bx ,

.w

a;
my?
. ,

.

ﬁ‘
u. :4: a.
, i
din».

Fl ‘

I

hﬁh.u...um

:-

I v
I. Irif. Pl.
$5.1

 

210:» l
1r. van? link...“

. iii. 1.1...

I .1. \ 1“!)

 

 

  

.}.4. 1‘4. 4 ...: vii...

 
 

». THESIS
I :9

am

This is to certify that the
dissertation entitled

TOWARD AN INTEGRATED CONCEPTUAL MODEL OF

RETAILER NEW PRODUCT EVALUATION AND NEW

 

I a: I PRODUCT success
4—.
CU
>- 4-0 >
a: w -;-_:-:'
at: g E: , presented by
m 9-: I
."" .C ." 5
-’ .2 -
2 Ying Huang

1

has been accepted towards fulfillment
of the requirements for the

Ph. D. degree in Advertising, Public Relations,
and Retailing

 

 

’7.

Major P'rofessor’sVSiﬁhature

43/ f4 9007

Date

MSU is an affirmative-action, equal-opportunity employer

cc---------n--------c-o-------o--o-v-a-v-p-I—I-u-o----v---I-.-o--v-.-o-o-u-a-a-u--I--o---I----o-a-3-u-u---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 I DATE DUE DATE DUE
Mm a '0 @013"?

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p:/CIRC/Date0ue.indd-p.1

 

TOWARD AN INT EGRATED. CONCEPTUAL MODEL OF RETAILER NEW
PRODUCT EVALUATION AND NEW PRODUCT SUCCESS

By

Ying Huang

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Advertising, Public Relations, & Retailing
Specialized in Retailing

2007

ABSTRACT

TOWARD AN INTEGRATED CONCEPTUAL MODEL OF RETAILER NEW
PRODUCT EVALUATION AND NEW PRODUCT SUCCESS

By

Ying Huang

The main objective of this study is to establish a link between two different
streams of literature devoted to new products in the consumer market: the ﬁrst is new
product (NP) literature, which focuses on manufacturers’ strategies and procedures in
launching new products; the second is retail buying (RB) literature, which focuses on
retailers’ buying behavior in relation to new products. To achieve this goal, we propose
an integrated model of retailer new product evaluation and new product success focusing
on two stages in the retail buying process: (1) the evaluation stage and (2) the post-
evaluation stage.

Using a sample collected from China, we tested the model. The results of this
study suggest that the retail buyer derives his/her evaluation of a new product from new
product advantage to consumers, new product advantage to retailers, and retailer trust of
the supplier. Further, this evaluation inﬂuences the extent of the retailer’s collaboration
with the supplier, and in turn, inﬂuences new product performance and retailer market
performance and retailer ﬁnancial performance. Due to the critical role retailers play in
i the potential market success of a new product, this study is an initial step in exploring the
importance of retailer new product buying in the new product development chain—an

area that connects RB literature with NP literature.

To Yingzi, Franklin, Jefferson, and William Wu, whose enduring love and support will
be cherished forever.

iii

ACKNOWLEDGEMENTS

I am deeply indebted to my wonderful doctoral committee for their support and
patience throughout my endeavor to complete my dissertation. First, I’d like to thank Dr.
Brenda Stemquist, my Academic Advisor and Chair of my doctoral committee; without
her encouragement and guidance, this dissertation would have been a mission impossible.
I particularly want to thank Dr. Roger Calantone, for his inspirational insights; I have
learned an immense amount from conversations with him. I’m grateful to Dr. Patricia
Huddleston, for being a wonderful mentor ever since the beginning of my doctoral
program. I also want to thank Dr. Kyoung-Nan Kwon for her helpful suggestions.

As always, love from my husband and my sons has been my impetus throughout
pursuing my doctoral degree. Completion of my degree would have been impossible
without my husband’s support and sacriﬁce. I am grateful for my parents, whose love and
support has helped me to go through the most difficult times of this journey.

I wish to thank my friends for their friendship and my colleagues for their
cheerleading. I also would like to extend my thanks to Dr. Richard Cole and Dr. Linda
Good for the financial support I received from the department.

Finally, I would like to thank the China Retail Research Center (CRRC) for

collaborating with data collection.

iv

TABLE OF CONTENTS

LIST OF TABLES ............................................................................... vii
LIST OF FIGURES .............................................................................. viii
CHAPTER 1. INTRODUCTION .................................................................. 1
Significance of the Study .................................................................. 6
Organization of the Study .................................................................. 7

CHAPTER 2. LITERATURE REVIEW AND THEORETICAL FRAMEWORK... . . . . ...8

Part One: Literature Review .............................................................. 8

A Review of Organizational Buying (OB) Behavior .......................... 8

A Review of New Product (NP) Literature .................................... 16

A Synthesized Conclusion of Literature Review .............................. 17

Part Two: Theoretical Framework ........................................................ 21

Theoretical Underpinnings ....................................................... 21

Part Three: Hypotheses Development .................................................... 29

New Product Advantage, its Antecedents, and its Effect .................... 29

Trust, its Antecedents, and its Effect ........................................... 33

New Product Evaluation and its Effect on Collaboration ................... 40

The Effect of Retailer Collaboration on New Product Performance ...... 41

New Product Performance and Retailer Performance ....................... 46

Control Variable .................................................................. 48

CHAPTER 3. METHODS ........................................................................ 49

Study Context .............................................................................. 49
Background Information of Retailing and New Product

Development in China .......... . .......................................................... 50

Survey Instrument .......................................................................... 52

Measures .................................................................................... 53

Data Collection ............................................................................. 55

Sample ....................................................................................... 57

CHAPTER 4. ANALYSES AND RESULTS ................................................... 67

Initial Data Analysis ....................................................................... 67

Measurement Validation .................................................................. 71

CFA of New Product Advantage ................................................ 71

CFA of Product Innovativeness and New Product Advantage ............. 74

CFA of Trust, Reputation, and Personal Relationships .................... 76

CFA of Cooperation, Communication, and Commitment ................... 77

CFA of New Product Performance and Retailer Performance .............. 79

The Path Model ......................................................................... 80

Results ....................................................................................... 8 1

CHAPTER 5. DISCUSSION ..................................................................... 88

Discussion of the Results .................................................................. 88
New Product Advantage ......................................................... 88

Trust ................................................................................. 90

New Product Evaluation ......................................................... 91
Collaboration ........................................................................ 92

New Product Performance and Retailer Performance ....................... 94
Theoretical Implications .................................................................. 94
Managerial Implications .................................................................. 96
Implications for Retailers ........................................................ 96
Implications for Suppliers ....................................................... 97
Limitations and Directions for Future Research ...................................... 99
Conclusion ................................................................................. 102
REFERENCES .................................................................................... 104

vi

Table 2.1

Table 3.1

Table 3.2

Table 3.3

Table 3.4

Table 3.5

Table 4.1

Table 4.2

Table 4.3

Table 4.4

Table 4.5

Table 4.6

Table 4.7

Table 4.8

LIST OF TABLES

A Comparison of this Dissertation to Previous Studies ........................... l9
Constructs and Measurement Items ................................................. 60
A Proﬁle of the Respondents’ Locations ........................................... 63
Characteristics of the Respondents .................................................. 64
Characteristics of Companies as Reported by the Respondents ................. 65
A Proﬁle of New Product Reported ................................................. 66
Construct Reliability and Item Information ........................................ 69
EFA Analysis of New Product Advantage ......................................... 73
CFA Results of Product Innovativeness and New Product Advantage... ......75
CFA Results of Trust, Reputation, and Personal Relationship .................. 77
CFA Results of Cooperation, Communication, and Commitment .............. 78
CFA Results of New Product Performance and Retailer Performance. . . . .....80
Covariance Matrix for the Path Model ............................................. 85
Results of the Path Model ............................................................ 87

vii

LIST OF FIGURES

Figure 2.1 An Integrated Model of New Product Evaluation and New Product

Performance ............................................................................. 20
Figure 2.2 A Reduced Form of Shem Model ................................................... 23
Figure 2.3 The S-Shaped Curve of New Product Diffusion ................................... 28
Figure 4.1 CFA Measurement Model of New Product Advantage .......................... 74
Figure 4.2 A Revised Model ...................................................................... 86

viii

CHAPTER 1
INTRODUCTION

New products are crucial to the survival and prosperity of modern firms.
According to the American Productivity & Quality Center (2003), new products launched
in the last three years account for more than one fourth of company sales. Because
successful new products help entice new customers and increase sales, profit, and market
share, generating a continuous stream of new products is important to every ﬁrm striving
to stay competitive and successful (Brown and Eisenhardt 1995; Cooper 2005; Hamel
and Prahalad 1991). Therefore, some companies, such as 3M, have been known for
having policies requiring that new products account for at least 30 percent of total sales
(see Kono and Lynn 2007).

