventory comes increased cost and risk of
obsolescence, especially in the food industry where product often has a short shelf life.
These results may also indicate a causal factor on the part of respondents (e. g. use of
J IT).

It is recommended that suppliers employ tracking technologies, such as Radio
Frequency Identification (RF ID) and Global Positioning Systems (GPS), to substitute
information for inventory. While this is likely to increase short-term costs, it is also
likely to synergistically improve delivery reliability and security. Firms that employ
tracking technologies will have greater knowledge of inventory availability. This
knowledge allows firms to more accurately match supply and demand without holding
excess inventory. Further, tracking technologies allow firms to uncover if product has
deviated from a preset route during manufacturing or transport. This deviation may

indicate a security incident has occurred.

132

Practitioner interviews indicate that delivery reliability may have a halo effect
with respect to security. Purchasing managers feel that suppliers who focus on delivery
reliability are likely to employ advanced security measures (i.e. if a supplier does one
thing well, they probably do all things well). This may explain the positive relationship
between these two factors. Interviews further indicated that price does not have a halo
effect with respect to security. High prices do not one purchasing managers to believe
that suppliers employ advanced security and low prices certainly do not indicate
advanced supplier security. This implies that suppliers may be able to “sell” security to
their customers by providing high levels of delivery reliability and simultaneously avoid
the liability of promising customers that their operations are secure.

Findings with respect to price, delivery reliability, and concern over security
incidents merit further investigation from a customer segmentation standpoint. It appears
that concerned respondents are not only less price sensitive, but they value other supplier
attributes (e. g. quality, security, and delivery reliability) more than their unconcerned
counterparts. Previous research has shown that security is currently viewed as more of a
strategic concern (Voss et al., 2006) in as much as security is primarily perceived to be a
method of brand protection. Practitioner interviews revealed that there are two general
classes of customers in the food industry: those that are “deal prone” (i.e. focus on price)
and those that are “service mavens” (i.e. focus on service/more strategic issues).
Customers who are deal prone are likely to place less value on security as it impedes their
efforts to purchase product at the lowest cost. Customers who are service mavens appear

to be more likely to sacrifice price for outcomes such as security, quality, and delivery

133

reliability. Given the results presented in Chapter Four, it appears that concerned
respondents are part of the service maven category.

As evidence, factor importance scores indicate that security and delivery
reliability contribute more to supplier preference for concerned respondents than
unconcerned respondents. Interestingly, quality is more important to unconcerned
respondents, but concerned respondents preferred the highest level of product quality
while unconcerned respondents actually preferred the lowest level of product quality.
Managers are advised to segment their customers according to their emphasis on price or
service with the knowledge that those customers who place more value on service are
more likely to also place more value on security.

This can be somewhat challenging. Suppliers who primarily handle commodity
goods are likely to sell product to customers who focus more on price. These suppliers
are likely to derive little, if any, competitive advantage from increasing security.
Suppliers who add significant value to a product, or sell “name brand” goods, are more
likely to sell these products to customers who are more service oriented. In order to
protect their brand, these service oriented customers are more likely to be willing to
sacrifice price for security. However, if commodity suppliers sell goods to customers
who are conscious of the effect of product quality on manufacturing performance, these
suppliers may be able to charge a higher price for their goods if they can use security
processes to deliver a higher quality product. Suppliers of all product types are cautioned
that neither price nor service oriented customers are likely to be willing to sacrifice

delivery reliability.

134

Focusing on this interaction between product characteristics (commodity v. value
added) and customer focus, Figure 5.1 illustrates recommendations of when security
should be implemented, should not be implemented, and when customers should be

educated on the importance of security. Figure 5.1 is presented below:

FIGURE 5.1 SECURITY IMPLEMENTATION MATRIX

Customer Focus

 

 

Price Service
3 i Do not Implement Educate Customer
‘6 1 UNLESS Implement 1f customer
g o 1 _. perceives that the
fé E i D031“ 1mplernent product is crucial to
3 8 1 security unless ”SK rs a their process or if risk is
E 1 concern a concern
fl
6 1 Educate Customer Implement
3 U ' Implement if customer A customer that is
,g g l perceives brand equity focused on service and
E g | to be important or if risk purchases a value added
9" m is a concern product is likely to value
‘ supplier security
1

 

 

 

Figure 5.] contains four quadrants delineated by various combinations of
customer focus and product characteristics. When products are characterized as
commodity goods, and a supplier’s customer focuses on price, the supplier will gain little
competitive advantage (perhaps will even be placed at a competitive disadvantage) from
security implementation. However, this does not mean that a supplier should not
implement security. If the supplier wishes to mitigate the demand and supply risk posed

by terrorists, and this risk is perceived to be significant, the supplier should implement

135

security in order to avoid the potential costs of an incident. When a supplier sells
branded (i.e. brand name) products to service oriented customers, they should implement
security in order to protect this brand name and also are likely to have the ability to
derive a competitive advantage from security because their customers are more likely to
pay a higher price to a secure supplier.

If a supplier sells a commodity product to a service oriented customer, or a
branded product to a price oriented customer, then it would behoove the supplier to
educate their customers on value of security and implement security under certain
circumstances. First, for both quadrants, suppliers should implement security if they
have a strong desire to mitigate risk. Commodity suppliers selling product to service
oriented customers should educate their customers on the value of security and emphasize
that/assess if their commodity good is crucial to the customer’s process. Provided that
the good is crucial to the customer’s process, and the customer recognizes this situation,
the supplier may have the ability to charge a higher price for their commodity. In other
words, security may serve as a way to differentiate one supplier’s commodity good from
another supplier’s. Suppliers selling branded products to price oriented customers should
implement security in order to protect their own brand, but should also educate their
customers on the value of security and attempt to convince them that security will help
protect their own assets. In this situation, the supplier has more incentive to implement
security, as a way to protect brand equity, regardless of whether a competitive advantage

can be garnered from security implementation.

136

PUBLIC POLICY IMPLICATIONS

The results of this research also have public policy implications. Results indicate
that international suppliers are able to charge prices one to two percent above industry
average if advanced security capabilities are employed, but results also indicate that
respondents strongly prefer domestic suppliers (see Table 4.5) and were not willing to
pay higher prices for their domestic suppliers to operate in a secure manner. Results
indicate that concerned respondents were willing to pay one to two percent higher prices
for their suppliers to operate in a secure manner, but roughly one-half of the sample is
composed of unconcerned respondents. Respondents were not willing to sacrifice
delivery reliability for security regardless of any mitigating circumstances.

Therefore, domestic suppliers will only be able to charge a higher price for
security if their customers are concerned about security incidents that have occurred at
their firm. However, even concerned respondents will not accept lower delivery
reliability for higher security. If service levels suffer as a result of security
implementation, concerned respondents are likely to drop a supplier with higher security
in favor of one with higher delivery reliability. Further, it has not been established if a
one to two percent price increase will cover the cost of security. If a one to two percent
price increase does not cover the cost of security, suppliers could only be incented to
increase security if they are able to gain enough market share (i.e. volume) by increasing
security to overcome its cost in the long-term. In the absence of this market share
increase, suppliers who increase security are placed at a competitive disadvantage in

terms of cost.

137

Given this information, and the current state of industry security implementation
(i.e. most firms have only implemented physical security measures that would have little
positive effect on delivery reliability) it appears that in order to achieve a critical mass of
supply chain security in the domestic food industry, the US. Government is going to
have to regulate its implementation. For widespread, voluntary security implementation
to occur without regulation: 1) a one to two percent price increase would have to cover
security expenses, 2) suppliers will have to find a way to hold delivery reliability constant
while simultaneously increasing security, 3) the number of firms concerned about
security incidents that have occurred at their firm will likely have to increase, and/or 4) a
major security event will have to occur in the food industry, which would increase the
number of firms concerned about security. In as much as 1) it is unknown whether a one
to two percent price increase will cover security expenses, 2) security has been shown to
negatively affect delivery reliability, 3) only one half of the sample was concerned about
security, and 4) a major security incident has not occurred in the food industry, in the
short-term it is likely that regulation will be necessary to ensure the food industry

implements security measures.

THEORETICAL IMPLICATIONS

This research represents one of the few empirical works to examine supply chain
security. Further, the results of this research are the first to examine the value managers
place on security. Perhaps the primary theoretical contribution of this work is that it lays
the foundation for future research. Future research opportunities are discussed in the next

section.

138

This work delineated the importance of security in relation to other supplier
selection criteria. Examination of factor importance scores presented in Table 4.5 reveals
that security plays less of a role in determining supplier preference than product quality,
delivery reliability, and price. However, results also indicate that security is more
important to firms under certain circumstances (e. g. international sourcing and concern
over security incidents). While the study of supply chain security is still in its infancy, it
is likely to grow more important in the future given the likelihood that security will have
to be regulated. Future supplier selection criteria research should include security as a
control variable, if not a primary factor of investigation.

This work uncovered that service oriented firms are likely to place more value on
security and to consider it part of their suppliers’ service offering. Future investigations
into the value of suppliers’ product quality and delivery reliability should control for

supplier security competence.

RECOMMENDATIONS

The results of this research have important change management implications.
Perhaps one of the overriding questions facing food industry executives is whether or not
they should implement security measures. Managers faced with this dilemma should first
research security best practices in order to uncover the steps needed to secure operations.
Second, managers should determine the extent to which they wish to implement these
best practices, assess the cost of doing so, and the effect these security measures will
have on delivery reliability, quality, and price. Third, managers should either formally or

informally “survey” their customers. These surveys should question customers’ quality,

139

delivery reliability, and price preferences as well as the value they place on security and
the extent to which they are concerned about security incidents that have occurred at their
firm. International suppliers should specifically survey customers on the extent to which
they are willing to sacrifice price and delivery reliability for security or simply expect to
pay a higher price and receive lower delivery reliability when they source internationally.
Given that respondents strongly prefer domestic suppliers to international suppliers (see
Table 4.5), it is possible they simply expect to pay a higher price if they source their
goods internationally. For example, if a commodity is available domestically and
internationally, respondents would source this product domestically all else being equal.
However, if the commodity were not available domestically, respondents would have no
choice but to source internationally. This situation is quite common in the food industry
given the year round demand for products with seasonal availability of inputs.
International suppliers of seasonal products are likely aware their customers have few
supplier choices, enabling them to charge a higher price.

Armed with this knowledge, suppliers will have a better grasp of customer
preferences. Customers who are service oriented (i.e. they place a high priority on
security, quality, and delivery reliability over price), or are concerned about security
incidents at their firm, are more likely to accept higher prices if they have been educated
as to suppliers’ advanced security capabilities. Suppliers should then assess whether this
price increase allows them to recoup the long-term costs of security. If the price increase
covers security expense, and the supplier can use tracking technology (or other means) to
hold delivery reliability constant, then it is recommended that this supplier undertake a

security initiative.

140

If price increases do not cover security expenses, or survey results reveal that
customers are price oriented (i.e. value low price above other supplier attributes), then
suppliers must weigh the benefits of security as a long term-cost avoidance mechanism
versus the short-term cost of security implementation. The cost of security measures in
the short-term may pale in comparison to the long-term costs of litigation, loss of
customer goodwill, sales, brand equity, and subsequent sale of equity investments that are
likely to occur as a result of a major security/contamination incident. However, the
supplier must also recognize that the likelihood of a major security incident is quite low.
It is recommended that suppliers whose security expenses are not covered by price
increases, or who service price oriented customers, secure their operations to the extent
that they are not placed at a severe competitive disadvantage.

Another, perhaps more likely, scenario is that some customers will value security
to the extent they are willing to pay more for it and others will not. In this case, suppliers
might consider the possibility of implementing security at select facilities, and servicing
customers who value security out of these facilities, while leaving other facilities
relatively unsecured. In this case, facilities with high levels of security should strive to
be “world class” in every respect. These world class facilities should offer the highest
levels of product quality, delivery reliability, and security and educate customers that
product from this facility comes at a price premium and why this premium is necessary.
Other facilities should focus solely on efficiency - producing product at the lowest total
cost. This would allow the firm to offer “menu pricing” to customers. Under a menu
pricing strategy, customers have the option of receiving product from the world class

facility that delivers high quality product, on time (perhaps utilizing the previously

141

mentioned tracking technologies to control delivery reliability), and in a secure manner.
Conversely, customers have the option of purchasing low cost product from the
unsecured facility. This strategy may also serve as a stop-gap measure for suppliers who
wish to test the security waters without fully committing to the expense of company wide
implementation.

However, this stop-gap measure is likely to be just that. Suppliers who secure
only certain facilities are more vulnerable at unsecured locations. While a security
incident may be avoided at a secure location, a security incident at an unsecured location
will have a negative halo effect on the firm as a whole. This is likely to raise doubts in
customers who were serviced fiom the world class facility and cause them to source
elsewhere, thus negating any positive benefits that may have been gained from the menu
pricing strategy. This strategy is only recommended in situations where a firm cannot
afford to simultaneously implement security measures at all facilities.

It should also be noted that suppliers increase their liability when they secure their
operations, educate their customers regarding security capabilities, and promise that
secure product will be delivered. If a security incident occurs, despite best efforts, this
security breach is likely to bring about greater liability because the supplier failed to live
up to its promises. If this is a significant concern, suppliers may wish to forego promises
of security in lieu of simply delivering high levels of delivery reliability, which
practitioner interviews revealed may have a halo effect on security.

To the extent that it does not compromise business viability, firms are encouraged
to 1) implement security measures, 2) sacrifice reasonable amounts of price and delivery

reliability from their suppliers in return for security, 3) encourage customers to

142

implement security, and 4) work with industry groups to form a critical mass of supply
chain security while simultaneously collaborating with supply chain partners on ways to
gain synergies from security measures such that price and delivery reliability are not
adversely affected in the long-term. These actions will be difficult but, in their absence,
the food industry is likely to be faced with heavy handed, one size fits all security

regulation leading to long-term, negative consequences for the industry as a whole.

LIMITATIONS

The findings of this research contribute significantly to the current body of supply
chain security knowledge, but certain limitations do exist. First, as is common with most
research, results may not be generalizable to other industries. The food industry is likely
to place more emphasis on security than other industries due to the vulnerability of the
food supply chain to terrorist intrusion as well as the serious repercussions of distributing
contaminated food to consumers. While only half of the respondents in this study
indicated they were concerned about security incidents at their firm, it would be logical to
assume that less than half of respondents from other industries would be concerned about
security. Therefore, significant industry differences could exist.

Further, CBC results do not allow one to differentiate between expectations and
actual preferences. Results indicate that respondents are willing to pay one to two
percent higher price to international suppliers employing advanced security measures.
However, these results may simply reflect respondent expectations. Respondents may

legitimately place more value on security when they source internationally, but their

143

willingness to pay a higher price for security (as indicated in the managerial implications
above) may simply be an artifact of their expectations. Therefore, security may simply
be an order qualifier for international suppliers. While the results indicate that
respondents are willing to pay more for security from these suppliers, this result may
simply be an indication that they expect to pay more to their international suppliers.

Another limitation present in this work is the inability to reliably determine if
significant differences exist between market preference percentages due to the small
number of observations. Thus, the researcher is left to rely on “directional” differences
(i.e. the market preference of one factor level is greater than the market preference of
another). For this reason, significant differences between market preference percentages
were not thoroughly discussed in this research but undoubtedly the results would be more
robust if one were able to reliably determine if significant differences exist. Due to the
use of multiple evaluation methods, this is not perceived to severely devalue the research
results, but perhaps future research should consider use of “bootstrapping” to simulate a
larger nrunber of observations and test for significant differences.

- Decisions made while designing this research also pose certain limitations. The
definitions of basic and advanced security presented to respondents are rather simplistic
in as much as they are firm centric. This research did not seek to expand security beyond
that employed by the supplier. A true security program encompasses more than a single
firm and a supplier who truly employs advanced security measures would do more than
the firm centric measures provided to respondents. Further, there is a significant
“security gap” between the conceptualization of basic security and the conceptualization

of advanced security. While this was necessary to illustrate a difference between the two,

144

one could posit that basic security is so basic as to be unrealistic and advanced security

may be so advanced as to be unrealistic.
FUTURE RESEARCH

These results present opportunities for future research. First, concluding
interviews with managers revealed that security adoption may follow an S-curve similar
to new product adoption wherein there are a small number of early adopters, a larger
number of early followers, and a small number of firms who are resistant to change with
respect to security adoption. It is likely that concerned respondents comprise the early
adopter group and possibly part of the early follower group, while unconcerned
respondents are likely to compose the bulk of the early follower group and the resistant to
change group. Future research should establish if this relationship holds with respect to
security adoption. Future research should also seek to uncover the characteristics of these
groups. It is likely that larger firms, or firms who produce “name brand” goods, are more
willing to sacrifice for security as they have more of a stake in brand protection. Smaller
firms who have not invested in creating mass brand awareness, or firms who primarily
trade in commodity goods, are likely to have less concern over security.