Every year, a great number of new consumer products are introduced into the
market. In the US. alone, consumer product manufacturers introduce 15,000 to 20,000
new products including food and non-food items each year (Gallo 1997). The process of
introducing a new product into the market is referred to as the term “new product
launch”. The new product launch is risky and typically the most expensive stage in the
new product development process (NPD), thus representing a pivotal factor in
determining the success of new consumer products (Calantone, di Benedetto and Stank
2005). Unfortunately, many new products fail during the new product launch process
(Kono and Lynn 200). The failure rate is found to vary from 40 to 90 percent depending
on the ﬁrms studied and the deﬁnition of failure employed (Mahajan, Muller and Wind
2000). The reasons for failures are many and range from the inferior technology of a new

product to intense competition to insufficient marketing support (Calantone and di

Benedetto 1988). Among the different factors relevant to new product failure, very little
attention is paid to distribution (Pellegrini and Zanderighi 1991). The literature tends to
underestimate the role played by distribution in the success/failure of new consumer
products.

Distribution has been traditionally viewed as a variable within the marketing mix:
something that is totally controlled by manufacturers through investment in marketing
support and the effort of a sales force (Crawford 1977; Kotler 1997). In this manner,
retailers and other distributors have been treated simply as intermediaries, passing
products to consumers in the distribution channel. For many years manufacturers have
dictated what retailersI should stock and have decided pricing, promotion, and presence
of merchandise in stores (Corstjens and Corstjens 1995). However, this picture is
completely out of date. Since the mid-19903, the talk in both academic and practitioner
journals as well as in the popular press has been that channel power has increasingly
shifted from manufacturers to retailers and it now rests with retailers (Kadiyali,
Chintagunta and Vilcassim 2000), particularly as retail concentration becomes more
prevalent. Moreover, in the conventional marketing view, distribution is assumed to be
transparent and passive (Corstjens and Corstjens 1995). It is assumed that retailers pass
back to the manufacturers all information of the consumer market, and, as a result,
retailers’ initiatives and behaviors are mostly ignored. In reality, retailers have their own
marketing strategies, and they do not necessarily conform to those of manufacturers
(Pellegrini and Zanderighi 1991). Equipped with advanced information technologies,

retailers can now choose to collaborate through sharing information such as scanner data

 

1 In this dissertation, “retailer” is used interchangeably with “buyer” unless otherwise speciﬁed and
“manufacturer” is used interchangeably with “supplier”.

and loyalty programs with manufacturers or simply keep the information for their own
benefits—for example, retailers may choose to promote their own private labels. Today,
modem retailers select products, allocate shelf space, plan promotion activities, and price
products according to their own objectives rather than following the manufacturer’s
recommended selling price (MRSP). Having totally changed the traditional view of
distribution, such retailers are called “power retailers” (Lusch and Vargo 1998).
Accordingly, during the new product launch process, today’s retailer needs to be
considered as an independent and integral force in the marketing channel who possesses a
dual role: (1) as a gatekeeper determining new products’ access to the market in the
marketing channel (Rao and McLaughlin 1989), and (2) as a value-added co-producer
determining the success of new products in the market (Vargo and Lusch 2004).

The retailer’s ﬁrst role as a gatekeeper has caught the attention of researchers and
has been examined by previous studies. For a new product, gaining entry into stores is
important in its initial introduction into the market because this determines the level of
consumer accessibility. Every week, hundreds of new consumer products are presented to
retailer buyers. For example, Wal-Mart receives about 2000 inquires each week asking
for adding a new product (Sebenius and Knebel 2006). The Progressive Grocer study
reports that, on average, buyers turn down about 65 to 70 percent of all products
presented to them (Desiraju 2001; McLaughlin and Rao 1991). Two major reasons
explain this relatively high rejection rate. First, risks related to the high failure rate of
new products and uncertainties associated with the potential demand of new products
explain retailers’ reluctance to accept new products (Desai, 2000). Each year, new

products that fail in the market cost food retailers in the US. an average of $956,800 per

store (Desiraju 2001). Second, scarce shelf space limits retailers’ capacity for stocking
new products. From 1970 to 1990, new store size has increased by only 47 percent, while
the number of new product introduced has grown by 795 percent (McLaughlin and Rao
1991). There is simply not room for all the new products. Therefore, previous studies
have been centered on developing models to predict retailers’ decisions about whether to
accept or reject a new product. However, these models in general proved to be poor
predictors of the accept decision. As a result, researchers have called for a better model
for explaining the accept decision (e.g. Rao, McLaughlin and Hawkes 1995). Indeed,
examining a dichotomous accept/reject decision fails to disclose the richness of the
accept decision. Such an approach seems to ignore the likelihood of different evaluations
given to different new products. For example, one new product might be evaluated as
“average”, and thus may be only marginally accepted, whereas another new product
might be evaluated as “superior”, and thus is accepted without reservation. Consequently,
retailers may exert different behavior (e. g. providing a large amount of promotional
support vs. providing no promotional support) after stocking new products based on their
evaluations of new products. Therefore, the extant approach to new product buying seems
to have a limited understanding of the retailer’s role of a gatekeeper in new product
introduction. To solve this problem, the ﬁrst objective of this dissertation is to propose a
realistic model of new product evaluation rather than the dichotomous accept or reject
model. As such, factors at the construct level instead of decision criteria at the item level
that inﬂuence new product evaluation will be identiﬁed. For example, one factor
identiﬁed in our model is new product advantage, a construct representing an overall

assessment of new product attributes.

The second role of the retailer as a co-producer in determining the success of a
new product launch has been virtually overlooked in the literature. Although previous
studies allude to the importance of retailers in determining new product success, they
usually “treat the variable in an ad hoc way or do not consider it at all” (Rao and
McLaughlin 1989, p. 81). Little research has been dedicated to investigating how retailers
inﬂuence new product performance. In this dissertation, we contend that retailers’
behavior towards new products inﬂuences new product performance in the market. The
retailer, being the customer of the manufacturer, is not only a receiver of a new product
but also a co-producer of the new product offering to consumers. When consumers make
a purchase from stores, they buy not only a physical item produced by the manufacturer,
but also the “service” (6. g. location, store brand, ambience, customer service, etc.)
produced by the retailer (Dawson 2000). The interaction of and coordination between the
retailer and the manufacturer with respect to the new product offering is therefore
important and determines the performance of the new product. This argument is
consistent with strategic partnerships advocated in the relationship marketing paradigm
(e. g. Anderson and N arus 1990; Cannon and Perreault 1999). Therefore, the second
objective of this dissertation is to investigate the inﬂuence of the retailer’s collaborative

behavior on new product performance.

Further, though new product performance has been frequently investigated from
the manufacturer’s perspective in the literature, neither the construct nor its impact on
retailer performance has been explored from the retailer’s perspective. This is surprising
given that the retail buyer’s responsibility is proﬁt generation. The performance of new

products in stores is a major concern of buyers and is consistently monitored by retailers

because new product performance contributes to store sales (Montgomery 1975). Thus
the third objective of this dissertation is to examine the impact of new product
performance on retailer performance.
Speciﬁcally, this dissertation seeks to answer the following research questions:
1) What are the factors that determine a retail buyer’s new product evaluation?
2) How does a retailer’s new product evaluation inﬂuence the retailer’s collaborative
behavior?
3) How does a retailer’s collaborative behavior inﬂuence new product performance?
4) How does new product performance inﬂuence retailer performance?
Significance of the Study
This dissertation is an attempt to answer the call for developing a theory of retail
buying behavior (see Rao, McLaughlin and Hawkes 1995, p. 22). The main objective is
to establish a link between two different streams of literature devoted to new products in
the consumer market: the ﬁrst is new product (NP) literature, which focuses on
manufacturers’ strategies and procedures in launching new products; the second is retail
buying (RB) literature, which focuses on retailers’ buying behavior in relation to new
products. To achieve this goal, we propose an integrated model of new product
evaluation and new product success focusing on two stages in the retail buying process:
(1) the evaluation stage and (2) the post-evaluation stage. We propose our model of new
product buying as a general model because Skytte and Blunch (2005) claim that it is
possible to generalize retailers’ buying behavior across countries after they conducted an
analysis in 16 Western European countries, and Stemquist and Chen (2007) assert that

Chinese retail buying behavior is similar to Western retail buying behavior. Thus, this

dissertation represents the first attempt to test our model in one country. We test the
model with a national sample of China. Later studies will be conducted in other countries
to validate the generalizability of this model. The results—from China as well as from
other countries—will provide important theoretical and managerial implications for both
retailers and manufacturers.

Organization of the Study

Chapter one introduces the objectives and the significance of this dissertation. An
overview of this dissertation is then presented.

Chapter two reviews the previous literature on the areas of RB and NP. Relevant
issues are identiﬁed in both areas in relation to new product buying behavior. A
comparison of this dissertation to the previous studies is also provided. Theoretical
underpinnings for the model are then presented. Drawn from the literature in both areas,
hypotheses are developed for empirical testing.

Chapter three describes the study context for this dissertation and the research
methods. Background information is provided regarding retailing in China. The
development of the survey instrument, sources of the measures, and the process of data
collection are described.

Chapter four deals with analyses and discusses results of hypotheses testing.

Chapter ﬁve delineates the contribution of this study and conclusions.
Implications for academics and managers, limitations, and future research are also

provided.

CHAPTER 2

LITERATURE REVIEW AND THEORETICAL FRAMEWORK

This chapter consists of three parts. The first part deals with a review of literature
in relevant areas. The second part provides theoretical underpinnings for the model. The
third part encompasses the hypotheses development.

Part One: Literature Review

The literature review for this dissertation covers two areas: organizational buying
(OB), in particular, RB and NP. We review previous studies of retail buying behavior in
these two areas and identify relevant issues in relation to retail buying behavior.

A Review of Organizational Buying (OB) Behavior

In an OB context, a buying process is often a dynamic, intricate, and multi-phase
process involving a complex set of variables affecting buying behavior (Johnson and
Lewin 1996). Although retail and industrial buying both fall under the rubric of
organizational buying (Wager, Ettenson and Parrish 1989), research on industrial buying
has dominated the area of DB, generated proliferated results over decades, and greatly
enriched our understanding of industrial buying behavior. In contrast, studies focusing on
RB are scanty and relatively little is known about retailers’ buying behavior (McLaughlin
and Rao 1991).

Three seminal works of Robinson, Faris, and Wind (1967), Webster and Wind
(1972), and Sheth (1973) together set the theoretical foundation for OB, or more
accurately, industrial buying. These three works delineated nine sets of constructs that
affect industrial buying behavior—environmental, organizational, group, participant,

purchase, seller, informational, and conﬂict/negotiation characteristics as well as the

stages in the buying process. A large number of empirical studies have been conducted to
test these three models and results have generally conﬁrmed the validity of the models
(see Johnson and Lewin 1996 for a comprehensive review).

In the area of RB, however, researchers have commonly noted the lack of a theory
of retail buying behavior (McGoldrick and Douglas 1983; Wagner, Ettenson and Parrish
1989). Although researchers have made a few attempts at conceptualizing retail buying
behavior (e. g. Sheth 1981), empirical studies on retail buying typically borrowed
industrial buying models, in particular, the Sheth (1973) model (6. g. Silva and Davies
2002; Wagner, Ettenson and Parrish 1989). This may be due to a large degree of the
similarity between Sheth’s (1973) model and Sheth’s (1981) model. However, there are
several characteristics that differentiate RB from industrial buying:

1) Retail buyer as proﬁt center (Sternquist I 998) vs. industrial buyer as cost
controller. One important difference between industrial and retail buying is that retail
buyers are responsible for generating revenue through their purchases whereas industrial
buyers are responsible for controlling costs, which suggests that retail markup (i.e. proﬁt
margin) is important to retailer buyers while low price is imperative to industrial buyers
(Wagner, Ettenson and Panish 1989).

2) Retail buying as an individual decision-making process vs. industrial buying as a
group decision-making process.2 Unlike industrial buying decisions, which are often
made within a buying center (Webster 1884), retail buying decisions are usually made
autonomously by the individual retail buyer (Kline and Wager 1994). Many retailers use

central buyers or store buyers responsible for buying, while in an industrial setting, at

 

2 We acknowledge certain retailers may use a buying committee. However, the existence of a committee
does not inﬂuence the buying decision in a major way because the buyer makes recommendations and the
committee normally adheres to the buyer’s recommendations (Hansen and Skytte 1998).

least the personnel from the purchasing, quality control, and manufacturing departments
are involved in the joint decision-making process, and the individual purchasing agent is
often a less critical member of the decision-making process in industrial buying (Sheth
1973).

3) Retail buyers purchase ﬁnished goods for resale t0 the consumer vs. industrial
buyers purchase raw materials and component parts for use in production.
Consequently, the right goods for industrial buyers are those necessary to support the
production process, while the right goods for retail buyers are likely to be those which
buyers expect to sell well (Wagner, Ettenson and Parrish 1989). Therefore, selling
performance of products is a major concern of retail buyers while stability and
availability of components are important factors to industrial buyers.

4) Retail buying focuses on product selection vs. industrial buying emphasizes
supplier selection. Two important buying decisions are the selection of products and the
selection of suppliers. While retail and industrial buying both involve consideration of
product characteristics as well as supplier characteristics, products are likely to be the
main focus of retail buying decisions because consumers buy products in stores.3 In
contrast, raw materials or components are more like commodities, thus industrial buyers
are more likely to select vendors who can supply low-cost, reliable products.

5) Retail buyers may require trial vs. industrial products do not require trial. New
products may be accepted by buyers for a trial period and will later be discontinued if
they fail to move. For instance, Stemquist and Chen (2007) reported that in China new

products are typically admitted for a three-month trial period. Similarly, McLaughlin and

 

3 We acknowledge that, in some product categories such as cosmetics, the supplier is as important as the
product.

10

Rao (1991) described that after a new product is accepted and stocked, its sales and
performance are monitored for a certain period (e. g. 12 months) and a review is made at
the end of the period to decide whether to continue carrying the new product or to
withdraw it from shelves. Therefore, McLaughlin and Rao (1991) checked the
continuity—whether the products remained on shelves after 12 months—to validate the
buyers’ acceptance decisions. Of the originally accepted products, 31.9 percent were still
on shelves after two year while 69.1 percent were discontinued within the first two years.
On the contrary, industrial buyers normally purchase products without a trial period.

In summary, it is clearly shown that RB and industrial buying are two distinct
buying processes employing different stages and behaviors. As such, researchers have
cautioned against blindly adopting the theory of industrial buying in the retail buying
process (Johansson 2001; Pellegrini and Zanderighi 1991; Sheth 1981) and call for
making the development of a theory of retail buying behavior as a high research priority
(Rao, McLaughlin and Hawkes 1995).

Along with lack of a theoretical framework in RB literature, a review of extant
empirical studies of new product buying also reveals eight issues in two main areas:
conceptualization and methodology. Our review confirms Hansen and Skytte’s (1998)
assertion that the ﬁndings on retail buying behavior appear scattered and unrelated. Next
we will discuss each of these issues and provide a brief argument as to how this
dissertation attempts to solve the observed problems in literature. More detailed
arguments will be provided in the third part—hypotheses development section. Table 2.1
summarizes a comparison of this dissertation to previous studies in retail buying.

In the area of conceptualization:

11

1) In terms of the framing of RB situations, product and supplier evaluation are not
readily seen as occurring simultaneously. Among a handful of studies investigating retail
buying behavior, some focused on supplier (e. g. McGoldrick and Douglas 1983) while
others focused on product buying (e.g. Skytte and Blunch 2005). To this end, the
literature appears to have supplier or vendor decision on the one hand and product
decision on the other.

This dissertation argues that new product buying involves simultaneous
consideration of both the new product and the supplier. Therefore, both factors relevant
to the new product and the supplier need to be integrated into the new product buying
model.

2) Most of studies on RB have adopted the accept/reject decision as a dependent
variable.4 For example, Grashof (1970) uses computer simulation experiments to examine
accept and reject criteria associated with alternative products for one product category.
Rao and McLaughlin (1989) and Rao, McLaughlin and Hawkes (1995) are by far the
most comprehensive studies in attempting to develop a model of predicting retailer
accept/reject decisions regarding new products. However, their models were considerably
more successful in accurately predicting the reject decision than the accept decision. For
instance, Rao, McLaughlin and Hawkes (1995) achieve a correct hit rate of 94 percent of
reject decisions compared to 54 percent of accept decisions, while Rao and McLaughlin

(1989) obtain a correct hit rate of 87 percent of reject decisions compared to 42 percent of

 

4 Two exceptions were Stemquist (1994), whose focus was on the buyers’ attitudes towards imported items,
and Skytte and Blunch (2005), who asked how likely it was that the respondents would buy a hypothetical
product from a hypothetical supplier.

12

accept decisions.5 The poor prediction of the accept decision seems to signal an
inappropriate tool used by previous studies.

This dissertation proposes to use new product evaluation as a dependent variable
in the new product buying model. We argue that the accept/reject decision is a simplistic
approach to the buying decision and thus fails to reveal the richness of the result of new
product evaluation by retail buyers.