Second, future research should seek to determine the optimal service offering for
each of these groups. The possibility exists that early adopters have different supplier
preferences than the early follower and resistant to change groups. Future research
should seek to uncover the optimal levels of quality, price, delivery reliability, and

security preferred by each of these groups.

145

Third, future research should seek to validate the relationship of security to
quality and delivery reliability. It appears respondents who are concerned about security
incidents may be more service oriented, and demand for supplier security may be
correlated with demand for other, more strategic, supplier offerings (e. g. quality and
delivery reliability). Future research should seek to provide empirical evidence
supporting this relationship between security and these more strategic factors.

Fourth, future research should seek to determine if a one to two percent price
increase covers the cost of security increases. This research is likely to be very industry
and product specific. For example, food is a relatively cheap commodity and a one to
two percent increase in price may not cover the cost of security implementation. If a one
to two percent price increase does not cover the cost of security, does the increase in
market share derived from increasing security pay for security increases? However,
researchers investigating the relationship of price to security should seek to go beyond
price as defined by initial monetary outlay and explore the total cost of ownership of
utilizing secure suppliers. It is possible that, while utilizing secure suppliers may cost
more initially, this cost may be offset over time by synergies between security and other
operational criteria. In other words, the cost of security may be offset by other factors
with the possibility of recouping this cost over time.

Fifth, practitioner interviews indicated that firms may not actually be willing to
pay a higher price for security when they source internationally. Respondents may
simply expect to pay more for goods when they source internationally because these

goods may not be available domestically. Future research should seek to determine if

146

willingness to pay more for internationally sourced goods is a function of increased
security or simply a function of expectations.

Sixth, practitioner interviews revealed that the high amount of vertical integration
in the food industry may play a role in the willingness of firms to sacrifice for supplier
security. If a firm sources goods internally, even if the sourcing location is overseas, this
firm may perceive that they have a lower risk of security incident because their suppliers
are under their control, thus lowering the purchasing agent’s uncertainty with respect to
the supplier’s level of security. Future research should explore the role of vertical
integration as it determines demand for security.

Finally, future research may seek to extend the present research and determine if
consumers, as opposed to firms, are willing to pay a higher price for security. A
correlation may be drawn between the value of certified organic food products and
certified secure food products. Manufacturers and retailers have found that some
consumers value organically produced food to the extent that they are willing to pay a
price premium for these goods. If retailers are able to certify that food product is secure,

consumers may be willing to pay a price premium for that product.

147

APPENDICES

148

APPENDIX A

COMPLETE SCENARIO LIST

Scenario List

.108

5:83 5:33

>oc2quoo btzoom

5.525 52.8

word 8391

E80 Heaven.

Scenano

10
11
12
13
14
15
16
17
18
19
20
21

22
23
24

25
26
27

28
29
30
31

32

33
34
35

149

36
37

38

39
40
41

42

43

44

45

46

47

48
49

50
51

52
53

2

2

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

150

86
87

88

89

90

91

92

93
94

95

96

97

98
99
100

101

102

103

104

105

106

107

108

151

APPENDIX B

SURVEY INSTRUMENT

Page 1

Page 1 asks the respondent to enter the password provided to them in order to access the
survey.

Page 2

Instructions

Thank you for your willingness to participate in this survey. To receive your executive
summary of the results of this research, please provide your e-mail address at the end of
the survey. Your answers will not be tracked by your e-mail address nor will your e-mail
address be shared with any outside parties.

This survey asks you to trade-off characteristics of a hypothetical food manufacturer that
supplies your firm with an important food product. A food manufacturer is defined as a
firm whose primary value added activity is to change the form of a food product. This
includes grain millers, protein processors, or any other supplier that receives food product
in one form, changes the physical form of the product in some way, and then ships the
product to your firm.

Supplier characteristics are defined in terms of five factors: mduct price, delivery
reliability, security competence, location, and mduct qualitv. Please only focus on these
five factors. You will be asked to compare four sets of supplier characteristics at a time
and then choose which one of the four you prefer MOST.

Brief definitions are presented in Section 1. In Section 2 you will be asked to choose the
most attractive set of supplier characteristics. If you need aprrore detailed defintion of a
supplier chaflteristic. click on the button containing the name of the characteristic and a
separate screen will appear with the definition. Following Section 2 are a few
demographic questions.

Each page of questions/evaluations has a "Next" button at the bottom of the page. Please
answer each question carefully before you hit the "Next" button to go to the next page.

This survey should take 15 minutes to complete. If you are interrupted during the survey,
log off the website and log back on later with your password. The purpose of using
passwords is to keep others out and to allow you to return. Your individual responses are
completely confidential and are not tracked by your password nor will they be shared

152

with anyone. I am the only person who will have access to the data that is gathered
through this research. Only the aggregate value all respondents place on the supplier
characteristics will be published.

Thank you for your time!

Doug Voss
Logistics Ph.D. Student
Michigan State University

(Click the Next button to continue)

Page 3

Section 1
Concept Definitions

Instructions:

Please read the following definitions and keep them in mind when you proceed to Section
2.

Product Price: Product price is defined as the monetary amount a supplier charges your
firm for their product. There are 3 levels of product price: Equal to or Below Industry
Average, Above Industry Averggg (1-2% above industry averge), and Significantly
Above Industry Avegrgp (3% or more above industry average).

Delivegy Reliability: Delivery reliability is defined as the percentage of time that a
supplier delivers product at the promised delivery time. There are 3 levels of delivery
reliability: Egufial to or Above Industry Average, Below Industry Average (1-2% below
industry average), and Sigpificantly Below Industry Averagg (3% or more below industry
average).

Product Quality:

Product quality is defined as how well a supplier's product conforms to your firrn's pre-
set specifications. There are 3 levels of product quality: Equal to or Below to Industry

Average, Moderately Higher than Indusfl Average, and Sigpificantly Higher than
Industry Avemgg.

Location of Supplier:
Supplier location is defined as the location where the supplier manufactures the food

153

product that is then shipped to your firm via the most appropriate mode of transportation.
There are 2 supplier locations: United States (located in Iowa and employs only US.
residents), and Middle East (located in Saudi Arabia and employs only Saudi residents).

Supplier Security Competence:

Security is defined as the application of policies, procedures, and technology to protect
food supply chain assets from theft, damage, or terrorism, and to prevent the introduction
of unauthorized contraband, people, or weapons of mass destruction into the food supply
chain.

There are 2 levels of supplier security: basic and advanced:

Basic Supplier Security Competence:

Suppliers with a basic security competence only utilize limited measures to protect their
product, processes, and facilities from internal (e. g. disgruntled employee) and external
(e. g. terrorist) food contamination threats.

Advanced Supplier Security Competence:

Suppliers with an advanced security competence make every effort to protect their
product, processes, and facilities from internal (6. g. disgruntled employee) and external
(e. g. terrorist) food contamination threats.

This concludes Section 1. Please proceed to Section 2

(Click the Next button to continue)

Page 4

Section 2
Choosing a Supplier

This section will present you with fifteen different scenarios. Each scenario contains four
sets of characteristics describing a hypothetical food manufacturer that supplies product
to your firm. Please carefully look over each of the four sets of characteristics in each
scenario then indicate which one of the four you would most prefer by clicking the circle
below the set of characteristics. If you need clarification on the defintion of a term, (e. g.
Security, Price, etc...) click on the button containing the name of that term.

Please begin with the scenario below:

154

Pages 5 — 20

Pages five through twenty present the conjoint scenarios in which respondents are asked
to choose between fifteen hypothetical suppliers. Due to the fact that the CBC software
randomly mixes the factor levels presented for each supplier in each scenario for each
respondent no examples are provided.

Pop-up windows
Each factor label in the above scenario is contained within a java script “button”. When
respondents click this button a pop-up window appears displaying the following k

factor/factor level definitions.

Supplier Securig Competence: Consider two different levels of security that a supplier
could employ.

 

Security is defined as the application of policies, procedures, and technology to protect a.
food supply chain assets from thefi, damage, or terrorism, and to prevent the introduction

of unauthorized contraband, people, or weapons of mass destruction into the food supply

chain.

The two levels of security, basic and advanced, are provided subsequently:

Basic Supplier SecurityCompetence:

Suppliers with a basic security competence only utilize limited measures to protect their
product, processes, and facilities from internal (e. g. disgruntled employee) and external
(e. g. terrorist) food contamination threats. A supplier with a basic security competence
typically does not employ any, or at least very few, of the following security measures:
lock their doors, employ guards, have fences around their facilities, perform background
checks on employees, secure inventory, and inspects product for possible contamination
before it is shipped to your firm.

Advanced Supplier Security Competence:
Suppliers with an advanced security competence make every effort to protect their

product, processes, and facilities from internal (e. g. disgruntled employee) and external
(e. g. terrorist) food contamination threats. A supplier with an advanced security
competence typically m employ most, or the majority, of the following actions: lock
their doors, has fences and gates protecting their facilities, employ guards protecting
entrances, perform thorough background checks on all employees and contractors,
constantly tracks inventory in a secure environment to make sure it is not contaminated,
and thoroughly inspects product for possible contamination before it is shipped to your
firm.

155

Product Price:

Product price is defined as the monetary amount a supplier charges your firm for their
product. These price levels are presented below:

Equal to or Below Industry Average - The supplier's price is equal to or below the

industry average.

Above Industry Average - The supplier's price is moderately higher than the industry
average (1-2% above industry average).

Significantly Above than Industry Averag - The supplier's price is significantly higher
than the industry average (3% or more above industry average).

Deliveg Reliabilig:

Delivery reliability is defined as the percentage of time that a supplier delivers product at
the promised delivery time. These delivery reliability levels are presented below.

Equal to or Above Industry Averggp: The supplier's delivery reliability is equal to or
above the industry average.

Below Industry Average: The supplier's delivery reliability is moderately worse than the
industry average (1 -2% below industry average).

Significpntly Below than Industry Averpgp: The supplier's delivery reliability is
significantly worse than the industry average (3% or more below industry average).

Location of Supplier:

Supplier location is defined as the location where the supplier manufactures the food
product that is then shipped to you via the most appropriate mode of transportation. The
two supplier locations are given below:

United States: This supplier is located in Iowa and employs only US. residents.

Middle East: This supplier is located in Saudi Arabia and employs only Saudi residents

156

Product Quality:

Product quality is defined as how well a supplier's product conforms to your firm's pre-
set specifications. Consider there to be three different levels of product quality a supplier
may attain:

Equal to or Below Industry Averggg - The supplier's product quality is equal to or below
the industry average.

Moderately Higher than Industry Average - The supplier's product quality is moderately
better than the industry average.

Sighificantly Higher than Indupm Averpgp - The supplier's product quality is
significantly higher than the industry average.

Page 21

This concludes Section 2.

Section 3 begins on the next page and asks you a few general questions about you
and your firm.

Section 3 is the final section of the survey and should take you no more than 5
minutes.

Please be sure to answer each question carefully before hitting the "Next" button.

Demographic questions

Beginning on Page 22, and concluding with the end of the survey, respondents were
asked to complete the series of demographic questions presented below.

1. What is your title?

t“ President (owner)
Vice-President
Director

Manager

 

533')

Other (please specify) i .. . 1 . . .

157

 

2. Please circle the category that best describes your current position (circle only one):
(‘ Purchasing
r‘ Supply Chain/Logistics
r Quality Assurance
r Security
("

 

Other (please specify) i

3. How long have you worked in this industry (in years)?
0-1 years

2-4 years

5-9 years

10-14 years

15-19 years

31331“)

20 or more years

4. How long have you worked in your current position (in years)?
0-1 years

2-4 years

5-9 years

10-14 years

15-19 years

”13337111

20 or more years

5. How long have you worked for your current employer (in years)?
0-1 years

2-4 years

5-9 years

10-14 years

15-19 years

33333”)

20 or more years

158

6. How many employees are in your division?

F Less than 50

r 50 - 250

F 250 - 1,000

r” 1,000 - 5,000

r More than 5,000

7. How many total employees work for your company in the US.

F Less than 50

F 50 - 250

r 250 - 1,000

r 1,000 - 5,000 .
C More than 5,000

8. Which best describes your firm's market area (check only one)?

C Local

r“ Regional
(‘ National
F Global

9. Which best describes your firm's supply scope (check only one)?

C Local

1" Regional
(‘ National
r“ Global

10. Which best describes your firm's scope of activities (check only one)?
(" Manufacturing

Merchandiser/Retailer

Carrier

Warehouse/Distributor

Logistics Service Provider

33333

Farm

159

11. Which best describes your firm's profit orientation?
r For-Profit
F Not-For-Profit

12. How would you assess your firm's level of vulnerability to intentional
contamination/security incidents?

r Minimal
Moderate

r“
r“ Average

r Above Average
F

Significantly Above Average

13. How concerned are you about intentional contamination/security incidents that have
occurred at your firm?

r Very Unconcemed
Unconcemed
Neutral

Concerned

Very Concerned
Not Applicable

33333

14. How do you assess the vulnerability of your suppliers to intentional
contamination/security incidents?

r Limited Vulnerability

(" Somewhat Vulnerable

1" Moderately Vulnerable
F Significantly Vulnerable

160

15. What role do the intentional contamination/security incidents experienced by your
firm play in selecting a supplier?

(t

33333

Very Limited Role
Limited Role
Moderate Role
Major Role

Very Major Role
Not Applicable

16. How likely would you be to pay a higher price for a supplier's product if they were
more secure than other potential suppliers of the same product?

("

("
("
(a
F

Very Unlikely

Somewhat Unlikely
Neither Likely nor Unlikely
Somewhat Likely

Very Likely

17. How likely would you be to accept a lower level of delivery reliability from a
supplier if they were more secure than other potential suppliers of the same product?

(‘
t"

F
(A
F

Very Unlikely

Somewhat Unlikely
Neither Likely nor Unlikely
Somewhat Likely

Very Likely

Please provide your e-mail address if you wish to receive an executive summary of the
results of this research.

 

(Click the Next button to continue)

161

APPENDIX C

WILCOXON RANK SUM TEST RESULTS

Description of Wilcoxon Rank Sum Test (Lehmann 1975)

The Wilcoxon Rank Sum test is a nonparametric method of determining statistical
differences between two independent sets of observations, whose properties violate the
assumptions of the Student’s t-test, by determining significant differences between the
sum of their ranks, which approximates a normal distribution as N -) 00. The procedure
for calculating the Wilcoxon Rank Sum test statistics and determining their significance,
or lack thereof, is as follows:

1. Determine if one wishes to assess if the means of two groups are equal (two-tailed

test) or if it is hypothesized that one group is greater than another (one-tailed test).

Divide data into two groups that are to be compared for significant differences.

Order the data from highest value to lowest value.

Rank the data from 1.. .n.

Calculate sum and average of the ranks for each group.

Sum the sums of each group.

Average the averages of each group.

Ta = sum of Group A ranks where Na = number of observations in Group A

Tb- — sum of Group B ranks where Nb= number of observations in Group B

0. Calculate Ta max and Tb max where Ta max: (N a*Nb)+[Na(Na+1)]/2 and Tb max

- (Na*Nb)+[Nb(Nb+1)]/2

11. Calcuate Ua and Ub (test statistics) where Ua = Ta max - Ta and Ub = Tb max — Tb.

12. Look up critical values for the test in a Wilcoxon Rank Sum table such as is found in
Lehmann (1975).

13. If both Ua and Ub fall outside of the critical values, the difference is significant. If
either fail to fall outside of the critical values, the difference is not significant.

14. For the following tests, the critical values are as follows:

“95”.“.0‘999’!”

Critical values for two tailed test @ .05 (6x6; .025 on each tail) = 5;30
Critical values for two tailed test @ .10 (6x6; .05 on each tail) = 7;28
Critical values for one tailed test @ .05 (3x3) = 0,8

Critical values for one tailed test @ .10 (3x3) = 1;7

The critical values above illustrate that it is almost impossible to find a significant
difference at the .05 level for a one-tailed test. It is difficult to find a significant
difference for a one or two-tailed test given the number of observations available.
Therefore, caution should be taken in interpreting the following results.