3) Most studies of RB have focused on decision criteria as independent variables.
The earlier studies have been focused on determining an array of decision criteria used by
retail buyers to make new product accept/reject decisions (e. g. Grashof 1970; Nilsson and
Host 1987; Pellegrini and Zanderighi 1991) and more recent ones have concentrated on
investigating the impact of one particular criterion—slotting allowances—on new
product acceptance (e. g. Rao and Mahi 2003; White, Troy and Gerlich 2000). One study
conducted by Nilsson and Host (1987) collects a detailed list of decision criteria and
identiﬁes a total of 394 decision criteria used by retail buyers in 34 studies. The large
amount of decision criteria, however, may reduce their practical implications (i.e. too
cumbersome for manufacturers to target) because different buyers are likely to adopt
different criteria for decision-making (Hansen and Skytte 1997). Moreover, the dominant
focus on decision criteria identiﬁcation seems to signify that much of the research to date
is exploratory in nature, rather than explanatory (i.e. emphasizes testing hypotheses of
effects and differences).

This dissertation seeks to test hypotheses and identifies factors at construct level

rather than decision criteria at item level that affect retail buyers’ evaluation of new

 

5 The authors followed up with buyers and found that “special characteristics seem to be associated with
some of those items accepted by buyers but predicted as rejects by the model” (Rao and McLaughlin 1989,
p.86).

13

products. Two factors identified are the product factor and the supplier factor. This
approach is considered more parsimonious than using a pool of decision criteria.

4) Most studies treat the RB process as one stage—the decision stage in which retail
buyers make accept or reject decisiOns, thus almost all of the models stop at the decision
point; little is known after new products are accepted. For example, White, Troy and
Gerlich (2000) study the impact of slotting fees and introductory allowances on new
product acceptance moderated by other variables such as gross margin and competition.

This dissertation represents an integrated model consisting of both new product
evaluation and post-evaluation stages. We contend that among new products accepted in
stores, retail buyers have different evaluations of the potential success of new products
and therefore exert different behavior in coordinating with suppliers.

5) New product performance and its impact on retailer performance were rarely
examined in RB literature. McLaughlin and Rao (1991) was the only study that examined
new product performance in stores. However, they used continuity—whether new
products are still on the retail shelves after two years of acceptance—as a proxy of new
product performance. This is an unsophisticated measure of new product performance.

In this dissertation, we argue that new product performance is an important driver
of retail performance. New products make a substantial contribution to total retail sales;
about one sales dollar in eight is due to extremely new products (Montgomery 197 5).
Since retail buyers are responsible for generating revenue, new products that fail to
achieve the expected performance will be delisted. Therefore, examining the impact of
new product performance on retailer performance represents a logical further step of

understanding retailer new product buying behavior. Accordingly, this dissertation

14

includes new product performance and retailer performance as dependent variables in the
model as well.
In the area of methodology:

1) Logistic regression is by far the most frequently used method for model testing in
RB literature. This is related to the issue of using the accept/reject decision as the
dependent variable. For example, Rao and McLaughlin (1989) use data collected on new
products presented to a large supermarket chain to estimate logistic regression models to
describe the retailer’s accept/reject decisions. Similarly, Gerlich, Walters and Heil (1994)
use logistic regression to test accept/reject decisions by buyers from one supermarket
chain. Although regression has its merits in identifying the predictors, the method shows
relationships rather than causation between independent and dependent variables.
Further, the interrelationships among the independent variables are not readily revealed.

In this dissertation, the new product buying model is tested by using the structural
equation model (SEM). This method provides testing of hypotheses in relation to
constructs as well as testing of model fit of the whole model.

2) Typical respondents in many previous studies are a small set of buyers from one
or a few retailers. For example, Montgomery (1975) studied the accept/reject results of
three supermarket buyers in the Boston area. Rao, McLaughlin and Hawkes (1995)
studied the accept/reject decisions of five buyers from a packaged grocer.

In this dissertation, we conducted a national survey of retail buyers from multiple
retail organizations.

3) Previous studies are mainly conducted in the US, European countries, and other

developed countries such as Japan (e. g Alpert et al. 2001) and New Zealand (e.g. Marr

15

and Thomas 1999). Only two studies took place in a developing country (Hansen 2001;
Stemquist and Chen 2007).6 This is probably because most of developing countries are
classiﬁed as sellers’ markets (Frazier, Gill and Kale 1989; Kale 1986); investigating
retailer new product buying behaviOr is meaningless because manufacturers dictate what
retailers sell.

This dissertation is conducted in China—a developing country in a stage of
economic transition. Chapter three will describe the study context.

In summary, past studies of new product buying behavior in the area of RB
typically examined accept/reject decisions made by buyers on either actual or
hypothetical products from a small number of retailers based on decision criteria
identiﬁed by retail buyers in developed countries.

A Review of New Product (NP) Literature

Among the stages of new product development, new product launch is relatively
under-researched in NP literature despite its importance in driving new product success
and high associated implementation cost (di Benedetto 1999). The literature reviewed
focuses on the area of new product launch and new product performance and reveals the
following characteristics:

1) Although distribution has been identiﬁed as one of the success factors in new
product launch (di Benedetto 1999), the majority of empirical studies have focused on
manufacturer strategies and procedures (e. g. Li & Calantone 1998). This is probably
because in NP literature, distribution is treated as a tactical decision rather than a strategic

decision. The only study involving retailers was Hultink, Tholke and Robben (1999).

 

6 Hansen (2001) reported an empirical study on supplier selection criteria as rated by seafood buyers for
Chinese supermarkets. Stemquist and Chen (2007) conducted a qualitative study on the new product
buying of Chinese retailers.

16

However, they followed the approach of Rao and McLaughlin (1989), which is typical in
RB literature, i.e., identifying decision criteria of Dutch retailers’ adoption of new
consumer durables.

2) The majority of studies inveSti gating new product performance examine the issue
from the manufacturer’s perspective (see Montoya-Weiss and Calantone 1994 for a
review). To date, none of the studies has measured new product performance from the
retailer’s perspective.

In summary, a review of the literature in the area of NP introduction draws the
same conclusion as did Doyle and Weinberg (1973) more than 30 years ago, who claim
that research on NP introduction has focused on the decisions and strategies of
manufacturers, virtually ignoring the retailer’s role. The retailer is truly a neglected link
in the new product development chain.

A Synthesized Conclusion of Literature Review

Quite surprisingly, despite researchers’ call for an expanded incorporation of RB
in NP research (see Rao, McLaughline and Hawkes 1995; Wind and Mahajan 1987),
little effort has been made so far in the NP area. The two sets of studies (RB and NP)
appear to run parallel to one another (Pellegrini and Zanderighi 1991) speaking their own
languages. For example, while new product performance is well-researched in NP
literature, it has not been examined in RB literature despite its obvious importance to
retailers. Further, although strategic factors such as new product advantage are among the
most studied factors in NP literature, new product advantage as a construct never
appeared in RB literature. However, some of the items included in new product

advantage have been frequently used as decision criteria (e. g. quality; uniqueness) in RB

17

literature. Therefore, RB and NP studies have the potential to be and should be integrated
to reﬂect the retailer’s involvement in the process of new product introduction. This

dissertation represents an attempt at achieving this goal.

18

Table 2.1 - A Comparison of this Dissertation to Previous Studies

 

Issues in Previous
Studies

Examples

In this dissertation

 

Conceptualization:

Product and supplier
evaluation are
examined separately.

Accept/reject
decision as a
dependent variable.

Decision criteria as
independent
variables.

Only the decision
stage is included.

New product
performance has not
been examined from
retailers’ perspective.

Methods:

Majority use logistic
regression analysis.

Respondents are
typically a small set
of buyers from one or
a few retailers.

Focuses are '
developed countries
in the Western.

Wagner, Ettenson and Parrish (1989)
examined retailer supplier selection
decision while Skytte and Blunch
(2005) studied retailer product selection
decisions.

Grashof (1970) used computer
simulation experiments to examine
addition and deletion of criteria
associated with alternative product for
one particular product category, dog
food.

Thomas and Marr (1993) asked
respondents to rate the importance of 41
criteria compiled by authors.

White, Troy and Gerlich (2000) studied
the impact of slotting fees and
introductory allowances on new product
acceptance.

One study that touched on new product
performance was McLaughlin and Rao
(1991) who checked whether new
products were still on shelves two years
after the acceptance.

Gerlich, Walters and Heil ( 1994) used
logistic regression to test one
supermarket chain’s accept/reject
decisions.

Rao, McLaughlin and Hawkes (1995)
studied ﬁve buyers from a retailer in
packaged grocery.

Hansen (2001) reported an empirical
study on supplier selection of Chinese
supermarkets.

Both the product factor and
the supplier factor are
considered.

New product evaluation
used as a dependent
variable.

The product factor (i.e. new
product advantage) and the
supplier factor (i.e. trust)
are independent variables.

Both retail new product
evaluation and post-
evaluation stages are
included. Retailer relational
behaviors are examined.

New product performance
and its impact on retail
performance are examined.

The structural equation
model is used.

186 buyers from more than
40 retailers are studied.

Focus is on a large
developing country—
China.

 

19

           

  
   

   

oozes—totem
SEE eve—gum

coca—Ectom
Hoe—.82 3:83—

coca—Eaton—
.eaee... sez

 

 

  

 

:oaSEEEEoU

22223000

5:323:00

\

A

 

 

own :55

users, 388

 

 

.25.

 

 

 

La

news-=95“
.2695 302

 

 

 

eemsem
9 E2
£083.50
9 SE

owS§>v< 835.5 302

 

 

 

 

             

E55323—
Esombm

oocotonxm
“mam

song—mom

      

mmoco>§>ocﬁ
8:85

8.883%?— uoscoum .392 can Sana—gm— 8355 302 no .352 tongue“...— :< I ﬂu 2:»?—

20

Part Two: Theoretical Framework

With the intention to connect two bodies of literature on RB and NP, we propose
an integrated conceptual model of retailer new product evaluation and new product
success (Figure 2.1). By “integrated”, we mean that (1) concepts from both RB literature
and NP literature are incorporated, and (2) the model includes both an evaluation stage
and a post-evaluation stage. Next we describe the theoretical underpinnings that provide

rationale for this model.

Theoretical Underpinnings
Focus on Evaluation

The central focus of our model is the evaluation of a new product formed by the
retail buyer, which stems from Mowen and Gaeth’s (1992) scheme of the evaluation
stage in marketing decision-making. Decision-making is vital to any ﬁrm facing choices
of investment, management, and deployment of resources in a competitive environment
(Crozier and Ranyard, 1997). For the retailer, new product buying decision are essentially
decisions of the re-allocation of resources (e. g. shelf-space), which offer both signiﬁcant
problems and substantial proﬁt opportunities (Doyle and Weinberg 1973). Therefore,
prior to making important buying decisions, an evaluation stage needs to occur in which
judgments about the new product are made.

Mowen and Gaeth (1992) contend that the evaluation stage is fundamental to any
marketing decision-making because judgments form the foundation of the decision.
Drawing from the decision-making models found in the behavioral decision theory
literature, Mowen and Gaeth (1992, p.178) deﬁne “evaluation” in marketing decision-

making as “the judgment of probability, the judgment of value, and the integration of

21

these two components into an overall assessment of the outcome.” This deﬁnition is
consistent with Hogarth’s (1987) argument that there are two different types of judgment
tasks in the evaluation process: the valuation of outcomes and the estimation of
probabilities. According to Mowen and Gaeth (1992), the judgment of value involves the
assessment of the goodness or badness of an event whereas the judgment of probability
refers to making a prediction about how likely an event is to occur or about how
frequently it will occur. Together, evaluation denotes the overall utility or assessment of
an option as well as its probability of occurrence.

In a new product buying setting, new product evaluation signiﬁes the assessment
of the attributes of a new product (i.e. the valuation judgment of superiority or inferiority
of a new product) and the prediction of its performance (i.e. the probability judgment of
the likelihood of future success). Attribute assessments and performance estimates are
integrated to form an overall evaluation of a new product. Consequently, we contend new
product evaluation is a broader and more central concept than the accept/reject decision
in understanding retailer new product buying behavior and therefore adopt new product
evaluation as the focus construct in our model.

The Evaluation Stage

A reduced form of Sheth’s (1981) model (see Figure 2.2) of merchandising
buying behavior is adopted as a general framework representing the evaluation stage of
our model. Sheth’s (1981) model of merchandising buying behavior is similar to Sheth’s
(1973) model of industry buying behavior in that both buying behaviors are affected by
two factors: “merchandise requirement” and “supplier accessibility” in Sheth’s (1981)

model and “product-speciﬁc factor” and “company-speciﬁc factor” in Sheth’s (1973).

22

These two models explicitly demonstrate that a buying process is an assessment of both

the product and the supplier. Moreover, the choice calculus process prior to the actual

choice in Sheth’s (1981) model is actually an evaluation stage. Based on our

aforementioned discussion about the importance of the evaluation stage in new product

buying, we use this reduced form of Sheth’s (1981) model to represent the new product

evaluation stage in our model. Consistent with the Sheth model, two factors that affect

new product evaluation are the product factor and the supplier factor.

Figure 2.2 - A Reduced Form of the Sheth Model

 

Product

 

 

 

 

 

 

 

 

Choice
Calculus

 

 

 

 

l

 

Suppﬁer

 

 

 

Source: Sheth (1981).

23

 

Ideal
Supplier/Product
Choice

 

——>

 

 

 

Actual
Supplier/Product
Choice

 

 

The product factor is denoted by new product advantage because new product
advantage refers to differences in important attributes between a new product and
alternatives (Day and Wensley 1988), and these attribute differences are the base upon
which value judgment is assessed inthe evaluation stage (Mowen and Gaeth 1992). The
supplier factor is denoted by retailer trust of the supplier in our model. Transaction Cost
Analysis (TCA) provides a theoretical explanation for using trust as a supplier factor.

Transaction Cost Analysis belongs to the “New Institutional Economics”
paradigm. The theory was ﬁrst suggested by Coase (1937) and further developed by
Williamson (1975; 1985; 1991; 1993). Earlier works by Williamson (1975, 1985)
polished the TCA framework by adding considerable precision to Coase’s general
arguments, and more recent works by Williamson (1991; 1993) expanded Coase’s initial
proposition of ﬁrms and markets as polar modes of governance structures to a relative
continuum consisting of market, hierarchy, and a variety of hybrid forms in between.
Although derived from economics, its substantive focus on exchange and transaction
makes TCA particularly relevant to a wide range of marketing phenomena (Rindﬂech and
Heide 1997). Over the past two decades, TCA has received considerable attention in the
marketing literature covering vertical integration in channel (e. g. Anderson and Weitz
1983), sales force management (e. g. Atuahene-Gima and Li 2002), foreign market entry
(e. g. Anderson and Coughlan 1987), buyer-seller relationship (e. g. Dwyer, Schurr and Oh
1987), and industrial purchasing strategy (e.g. Stump and Heide 1996). None of this
attention, however, has been paid to retail buying behavior as yet. This is somewhat

surprising because two main assumptions of human behavior (Rindﬂech and Heide 1997)

24

make the theory particularly suitable for the context of the retailer new product buying
situation.

The ﬁrst assumption, bounded rationality, is the cognitive assumption that
decision makers have constraints on their information collecting and processing
capabilities and limits on their rationality (Simon 1957). This is a common phenomenon
for decision makers during the evaluation stage (Mowen and Gaeth 1992). For example,
the retail buyer may not have the complete information to assess all of the attributes of a
new product because they are not privy to the entire research and development (R&D)
process behind each new product (Patterson and Richards 2000). This type of problem is
also called hidden information. Further, a retail buyer might have difﬁculty in predicting
the future performance of the supplier in areas such as reliable delivery ex ante. To the
extent that a new product success is partially dependent on a supplier’s collaboration and
performance, the behavior uncertainty related to the supplier may augment the problem
caused by the bounded rationality of the buyer. This type of problem is also called hidden
action.

The second assumption, opportunism, is the behavioral assumption that a party
may seek to serve self-interest given the chance, leading to such behaviors as lying,
cheating, violating agreements, and/or other more subtle forms of dishonesty (Williamson
1993). For example, a supplier may break an exclusive selling contract. The opportunistic
behavior may happen particularly when there is asset speciﬁcity involved. For example, a
retailer may have already invested in advertising or promotion for selling an exclusive
product. This “lock-in” situation may prompt the manufacturer’s opportunistic behavior.

In a new product buying setting, the potential problems associated with bounded

25

rationality of the retail buyer and opportunistic behavior of the supplier increase
transaction costs of accepting a new product. In order to minimize transaction costs,
exchanges need to be organized so as to economize on bounded rationality of the buyer
while simultaneously safeguarding them against the hazards of the potential opportunism
of the supplier (Williamson 1993). According to the current theory of TCA, retailer trust
of the supplier may serve as a hybrid governance mechanism that reduces transaction
costs and mitigates opportunism in the new product buying context characterized by
uncertainty and bounded rationality. Williamson (1991) refers to such trust as
“calculative trust”,7 which is essentially inferred trust, because an individual or
organization calculates the costs and/or rewards of another party cheating or not cheating
and decides whether the other party is trustworthy. For example, a supplier with a good
reputation may be perceived by a retailer as trustworthy because of the efﬁcacy of
reputation in the business community—the immediate gain from opportunism in the ﬁeld
must be traded off against future costs and sanctions (Williamson 1991). In other words,
the retailer weights the cost and infers that, for the supplier, the repercussions of being
caught cheating are larger than the rewards of cheating, and the supplier would rather not
cheat and therefore stay trustworthy.