162

TABLE C.l HYPOTHESIS 5 RESULTS: SECURITY BY PRICE
INTERACTION

 

 

 

 

 

 

 

 

 

 

 

 

Security Market Preference
Level Price Level Utilities %
Basic Equal to or below 0.083 27.06
Basic 1-2% higher (0.076) 6.82
Basic 3% or more higher (0.007) 2.03
Advanced Equal to or below (0.083) 37.1 1
Advanced 1-2% higher 0.076 22.50
Advanced 3% or more higher 0.007 4.48

 

TABLE C.2 TEST FOR SIGNIFICNT DIFFERENCE IN MARKET
PREFERENCE FOR BASIC AND ADVANCED SECURITY FOR ALL LEVELS
OF PRICE

Advanced Basic
A B
37.1 1% 27.06%
22.50% 6.82%
4.48% 2.03%

Ordered Data
37.1 1%
27.06%
22.50%

6.82%
4.48%
2.03%

A

Ta=9 Tb=12
TaandTbMaxforN=3= 15
Ua=6

Ub=3

 

163

 

TABLE C. 3 HYPOTHESIS 6 RESULTS: SECURITY BY DELIVERY

RELIABILITY INTERACTION

 

 

 

 

 

 

 

 

 

 

 

Security Market Preference
Level Delivery Reliability Level Utilities %
Basic Equal to or Above 0.054 30.28
Basic 1-2% Below (0.098) 5.11
Basic 3% or more Below 0.044 1.65
Advanced Equal to or Above (0.054) 43.36
Advanced 1-2% Below 0.098 16.90
Advanced 3% or more Below (0.044) 2.70

 

TABLE C.4 TEST FOR SIGNIFICANT DIFFERENCE IN MARKET
PREFERENCE FOR BASIC AND ADVANCED SECURITY FOR ALL LEVELS

OF DELIVERY RELIABILITY

Advanced Basic
A Gro B
43.36% 30.28%
16.90% 5.1 1%
2.70% 1.65%

Ordered Data Gro
43.36%
30.28%
16.90%
5.1 1%
2.70%
1.65%

Ta=9 Tb=12

TaandTbMax forN=3= 15

Ua=6

Ub=3 NotSi

 

164

 

 

TABLE C. 5

HYPOTHESIS 8 RESULTS: SECURITY BY DELIVERY
RELIABILITY BY INCIDENT CONCERN INTERACTION

 

 

 

 

 

 

 

 

 

 

 

Concerned Unconcemed
Respondents Respondents
Market Market
Security Delivery Reliability Preference Preference

Level Level Utilities % Utilities %
Basic Equal to or Above 0.098 28.31 0.023 32.1 1
Basic 1—2% Below (0.017) 5.00 (0.195) 4.90
Basic 3% or more Below (0.082) 0.65 0.172 3.38
Advanced Equal to or Above (0.098) 47.18 (0.023) 38.71
Advanced 1-2% Below 0.017 15.64 0.195 18.97
Advanced 3% or more Below 0.082 3.23 (0.172) 1.94

 

 

 

165

 

 

TABLE C.6 TEST FOR SIGNIFICANT DIFFERENCE IN MARKET
PREFERENCE FOR SECURITY AND DELIVERY RELIABILITY BETWEEN
CONCERNED AND UNCONCERNED RESPONDENTS

Concerned Unconcemed
Gro A B

28.31% 32.11%

5% 4.90%

0.65% 3.38%
47.18% 38.71%
15.64% 18.97%
3.23% 1.94%

Ordered Data
47.18%
38.71%
32.11%
28.31%
18.97%
15.64%

5%
4.90%
3.38%

OO\IO\£II«§UJNv-‘

\O

3.23%
1.94%
0.65%

NUWUO‘NNC‘NO‘C‘W

A

Ta=40 Tb=38

TaandTbMax forN=6= 57

Ua=17

Ub=19 NotSi

 

166

TABLE C.7 TEST FOR SIGNIFICANT DIFFERENCE BETWEEN MARKET
PREFERENCE OF SUPPLIERS WITH ADVANCED SECURITY V. MARKET
PREFERENCE OF SUPPLIERS WITH BASIC SECURITY FOR CONCERNED
RESPONDENTS

Concerned Unconcemed
A B
47. 18% 28.31%
15.64% 5%
3.23% 0.65%

Ordered Data
47.18%
28.31%
15.64%

5%
3.23%
0.65%

Ta=9 Tb=12

TaandTbMax forN=3= 15

Ua=6

Ub=3 NotSi

 

167

TABLE C.8 TEST FOR SIGNIFICANT DIFFERENCE BETWEEN MARKET
PREFERENCE OF SUPPLIRES WITH ADVANCED SECURITY V. MARKET
PREFERENCE OF SUPPLIERS WITH BASIC SECURITY FOR
UNCONCERNED RESPONDENTS

Concerned Unconcemed
Gro A B
38.71% 32.11%
18.97% 4.90%
1.94% 3.38%

Ordered Data Gro
38.71%
32.1 1%
18.97%
4.90%
3.38%
1.94%

Av

Ta=11 Tb=10
TaandTbMax forN=3= 15
Ua=4

Ub=5

 

168

TABLE C.9 HYPOTHESIS 9 RESULTS: SECURITY BY PRICE BY
INCIDENT CONCERN INTERACTION

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Concerned Unconcemed
Respondents Respondents
Market Market

Security Preference Preference

Level Price Level Utilities % Utilities %
Basic Equal to or below 0.063 23.29 0.091 31.36
Basic 1-2% higher (0.061) 7.52 (0.083) 6.54
Basic 3% or more higher (0.002) 2.38 (0.008) 1.79
Advanced Equal to or below (0.063) 37.42 (0.091) 36.70
Advanced 1-2% higher 0.061 24.13 0.083 20.04
Advanced 3% or more higher 0.002 5.27 0.008 3.57

 

169

 

TABLE C.10 TEST FOR SIGNIFICANT DIFFERENCE IN MARKET
PREFERENCE FOR SECURITY AND PRICE FOR CONCERNED AND

UNCONCERNED RESPONDENTS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Concerned Unconcemed
Group A Group B
23.29% 31.36%
7.52% 6.54%
2.38% 1.79%
37.42% 36.70%
24.13% 20.04%
5.27% 3.57%
Ordered Data Group Rank
37.42% a 1
36.70% b 2
31.36% b 3
24.13% a 4
23.29% a 5
20.04% b 6
7.52% a 7
6.54% b 8
5.27% a 9
3.57% b 10
2.38% a 11
1.79% b 12
Group A Ranks Group B Ranks
1 2
4 3
5 6
7 8
9 10
11 12 Total
Sum 37 41 78
Average 6.166667 6.833333 6.5
Ta = 37 Tb = 41
TaandTb Max forN =6 = 57
Ua = 20
Ub = 16 Not Significant (p > .10)

 

170

 

TABLE C.11 TEST FOR SIGNIFICANT DIFFERENCE BETWEEN MARKET
PREFERENCE OF SUPPLIERS WITH ADVANCED SECURITY V. MARKET
PREFERENCE OF SUPPLIERS WITH BASIC SECURITY FOR CONCERNED

RESPONDENTS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Concerned Unconcemed
Group A Group B
37.42% 23.29%
24.13% 7.52%
5.27% 2.38%
Ordered Data Group Rank
37.42% a 1
24.13% a 2
23.29% b 3
7.52% b 4
5.27% a 5
2.38% b 6
Group A Ranks Group B Ranks
1 3
2 4
5 6 Total
Sum 8 13 21
Average 2.666667 4.333333 3.5
Ta = 8 Tb = 13
Ta andTbMax forN=3 =15
Ua = 7
Ub = 2 Not Significant (p > .10)

 

171

 

TABLE C.12 TEST FOR SIGNIFICANT DIFFERENCE BETWEEN MARKET
PREFERENCE OF SUPPLIERS WITH ADVANCED SECURITY V. MARKET

PREFERENCE OF SUPPLIERS WITH BASIC SECURITY FOR
UNCONCERNED RESPONDENTS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Concerned Unconcemed
Group A Group B
36.70% 31 .36%
20.04% 6.54%
3.57% 1.79%
Ordered Data Group Rank
36.70% a 1
31 .36% b 2
20.04% a 3
6.54% b 4
3.57% a 5
1.79% b 6
Group A Ranks Group B Ranks
1 2
3 4
5 6 Total
Sum 9 12 21
Average 3 4 3.5
Ta = 9 Tb = 12
TaandTbMax forN=3 =15
Ua=6
Ub = 3 Not Significant (p > .10)