In the literature, trust serving as an alternative to formal control mechanisms has
been validated in many studies (e. g. Bradach and Eccles 1989; Doney and Cannon 1997).
Therefore, we propose trust as a supplier factor that helps to mitigate risks associated

with the opportunism of a supplier and the uncertainties related to the retail buyer’s

bounded rationality in new product buying.

 

7 Williamson (1991) argues that personal trust is nearly noncalculative, thus, nonexistent in commercial
exchange relationships.

26

The Post-Evaluation Stage

The success of new products depends ultimately on consumers accepting them
(Hauser, Tellis and Grifﬁn 2006). The theory of the diffusion of new products argues that
the sales of successful new products typically follow an S-shaped developmental curve as
Figure 2.3 shows: the sales start slowly in the initial introduction stage; after some time,
sales take off, showing a strong increase in growth rate; finally, sales ﬂatten as the market
saturates when a certain level of marketing penetration has been reached (Rogers, 2003).
When a new product enters the market, sales increase slowly. During this time, sales are
mostly driven by external inﬂuences, such as promotions and mass media advertising
aimed at making the product take off (Delre et al. 2007). Many promotional and
advertising activities are co-planned with retailers. Therefore, the behavior of the retailer
after stocking a new product is vital to the new product’s success as well.

Further, in arguing that the foci of economics and marketing have shifted from the
producer to the customer, Vargo and Lusch (2004) propose a new logic of market—
“service marketing”—in which intangibility, exchange processes, and relationships are
keys to gaining competitive advantage. A service marketing orientation embraces a view
of involving the customer as a co-producer of a market offering; thus, the value of the
market offering is co-determined by both the producer and the co-producer. Applying this
logic to our context, we argue that the manufacturer and the retailer co-create the value of
a new product offering, and the behavior of both parties will determine the success of the
new product in the market. Moreover, Vargo and Lusch’s (2004) viewing of a customer

as co-producer is also consistent with the relationship marketing paradigm. The

27

philosophy of relationship marketing maintains that developing strategic partnerships
characterized by coordination and collaboration between buyers and sellers is crucial to
ﬁrm success. Retailers and suppliers typically have tended to deal with each other at
arms—length, viewing each other as adversaries where one party wins at the expense of
the other (W illson 1994). The view is now obsolete and especially wrong during the new
product introduction situation. A successful new product brings beneﬁts to both
manufactures and retailers. Therefore, coordinating retailer and supplier behaviors during
the introduction of a new product and emphasizing cooperation, information-sharing, and

commitment are likely to bring mutual beneﬁts and drive the success of the new product.

Figure 2.3 - The S-Shaped Curve of New Product Diffusion.

 

u Emma
Sales Inﬂuence
A /
Takeoff
External
Inﬂuence

 

 

 

I

 

 

 

 

Time

Source: Delre et a1. (2007)

28

Part Three: Hypotheses Development
New Product Advantage, its Antecedents, and its Effect

New product advantage. New product advantage is defined as “the customer’s
perception of product superiority with respect to quality, cost-benefit ratio, or functions
relative to alternatives” (Montoya-Weiss and Calantone, 1994, p. 415), or “the beneﬁts
that customers get from the new product” (Langerak, Hultink and Robben 2004, p. 82).
The construct has been extensively studied in NP literature. Though the definitions of
new product advantage embrace the customer’s perception, almost all studies examining
this construct have been from the manufacturer’s perspective (e. g. Li and Calantone
1998; Song and Parry 1994). Our approach to new product advantage however differs
from the one in NP literature. We take the perspective of a retailer, a customer of the
manufacturer.

Retailers not only buy new products but also sell new products to consumers. In
deciding on new product selection, the retailer takes into account the ability of the new
product in generating store trafﬁc, margin, cannibalization, etc. (Desiraju 2001). As such,
during the new product evaluation stage, the retailer buyer considers product advantages
to retailers as well as to consumers. A new product that provides an opportunity for a
higher margin is a product advantage to retailers, while a new product with a new feature
may represent a product advantage to consumers. For example, Sony’s PlayStation
followed Nintendo into the market, but it used CDs instead of tapes, which allowed the
consumer a wider array of content. Similarly, Canon successfully followed Konica into
the market for automatic focus cameras by adding a device that made it possible for the

auto focus to work at night (Kono and Lynn 2007). Product advantage to consumers and

29

product advantage to retailers are also interrelated. A new product generating high store
trafﬁc—a product advantage to retailers—does so may because the new product meets a
previously unmet need of consumers—a product advantage to consumers. Corstjens and
Corstjens (1995) argue that consumer product manufacturers can improve their new
product success rates if they put more effort into creating product value for retailers as
well as differential product advantage for consumers. Consequently, we propose new
product advantage as a second-order construct with two dimensions—product advantage
to consumers and product advantage to retailers.
The Antecedents of New Product Adentage

Product Innovativeness. Product innovativeness is deﬁned as the degree of
newness of a new product, or the degree to which a new product differs from other
existing products (White, Troy and Gerlich 2000). In RB literature, product
innovativeness is usually represented by one item: either newness (e. g. Grashof 1970) or
uniqueness of a new product (Rao and McLaughlin 1989). For instance, in one study,
Montgomery (1975) studies retail buyers for a supermarket chain and ﬁnds that the
newness of a new product is one of the top criteria used by retail buyers in new product
acceptance decisions. In another study, Rao, McLaughlin and Hawkes (1995) examine
the buying decisions of both actual products and hypothetical products and ﬁnd that
product uniqueness is a significant factor in determining new product acceptance. In
contrast, the conceptualization of product innovativeness in NP literature is much more
complex. For instance, Lee and O’Connor (2003) conceptualize product innovativeness
as a multidimensional construct consisting of four dimensions: product newness to the

firm, market newness to the ﬁrm, product superiority to the customer, and ad0ption

30

difficulty for the customer. Yet, Garcia and Calantone (2002) classify new product
innovativeness as having two dimensions: a macro (i.e. industry) level and a micro (i.e.
firm) level. Nevertheless, despite different conceptualizations adopted by researchers,
past studies in general found a positive relationship between new product innovativeness
and new product advantage (Calantone, Chan and Cui 2006; Gatignon and Xuereb 1997).
This logic is intuitive because an innovative new product created with the purpose of
being superior to the existing products must possess certain product advantages to be
superior. Thus, consistent with the literature, we posit:

H 1: Product innovativeness is positively related to a) new product advantage to
consumers and b) new product advantage to retailers.

The Effect of New Product Admtage

The eﬁect of new product advantage on new product evaluation. The retailer’s
evaluation of a new product includes the judgment of the total utility value of the new
product represented by the attributes of the new product and the likelihood of its potential
success (Mowen and Gaeth’s 1992). The attribute superiority of a new product increases
the likelihood of its success. Based on a synthesis of over 3000 studies, Rogers (1983)
concludes that a new product will be successful when it has a relative advantage over
existing products. The Theory of Innovation Diffusion (Rogers 1995) also holds that
there will be an increased rate of diffusion if consumers perceive the new product to have
a relative advantage. On the contrary, new products without a relative advantage do not
meet consumer needs and are likely to fail. For example, many food manufacturers have
recently been launching new low-carbohydrate products. In 2004, the number of new
low-carbohydrate products jumped from around 500 to over 5,000. However, many of

those were “me-too” products and thus failed quickly. Not surprisingly, now the talk in

31

the industry is of products being axed and of truckloads of unsold low-carbohydrate
snacks and meals being donated to charities for the homeless (Economist 2005).

New superior and differentiated products are products with superior advantages
delivering unique benefits and greater value to the customers (Cooper 2005); new
products without advantages imply product parity. Product parity often happens when
several competitors produce “me-too” products with almost identical quality and
indistinguishable attributes, which leads to substitutability and therefore provides less
incentive for retailers to stock the product (Corstjens and Corstjens 1995). Empirical
studies also provide evidence of the retailer’s preference of new products with product
advantage. For example, gross profit margin, a product advantage to retailer, is found as a
highly signiﬁcant factor in determining retailer new product acceptance (Rao,
McLaughlin and Hawkes 1995; Wagner, Ettenson, and Parrish 1989). Similarly, Alpert,
Kamins and Graham (1992) study the retailer preference of new products and ﬁnd that a
pioneer brand, which is defined as a new product that is signiﬁcantly different from any
other products in the judgment of a retail buyer, is greatly preferred by retailer buyers
than a “me-too” product, which is deﬁned as a new SKU that the retail buyer perceives to
be about the same as a previously introduced item. Thus, we propose that new product
advantage is positively related to the retailer’s evaluation of the new product.