 

 

 

172

 

TABLE C. 13

HYPOTHESIS 11 RESULTS: SECURITY BY DELIVERY
RELIABILITY BY SOURCING LOCATION INTERACTION: MARKET

 

 

 

 

 

 

 

 

PREFERENCE
Domestic Sourcing International Sourcing
Security Delivery Reliability
Level Level Market Preference % Market Preference %

Basic Equal to or Above 30.28 20.52
Basic 1-2% Below 5.11 3.29
Basic 3% or more Below 1.65 1.13
Advanced Equal to or Above 43.36 52.33
Advanced 1-2% Below 16.90 19.55
Advanced 3% or more Below 2.70 3.19

 

 

 

 

 

173

 

TABLE C. 14 TEST FOR SIGNIFICANT DIFFERENCE IN MARKET
PREFERENCE FOR SECURITY AND DELIVERY RELIABILITY FOR
DOMESTIC AND INTERNATIONAL SOURCING

Domestic Intemational
A B
30.28% 20.52%
5.11% 3.29%
1.65% 1.13%
43.36% 52.33%
16.90% 19.55%
2.70% 3.19%

Ordered Data
52.33%
43.36%
30.28%
20.52%
19.55%
16.90%

5.11%
3.29%
3.19%
2.70%
1.65%
1.13%

\OOONGMJXMNH

p—Aj—a
HO

o‘wmo‘c‘mmc‘o‘mmo‘
DU
pa
.. E ..
NOOOUIA—‘m N

.o‘w
mxo

A

Ta=39 Tb=39
TaandTbMax forN=6= 57
Ua=18

Ub=18

 

174

TABLE C. 15 TEST FOR SIGNIFICANT DIFFERENCE BETWEEN MARKET
PREFERENCE OF SUPPLIERS WITH ADVANCED SECURITY WHEN
SOURCING DOMESTICALLY V. SOURCING INTERNATIONALLY

Domestic International
A Gro B
43.36% 52.33%
16.90% 19.55%
2.70% 3.19%

Ordered Data
52.33%
43.36%
19.55%
16.90%

3.19%
2.70%

Av

Ta=12 Tb=9
TaandTbMax forN=3= 15
Ua=3

Ub=6

 

175

TABLE C.16 TEST FOR SIGNIFICANT DIFFERENCE BETWEEN MARKET
PREFERENCE OF SUPPLIERS WITH BASIC SECURITY WHEN SOURCING
INTERNATIONALLY V. SUPPLIERS WITH ADVANCED SECURITY WHEN
SOURCING INTERNATIONALLY

International International
Gro A B
52.33% 20.52%
19.55% 3 .29%
3.19% 1 . 13%

Ordered Data Gro
52.33%
20.52%
19.55%
3.29%
3.19%
1.13%

Ta=9 Tb=12
TaandTbMax forN=3= 15
Ua=6

Ub=3

 

176

TABLE C. 17 HYPOTHESIS 12 RESULTS: SECURITY BY PRICE BY
SOURCING LOCATION INTERACTION

 

Domestic Sourcing

International Sourcing

 

Market Preference

Market Preference

 

 

 

 

 

 

 

 

 

 

Security Level Price Level % %

Basic Equal to or below 27.06 18.33
Basic 1-2% higher 6.82 4.39
Basic 3% or more higher 2.03 1.33
Advanced Equal to or below 37.11 44.74
Advanced 1-2% higher 22.50 25.94
Advanced 3% or more higher 4.48 5.28

 

177

 

TABLE C. 18 TEST FOR SIGNIFICANT DIFFERENCE BETWEEN MARKET
PREFERENCE FOR DOMESTIC AND INTERNATIONAL SOURCING AT ALL
LEVELS OF SECURITY AND PRICE

Domestic International
A B
27.06% 18.33%
6.82% 4.39%
2.03% 1.33%
37.1 1% 44.74%
22.50% 25.94%
4.48% 5.28%

Ordered Data Gro
44.74%
37.11%
27.06%
25.94%
22.50%
18.33%

6.82%
5.28%
4.48%
4.39%
2.03%
1.33%

\OOO\IO\UIJ>DJN-‘

O‘NU‘NO‘AIU‘NO‘NNO‘

A

Ta=37 Tb=41
TaandTbMaxforN=6= 57
Ua=20

Ub=16

 

178

TABLE C.19 TEST FOR SIGNIFICANT DIFFERENCE BETWEEN MARKET
PREFERENCE OF SUPPLIERS WITH ADVANCED SECURITY WHEN
SOURCIN G DOMESTICALLY V. SOURCIN G INTERNATIONALLY

Domestic Intemational
A B
37 . 1 1% 44.74%
22.50% 25.94%
4.48% 5.28%

Ordered Data
44.74%
37.11%
25.94%
22.50%

5.28%
4.48%

Sum
Av

Ta=12 Tb=9
TaandTbMax forN=3= 15
Ua=3

Ub=6

 

179

TABLE C. 20 TEST FOR SIGNIFICANT DIFFERENCE BETWEEN MARKET
PREFERENCE OF SUPPLIERS WITH BASIC SECURITY WHEN SOURCING
INTERNATIONALLY V. SUPPLIERS WITH ADVANCED SECURITY WHEN
SOURCING INTERNATIONALLY

 

 

 

 

 

 

 

 

 

 

 

 

 

International International
Group A Group B
18.33% 44.74%
4.39% 25.94%
1.33% 5.28%

Ordered Data Group Rank
44.74% b 1
25.94% b 2
18.33% a 3

5.28% b 4
4.39% a 5
1.33% a 6

 

 

Group A Ranks Group B Ranks

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3 l
5 2
6 4 Total
Sum 14 7 21
Average 4.67 2.33 3.5
Ta = 14 Tb = 7
Ta and Tb Max forN=3 =15
Ua = l
Ub = 8 Significant Difference (p < .10)

 

180

APPENDIX D
PRE-NOTIFICATION LETTER
Dear ,

My name is Doug Voss and I am currently working to complete my Ph.D. in logistics at
Michigan State University. In order to complete my degree I am required to collect data
and produce a dissertation from this data. You have been randomly selected to
participate in this study. Your participation in this research is purely voluntary. Your
responses, should you choose to participate, will be stored separately from your name.
This will prevent the matching of your responses to your name. Should you choose to
participate, you will be provided an executive summary of the results.

The purpose of my research is to determine how important supplier security is to
purchasing managers employed by food manufacturers. This research will consist of a
web-based survey which presents profiles of four hypothetical suppliers, as well as
placing you in a hypothetical situation, and asks you to pick one of the two suppliers
given their hypothetical profile and your hypothetical situation. From the combined
responses I receive from all of the respondents to this survey I will be able to determine
how much importance purchasing managers place on security, product quality, delivery
reliability, and product price. I will also discover if certain conditions make security
more valuable to purchasing managers.

Again, your participation in this research is purely voluntary. There are no adverse
consequences for choosing to not participating. Your responses, should you choose to
participate, will be stored separately from your name. This will prevent the matching of
your responses to your name. You will be provided an executive summary of the results
upon completion of this research. The expected date of completion is June, 2006.

You will receive a letter in the next seven days that contains a link to a web page. This
web page contains the survey. Please do not hesitate to contact me with any questions or
concerns you may have. Thank you fOr your time.

Best regards,

Doug Voss David J. Closs, Ph.D.
PhD. Candidate in logistics

The Eli Broad College of Business

Michigan State University

vossmic1@msu.edu

(517) 353-6381

181

APPENDIX E
NOTIFICATION LETTER

Dear ,

You have been sent a postcard requesting your participation in Doug Voss’ dissertation
research pertaining to food supply chain security. To protect respondent anonymity, I am
not tracking who has completed the survey. If you have completed it, I would like to
extend my sincere “Thanks”. If you have not completed the survey, YOUR HELP IS
GREATLY NEEDED, and will allow you to receive an executive summary of the
results.

As someone who has worked in the private sector, I fully realize the value of your
time and know you receive numerous surveys. I cannot stress enough that this
survey is only for academic use, nobody will profit from your participation, and you
are only asked to provide anonymous preferences about a hypothetical supplier.

If you have a spare 15 minutes to complete the survey, please direct your web browser to
www.purchasingsurvey.com with your password: PW

Thank you very much for your time! Your help means a great deal to me personally
and professionally. Please do not hesitate to e-mail or call if you have any questions.