H2: New product advantage— a) new product advantage to consumers and b) new

product advantage to retailers—is positively related to retailer new product
evaluation.

32

The effect of new product advantage on new product performance. The superior
product is its own promoter.8 The sensory experience of a product is a key factor
determining consumer response and evaluation of a new product (Mooy and Robben
2002). Therefore, new product advantage consistently appears as the most important
factor in explaining the adoption and success of the new product (Montoya-Weiss and
Calantone 1994). Empirical studies in NP literature show that new product advantage
leads to superior product performance. For example, one study shows that superior new
products contribute to incremental sales, generate excitement about going shopping, meet
an unmet need, and have potential to achieve high volume (Alpert, Kamins and Graham
1992). And another study reports that superior new products have ﬁve times the success
rate, more than four times the market share, and four times the proﬁtability of products
without product advantage (Cooper and Kleinschmidt 1990). Cooper (2005) synthesizes
numerous research studies on new product development and concludes that the number
one driver of new product success is the product itself—superior and differentiated
products. Thus, we propose:

H3: New product advantage— a) new product advantage to consumers and b) new
product advantage to retailers—is positively related to new product performance.
Trust, its Antecedents, and its Effect

Trust. Trust is generally deﬁned as the belief by one party that the other party is
reliable and will fulfill obligations in an exchange relationship (Morgan and Hunt 1994).
In short, trust refers to the buyer's belief that the seller is reliable. In a new product

buying situation, the buyer may not have complete information about the new product;

 

8 Though it is generally true, some of the innovative products may be too radically different from existing
products; thus lack of familiarity from consumers and results in low consumer acceptance (Calantone, Chan
and Cui 2006).

33

thus, the buyer’s trust toward the supplier serves an important role in reducing transaction
costs and mitigating opportunism in exchange contexts characterized by uncertainty
(Doney and Cannon 1997).

Trust as an important construct has received a great deal of attention from
sociology, social psychology, economics, and marketing. Trust literature suggests that
trust can be individual-to-individual (Rotter 1967), individual-to-institution (La Porte and
Metlay 1996), individual-to-organization (Morgan and Hunt 1994), organization-to-
individual (Doney and Cannon 1997), and organization-to-organization (Zucker 1986).
Dooney and Cannon (1997) maintain that trust in the context of buyer-seller relationships
mainly take two forms: buyer’s trust of the supplier firm and buyer’s trust of the
salesperson.

Buyer’s trust of the salesperson is indeed personal trust, which is different in
nature from interorganizational trust (Anderson and Narus 1990). Williamson (1991)
argues that personal trust is not likely to exist in commercial exchange relationships.
Other researchers often have a similar opinion. For example, Zucker (1986) contends that
personal trust is by no means sufﬁcient to produce the quantity and quality of trust
needed in highly differentiated socio-economic systems. In other words, trust based on an
individual’s integrity can only fulﬁll a supplementary function, compared with
interorganizational trust. In the same vein, Lane and Bachmann (1996) show that trust-
based relations between buyer and supplier ﬁrms rarely evolve spontaneously on the level
of individual interaction, but, rather, emerge at organizational level. Interestingly,
although claiming that both buyer’s trust of the supplier ﬁrm and buyer’s trust of the

salesperson are important in affecting industrial buyer’s purchasing behavior, Dooney

34

and Cannon (1997) fail to find a direct inﬂuence of buyer’s trust of the salesperson on
buyer’s behavior in their empirical study. Therefore, in this dissertation, we examine
interorganizational trust in retailer new product buying behavior. Nevertheless, we do
acknowledge the inﬂuence of a salesperson on interorganizational trust. However, we
suspect the inﬂuence is likely in the form of personal relationships between buyers and
salespersons and therefore treat the variable as an antecedent of buyer’s trust of the
supplier.

The Antecedents of Trust

Williamson (1991) suggests that trust primarily involves a calculative process,
that is, one party calculates the costs and/or rewards of the other party cheating or staying
in the relationship. To the extent that the beneﬁts of cheating do not exceed the costs of
being caught, one party infers that it would be in the other party’s best interest not to 0
cheat and therefore the other party can be trusted (Akerlof 1970). As such, calculative
trust is in effect an inferred trust that can be derived from personal and indirect
experiences.

Supplier Reputation. Supplier reputation is defined as the extent to which ﬁrms
and people in the industry believe a supplier is honest, reliable, and caring for its
customers (Doney and Cannon 1997). Supplier reputation is a risk-reducing factor in
retail buying decisions (White, Troy and Gerlich 2000), and thus serves as a mechanism
to enforce agreements (Boyd and Richerson 1985). A good or bad reputation of a supplier
is easily spread across ﬁrms in the industry. A retail buyer thus “knows” a supplier
through interactions with others. Granovetter (1985) argues that economic action is

embedded in structures of social relations. In other words, ﬁrms do not behave or act as

35

atoms outside a social context; instead, their actions are inﬂuenced by interactions with
others and between others. In this way, retail buyers can acquire information from the
words and experience of other ﬁrms and people. Using this indirect experience, buyers
can infer the trustworthiness of a supplier (Doney and Cannon 1997).

Further, to the extent that a good reputation takes a long time to build, buyers can
also calculate whether the costs of a supplier acting opportunistically exceed the rewards
of maintaining a good reputation, and if they do not, the buyer can infer that the supplier
will keep acting in a trustworthy manner (Telser 1980). The positive relationship between
supplier reputation and buyer trust has been well supported by empirical results in
previous studies. For example, a study of retail buyers and their suppliers ﬁnds that a
supplier’s favorable reputation leads to the retailer’s trust (Ganesan 1994). Likewise,
Anderson and Weitz (1989) ﬁnd that a manufacturer’s reputation is positively related to
its channel partner’s trust. Thus, we posit:

H4: Supplier reputation is positively related to retailer trust of the supplier.

Past Experience. A retailer’s past experience with a supplier is deﬁned as the
length of relation between the buyer and the supplier. The retailer’s past experience
implies that the retailer has a prior relationship with the supplier. Manufacturers
constantly develop and introduce a stream of new products into the market. Therefore,
not all new products are introduced by new suppliers. Many times, a new product
presented to a retailer is from a long-time supplier. For example, P&G introduces many
new products to Wal-Mart every year. Retailers treat old and new suppliers differently. A
study of all Italian major retailers finds that new products introduced by established

suppliers have one in two chances of being accepted, while those by new suppliers have

36

 

less than a one in five chance (Pellegrini and Zanderighi 1991). A prior relationship
between the buyer and the supplier helps reduce the perceived level of purchase related
risk (Johnson and Lewin 1996) because over time, two parties become well acquainted
with each other, which allows them to predict each other’s actions (Anderson and Weitz
1989). Long time relationships lay a foundation for trust because adjustments are made
and only satisfactory and trustful relationships are maintained (Anderson and Weitz

1989). Thus, we propose:

H5: Retailer’s past experience with the supplier is positively related to retailer trust of
the supplier.

Personal Relationships. Personal relationships refer to the social relationships
between buyers and salespersons. Granovetter’s (1985) seminal work provides an
argument on the ubiquity of social embeddedness in economic exchange. “Most behavior
is closely embedded in networks of interpersonal relations” (Granovetter, 1985, p.504).
His insight highlights a simple but critical regularity in economic exchange that is
commonly observed (Moran 2005): social relations and economic relations are often
intertwined.

Because it is personal experience and known to be accurate, a buyer’s personal
relationship with a salesperson is the best source of information that he can obtain to
make inferences about the seller (Williamson 1991). Personal relationships between
buyers and salespersons facilitate problem solving and are important in overcoming
barriers to communication (Metcalf, Frear and Krishnan 1992). In the literature, Weitz
and J ap (1995) argue that personal relationships between boundary-spanning members

play an important role in aiding and enhancing interorganizational exchange. Personal

37

relationships between individuals are found to shape the context for new exchanges
between ﬁrms by reducing risks and uncertainty about the motives and intentions of the
other member (Larson 1992). From a transaction cost perspective, Standiﬁrd and
Marshall (2000) claim personal relationships help to reduce transaction costs in an
exchange and ensure the other firm’s trustworthiness and abstention from opportunism.
Therefore, we propose:
H6: Personal relationships between the retail buyer and the salesperson are positively

related to retailer trust of the supplier.
The Effect of Trust

Effect of trust in new product evaluation. Trust has been widely recognized as a
mechanism to reduce risk and uncertainty in the literature. Three studies have been found
to examine the role of trust in buying/purchasing decisions. Using in-depth face-to-face
interviews with retail buyers and managers in ﬁve European markets (Germany, Greece,
Italy, Netherlands, UK), Knight, Holdsworth and Mather (2007. p. 121) conclude that
“trust appears to be the central component that enhances perceived quality while
minimizing perceived monetary sacriﬁce—thereby leading to enhanced perceived value,
and therefore a willingness to buy.” Banting and Blenkhom (1988) conduct a study on
non-food chain store buyers in Canada and ﬁnd trustworthiness of suppliers is among the
highest ranked criteria of chain store buyers in decisions of new product selection. In the
situations of buying food products from a foreign country, Knight, Holdsworth and
Mather (2007) find that trust in integrity of production and supplier appears to the central

determining factor of the retailer’s buying food imports.