Best regards,

Doug Voss David J. Closs, Ph.D.
PhD. Candidate in logistics

N370 Business College Complex

The Eli Broad College of Business

Michigan State University

voss@bus.msu.edu

(517) 432-5535 ext. 258

182

APPENDIX F
FINAL NOTIFICATION LETTER
Dear ,

You have been sent a postcard and letter requesting your participation in Doug Voss’
dissertation research pertaining to food supply chain security. To protect respondent
anonymity, I am not tracking who has completed the survey. If you have completed it, I
would like to extend my sincere “Thanks”. If you have not completed the survey, YOUR
HELP IS GREATLY NEEDED, and will allow you to receive an executive summary of
the results.

As someone who has worked in the private sector, I fully realize the value of your
time and know you receive numerous surveys. I cannot stress enough that this
survey is only for academic use, nobody will profit from your participation, and you
are only asked to provide anonymous preferences about a hypothetical supplier.

If you have a spare 15 minutes to complete the survey, please direct your web browser to
www.purchasingsurvey.com with your password:

Thank you very much for your time! Your help means a great deal to me personally
and professionally. Please do not hesitate to e-mail or call if you have any questions.

Warmest regards,

Doug Voss David J. Closs, Ph.D.
PhD. Candidate in logistics

N370 Business College Complex

The Eli Broad College of Business

Michigan State University

voss@bus.msu.edu

(517)432-5535 ext. 258

183

APPENDIX G

UCHRIS ASSURANCE OF EXCEMPT RESEARCH PROTOCOL

Principal Investigator Assurance of An Exempt Protocol

Name of Principal Investigator: Cheri Speier

Title of Project: "How Much Security is Enough? The Importance of Security in the Supplier
Selection Decision"

IRB #: X05-862

The University Committee on Research Involving Human Subjects has deemed this project as
exempt, in accord in federal regulations of projects exempt from IRB review. As an exempt
protocol, the IRB will not be further involved with the review or continued review of the projects,
as long as the project maintains the properties that make it exempt.

0 Since the IRB is no longer involved in the review and continued review of this project, it
is the Principal Investigator who assumes the responsibilities of the protection of human
subjects in this project and ensures that the project is performed with integrity and within
accepted ethical standards, particularly as outlined by the Belmont Report (see exempt
educational materials).

0 The Principal Investigator assumes responsibility for ensuring that the research subjects
be informed of the research through a documented or undocumented consent process, if
appropriate.

0 The Principal Investigator assumes the responsibility to maintain confidentiality of the
subjects and the data, and maintain the privacy of the subjects and the protection of the
data through appropriate means. If data is anonymous, the investigators will make no
attempt to identify any individuals.

0 The Principal Investigator assumes the responsibility that co-investigators and other
members of the research team adhere to the appropriate policies to protect human
subjects, maintain confidentiality and privacy, and adhere to accepted ethical standards.

0 If the Principal Investigator adds additional investigators to an exempt protocol, he/ she
may inform UCRIHS of the additions. This may be of particular importance to graduate
students if the Graduate School requires proof of IRB approval.

0 Any complaints from participants regarding the risk and benefits of the project must be
reported to the UCRIHS.

0 Since the Principal Investigator and co-investi gators are charged with human subject
protection and adhering to ethical principles in exempt research, it is appropriate that
investigators be trained in human subject principles. The Principal Investigator and all
members of the research team are required to complete MSU IRB educational
requirements or equivalent.

184

0 Any change in the protocol which may raise the project from exempt to an expedited or
full review category, must be presented to UCRIHS. If there is any question about a
change in protocol the Principal Investigator should consult the Chair of UCRIHS.
Failure to submit changes which raise the protocol out of the exempt category will be
considered non-compliance and will be subject to investigation and action by UCRIHS.

By signing below, the Principal Investigator assures that he/she will abide by the temis of this
assurance and the UCRII-IS exempt policy.

 

 

Signature of Principal Investigator Date

185

BIBLIOGRAPHY

186

BIBLIOGRAPHY

Aberdeen Group (2004), “How Supply Chain Leaders Protect Their Brands: A
Benchmark Report on Regulatory Compliance and Product Safety Mandates for
Food, Pharmaceuticals, Consumer Products, and Medical Devices,” September

Armstrong, J. Scott and Terry S. Overton (1977), “Estimating Nonresponse Bias in Mail
Surveys,” Journal of Marketing Research, 14 (August), 396-402.

Blalock, Cecelia (2003), “Tackling Transportation Security,” Advantage, October,
available at www.fmi.org/advantage.

Bledsoe, Gleyn E. and Barbara A. Rasco (2002), “Addressing the Risk of Bioterrorism in
Food Production,” Food Technology, 56(2), February, p. 43-47.

Bowersox, Donald J ., David J. Closs, M. Bixby Cooper (2002), “Supply Chain Logistics
Management,” New York, NY: McGraw Hill.

, , and Theodore P. Stank (1999), “21St Century Logistics: Making Supply
Chain Integration a Reality,” Oak Brook, IL: Council of Logistics Management.

 

Braglia, Marcello and Alberto Petroni (2000), “A Quality Assurance-Oriented
Methodology for Handling Tradeoffs in Supplier Selection,” International
Journal of Physical Distribution and Logistics Management, 30 (2), p. 96-1 1 l.

Bremrner, Paul (2001). New risks in international business. Retrieved on October 10,
2003, from: http://www.mmc.com/views/autumn 01 bremershtml.

Bruemmer, Barbara (2003), “Food Biosecurity,” Journal of the American Dietetic
Association, June, 103 (6), p. 687 —— 691.

Cattin, Phillipe and Dick R. Wittink (1982), “Commercial Use of Conjoint Analysis,”
Journal of Marketing, 46 (3), pg. 44-53.

Chalk, Peter (2001), “Terrorism, Infrastructure Protection, and the US. Food and

Agricultural Sector,” Rand Corporation, Senate Testimony given October 10,
2001.

Choi, Thomas Y. and Janet L. Hartley (1996), “An Exploration of Supplier Selection
Practices Across the Supply Chain,” Journal of Operations Management, 14 (?),
p. 333-343.

Churchill, Jr., Gilbert A. (1979), “A Paradigm for Developing Better Measures of
Marketing Constucts,” Journal of Marketing Research, 16 (February), p. 64-74.

187

-------- and Dawn Iacobucci (2002), Marketing Research: Methodological Foundations,
eighth edition, Harcourt College Publishers, Ft. Worth, pgs. 1006.

Closs, David J. and Edmund F. McGarrell (2004), “Enhancing Security Throughout the
Supply Chain,” IBM Center for the Business of Government, available at
www.businessofgovemmentorg, pgs. 52.

Cremin, Brian (2001), Extortion by product contamination: A recipe for disaster within
the food and drink industry, The American Behavioral Scientist, February, 44(6),
p. 1042-1052.

Cullen, David and Sean Kilcarr (2004), “Lockdown 2: Security and Borders,” F [ect
Owner, June, 99(6), p. 20-26.

Cupp, O. Shawn, David E. Walker II, and John Hillison (2004), Agro terrorism in the
US: Key security challenge for the 21St Century, Biosecurity and Bioterrorism,
2(2), p. 97-105.

Department of Transportation Bureau of Transportation Statistics (2005), “Pocket Guide
to Transportation,” available at http://www.bts.gov/publications/pocket
guide to transportation/2005/pdf/entire.pdf, accessed 1-23-05.

Dillman, Don A. (2000), Mail and Internet Surveys: The Tailored Design Method,
Second Edition, John Wiley and Sons, Inc., New York, pgs. 464.

Dobie, Kathryn, L. Milton Glisson, and James Grant (2000), “Terrorism and the Global
Supply Chain: Where Are Your Weak Links?,” Journal of Transportation
Management, 12(1), Spring, p. 57-66.

Dunn, Steven C., Robert F. Seaker, and Matthew A. Waller (1994), “Latent Variables in
Business Logistics Research: Scale Development and Validation,” Journal of
Business Logistics, 15 (2), p. 145-172.

Evans, Richard H (1982), “Product Involvement and Industrial Buying,” Journal of
Purchasing and Materials Management, 18 (Summer), p. 23 — 28.

EyeforTransport Global Research (2004), North American supply chain security —- An
analysis of eyefortransport’s recent survey, July 2004.

EyeforTransport (2005), “Cargo and Supply Chain Security Trends 2005,” available at
http://www.eyefortransport.com, accessed 9/21/05.

Fransella, Fay, Richard Bell, and Don Bannister (2004), A Manual for Repertory Grid
Technique, Second Edition, John Wiley and Sons, Ltd., West Sussex, pgs. 266.

188

Gips, Michael A. (2003), The first link in the food chain, Security Management, February
2003, Available at http://www.securitymanagement.com/library/001379.html,
accessed August 26‘“, 2004.

Goffin, Keith, Marek Sywejczewski, and Colin New (1997), “Managing Suppliers:
When Fewer Can Mean More,” International Journal of Physical Distribution
and Logistics Management, 27 (7), p. 422-436.

Green, Paul E. and V. Srinivasan (1978), “Conjoint Analysis in Consumer Research:
Issues and Outlook,” Journal of Consumer Research, 5 (2), p. 103-123.

Hannon, David (2004), Securing the supply chain: What you need to know today,
Purchasing, January 15, 133(1), p. 50-52.

Hair, Joseph F. J r., Rolph B. Anderson, Ronald L. Tatham, and William C. Black (1998),
Multivariate Data Analysis, fifth edition, Upper Saddle River, NJ: Prentice
Hall, pgs. 730.

Harl, Neil E. (2002), US. agriculture, food production is threatened by bioterrorism
attacks, Ag Lender April, 2002: 1.

Helferich, Omar Keith and Robert L. Cook (2003), “Securing the Supply Chain,” Council
of Logistics Management, Oak Brook, IL.

Hulme, George V. (2005), “Food Chain’s Fear Factor,” Information Week, May, p. 39—
42.

Johnson, M. Eric (2001), “Leaming from Toys: Lessons in Managing Supply Chain Risk
from the Toy Industry,” California Management Review, 43 (3), p. 106-124.

J orgensen, Barbara (2003), Remedies for an insecure supply chain, Electronic
Business, November, 29(14), p. 37-38.

Kahaner, Larry (2003), “Lock ‘em Up,” Fleet Owner, 98(2), p. 8.

Kent, John L. and Daniel J. Flint (1997), “Perspectives on the Evolution of Logistics
Thought,” Journal of Business Logistics, 18(2), p. 15-29.

Lehmann, Donald R. and John O’Shaughnessy (1974), “Difference in Attribute
Importance for Different Industrial Products,” Journal of Marketing, 38 (April),
p. 36-42.

-------- and -------- (1982), “Decision Criteria Used in Buying Different Categories of
Products,” Journal of Purchasing and Materials Management, 18(1), p. 9-15.

Lehmann, Erich L. (1975), Nonparametrics: Statistical Methods Based on Ranks,

189

McGraw-Hill, New York, p. 457.

Louviere, Jordan J. and George Woodworth (1983), “Design and Analysis of Simulated
Consumer Choice or Allocation Experiments: An Approach Based on Aggregate
Data,” Journal of Marketing Research, 20 (November), p. 350 — 367.

-------- (1988), Analyzing Decision Making: Metric Conjoint Analysis, Sage
Publications, London, pgs. 95.

Meriwether, Gordon and Rita Goodell (2004a), “A Food Security Primer: Risk to
Recovery Part 1: Overview,” available at http://www.idmagonline.com, accessed
1-3-05.

Michigan State University Food Supply Chain Security Research Team (in progress).
Min, Hokey (1994), “International Supplier Selection: A Multi-Attribute Utility

Approach,” International Journal of Physical Distribution and Logistics
Management, 24 (5), p. 24-33.

National Academy of Sciences (2004), Countering agricultural bioterrorism, available at
http://books.n_ap.edu/catalog/10505.htm1, accessed August 26"”, 2004.

Nganje, William, William W. Wilson, and James Nolan (2003), “Terrorism and the Grain
Handling System,” CAES Annual Meeting, July 26-30, 2003, Montreal, Canada.

Nunnally, J um (1978), Psychometric Theory, Second Edition, McGraw Hill, New York,
pgs. 701.

Onne, Bryan K. (2006), “Getting Started with Conjoint Analysis: Strategies for Product
Design and Pricing Research,” Madison, WI; Research Publishers, LLC.

Phillips, Lawrence T. (2001), “A Crisis of Security and Economics,” Regulation, 24(4),
p. 53-56.

Porter, Michael E. (1985), Competitive Advantage: Creating and Sustaining Superior
Performance Collier McMillan, New York, pgs. 557,

 

Radjou, Navi (2003), Securing the supply chain: Constructive paranoia, Network
Computing, September 25, 14(19), p. 2.

Rand Corporation (2003), “Agro terrorism: What is the Threat and What Can Be Done
About It?,” Rand Corporation Research Brief Series, available at www.rand.org.

 

Rice, James B. and Federico Caniato (2003), “Building a Secure and Resilient Supply
Network,” Supply Chain Management Review, 7(5), September-October, p. 22-
30.

190

-------- and Philip W. Spayad (2005), “Investing in Supply Chain Security: Collateral
Benefits,” IBM Center for the Business of Government, available at
www.businessofgovernmentorg, pgs. 28.

Rogers, David, David Lockman, Gregg Schwerdt, Brian O’Donnell, and Roger Huff
(2004), Supply chain security, Material Handling Management, February, 59(2),
p. 15-17.

Russell, Dawn M. & John P. Saldanha, (2003), “Five Tenets of Security-Aware Logistics
and Supply Chain Operation,” Transportation Journal. 42(4), p. 44-54.

Segal, Madhav N. (1982), “Reliability of Conjoint Analysis: Contrasting Data Collection
Procedures,” Journal of Marketing Research, 19 (February), p. 139-143.

Sheffi, Yossi (2001). “Supply chain management under the threat of international
terrorism,” International Journal of Logistics Management. 12(2), p. 1-11.

Spekman, Robert E. and Edward W. Davis (2004), “Risky Business: Expanding the
Discussion on Risk and the Extended Enterprise,” International Journal of
Physical Distribution and Logistics Management, 34 (5), p. 414 -— 433.

Strozniak, Peter (2004), “If the Fight Comes to Food,” Industry Week, October, 25 3(10),
p. 15.

Transportation Security Administration (2003), Report to Congress on transportation
security, March 31, 2003.

US. Customs and Border Protection (2004), “CSI in Brief ’, available at
http://www.cbp.gov/xp/cgov/border security/international activities/csi/csi in briefxml,
accessed 1-25-04.

US. Food and Drug Administration (2004), “Fact Sheet on FDA’s New Food
Bioterrorism Regulation: Establishment and Maintenance of Records,” December
6, available at http://www.cfsan.fda.gov/~dms/fsbtac23.html, accessed 1-2-05.

-------- (2003), Risk Assessment for food terrorism and other food safety concerns,
October 7, 2003. Available at http://www.cfsan.fda.gov/~dms/rabtact.html.
Accessed August 26th, 2004.

Vonderembse, Mark, Michael Tracey, Chong Leng Tan, and Edward J. Bardi (1995),
“Current Purchasing Practices and J IT: Some of the Effects on Inbound
Logistics,” International Journal of Physical Distribution and Logistics
Management, 25 (3), p. 33-48.

Warehousing Education and Research Council (2004), “The Cost of Security,”

191

Warehousing Education and Research Council Newsletter, September, p. 1-3.

Wheelis, Mark, Rocco Casagrande, and Lawrence V. Madden (2002), Biological attack
on agriculture: Low tech, high impact bioterrorism, Bioscience, July, 52(7), p.
569 — 576.

Wilson, Elizabeth J (1994), “The Relative Importance of Supplier Selection Criteria: A
Review and Update,” International Journal of Purchasing and Materials
Management, 30 (3; Summer), p. 35-41.

Wolfe, Michael. (2001), Fifih EU/U S Forum on Intermodal Freight Transport, Retrieved
on September 2, 2003, from:
http://ops.fhwa.dot.gov/freight/documents/frt sec ex sum4 4 02.doc.

-------- (2004), “The Dynamics of Supply Chain Security,” The Monitor, 10
(2), available at http://www.maritimeexpo.com, accessed 9/21/05.

World Health Organization (2002), “Terrorist Threats to Food: Guidance for
Establishing and Strengthening Prevention and Response Systems,” Geneva.

Zsidisin, George A. (2003), “A Grounded Definition of Supply Risk,” Journal of
Purchasing and Supply Management, 9, p. 217 — 224.

192

IIIIIIIIIIIIIIIIIIIIIIIIIIII

l1111111111111111111151115lel