38

Retailers argue that stocking new products is risky because new products incur
extra costs, which retailers consider fairly high, at both warehouse and store level, and
new product failure rates are quite high (Pellegrini and Zanderighi 1991). Therefore,
during the new product evaluation stage, retailers carefully assess the value of a new
product and the probability of its success. The trustworthiness of the supplier helps to
increase the buyer’s confidence in the new product’s success and reduces the perceived
transaction costs. Thus, we posit:

H7: Retailer trust of the supplier is positively related to retailer new product
evaluation.

Effect of trust on collaboration. Trust—or conﬁdence in the reliability and
integrity of an exchange partner (Morgan and Hunt 1994) and a willingness to rely on
that conﬁdence (Moorman, Deshpande, and Zaltman l993)——is thought to be a building
block or foundation for collaboration including such behavior as cooperation,
communication, and commitment. In prior studies, trust had a positive effect on
cooperation (Morgan and Hunt 1994), communication (Dwyer, Schurr and Oh 1987), and
commitment (Geyskens et a1. 1999). Retailer trust of the supplier positively inﬂuences a
retailer’s cooperation with, communication with, and commitment to a long-term
relationship with the supplier. In buyer-seller bargaining situations, Schurr and Ozanne
(1985) find trust to be central to the process of achieving cooperative problem solving
and constructive dialogue. Ganesan (1994) argues that a retailer’s trust in a supplier
affects the retailer’s commitment in three ways: 1) it reduces the perceived risk related to
possible opportunistic behavior by the supplier, 2) it lowers costs in exchange
transactions, and 3) it increases the chances of forming a long-term relationship with the

supplier. Trust in a relationship builds the desire to strengthen and continue—that is, to

39

cooperate and to communicate with each other, and to commit to a relationship
(Andaleeb 1996). Thus, consistent with prior research, we posit that:
H8: Retailer trust of the supplier is positively related to retailer collaboration with the

supplier.
New Product Evaluation and its Effect on Collaboration

Modern retailers do not act simply as intermediaries passing products to
consumers; they also “manufacture the service” to consumers (Dawson 2000) and co-
produce the product offering in-store through in-store display, promotion, and
communication with consumers in and/or out of stores (Corstjens and Corstjens 1995).
For example, a similar style of shoes sold at Target and Nordstrom will have different
sales records because different service levels and store images are likely to inﬂuence the
quality perception and image of the shoes.9 In other words, similar shoes sold in these
two stores actually represent different product offerings and thus render different values
in the eyes of consumers. Consequently, we can say that retailers have leverage over the
perceived value of a product. The retailer can also inﬂuence the sales of a product
through local advertising and other selling efforts such as a selling presentation, personal
recommendation or advice solicited by the consumer (Porter 1974). Therefore, the
retailer’s collaborative behavior affects product performance. A Harvard Business case
reports how collaboration between Wal-Mart and P&G helps to reduce operational costs
and increase revenues for both parties (Sebenius and Knebel 2007). Wal-Mart and P&G
have developed a collaborative relationship through the development of sophisticated

programs such as JIT (J ust-In-Time) delivery, EDI (Electronic Data Interchange), and

 

9 We acknowledge that Target and Nordstrom are not likely to sell the same shoes. However, we use this
imaginative example to illustrate different values added by different retailers on the same product.

40

ECR (Efﬁcient-Consumer-Response) Systems. Because of these programs, P&G is able
to monitor sales of products in Wal-Mart stores as well as promote and ship its goods in
response to actual consumer demands. In this way, P&G and Wal-Mart have turned what
used to be an adversarial, win-lose situation—with no sharing of information, no
coordination—into a win-win situation—with a partnership beneﬁcial to both. Thus,
collaborative behavior creates ﬁnancial linkages that make ﬁrms dependent upon mutual
performance (Calantone, diBenedetto and Stank 2005).

After a new product is accepted in stores, the performance of the product is likely
to be inﬂuenced by the retailer. A new product evaluated as highly favorable by the retail
buyer indicates the perceived superiority of the product with a high probability to
contribute to retailer profitability, and thus is likely to induce a higher degree of
cooperation and receive more support from retailers. A study by Besanko, Dubé and
Gupta (2005) ﬁnds that retailers provide more local advertising support to products that
contribute more to the proﬁt of stores. Thus, the extent of the retailer’s collaborative
behavior towards a new product seems dependent on the evaluation of the new product.
We therefore propose the following:

H9: Retailer new product evaluation is positively related to retailer collaborative
behavior including cooperation, communication, and commitment.

The Effect of Retailer Collaboration on New Product Performance

Retailers perform a critical role in the initial sales of a new product (Ottum 1996).
The actions of retailers can have a major effect on the diffusion of a new product
launched in the market (Kuester, Gatignon and Robertson 2000). Collaborative
relationships between manufacturers and retailers are crucial to a successful new product

launch (Bowersox, Closs and Stank 1999). Collaborative relationships require the

41

retailers and manufacturers to participate in strategic co-planning and encourage cross-
organizational coordinating behavior such as coordinating with daily operation, sharing
information, and staying committed to the relationship. Therefore, cooperation,
communication between the retailer and the supplier and mutual commitment to the
relationship are alike manifesto of collaboration between the retailer and the supplier.

Cooperation. Cooperation between the manufacturers and retailers in the
consumer products industry may include collaborative planning, forecasting, and
replenishment. Poor cooperation in these activities between the manufacturer and channel
members can induce new product failure (Kono and Lynn 2007). For example, in many
consumer product industries, manufacturers provide promotional support in various
forms to retailers, including promotional material (e. g. display ﬁxtures), monetary
support (e. g. trade promotion), selling aids in stores, and motivated manufacturer
salespeople (Kincade, Woodard and Park 2002). Promotional activities that support the
launch of a new product play a crucial role in the success of the new product because
they determine to a large extent the diffusion of a new product and drive the sales of the
product (Walters and MacKenzie 1988; Tellis 1998). The absence of promotional support
may lead to the failure of a new product (Delre et al. 2007).

However, not all of the promotion activities provided by manufacturers are
supported by retailers (Walters 1988). It is reported that retailers sometimes do not pass
trade-promotion money provided by manufacturers on to consumers (Besanko, Dubé and
Gupta 2005). Further, retailers are able to inﬂuence the new product performance through
withholding selling effort, which might entail not providing local advertising for a new

product or inﬂuencing the consumer to purchase another product (Huang, Li and

42

Mahajan 2002). Ensuring retailers’ cooperation in the new product introduction process is
therefore critical to new product performance. As an example, Nabisco’s Planters saw its
sales rise by 40 percent as a result of better-planned promotions and joint replenishment
with Wegmans stores (as cited in Lee 2002). Similarly, Useem, Schlosser and Kim
(2003) report how Wal-Mart contributed to the success of Shrek, an animated film by
DreamWorks. The movie was a big hit in summer 2001 and later released on DVD on
November 2001. Wal-Mart’s nation wide promotion—its stores bristling with displays of
the green ogre—helped Shrek become the year's bestselling DVD. All of these examples
demonstrate that retailers’ cooperative actions can lead to new product success.
Communication. Communication includes the ways in which information is
exchanged and shared between partners and the openness between partners in their
exchanges of information (Fontenot and Wilson 1997). Communication is a vital factor
for a good coordination between retailers and suppliers in new product introductions
(Kumar and Phrommathed 2005), especially when high environmental uncertainty and
competition are involved (Frazier 1999). A meta-analysis of 47 empirical studies in the
area of new product performance conducted by Montoya-Weiss and Calantone (1994)
conclude that one determinant of new product performance is the communication and
information exchange between the manufacturer and other ﬁrms in the distribution
channel. Having gathered and analyzed managerial perceptions regarding the key success
factors in launch activities pertaining to new product performance, di Benedetto (1999)
concludes that information-gathering activities even after the new product is launched
into the market are important to new product success. The rationale is that continuous

updating and modification of both marketing and production plans are necessary

43

throughout the product launch and post-launch phases in response to customer and
competitive reactions, and technological or economic environment changes. Therefore,
information about consumers and the effectiveness of the marketing activities undertaken
is important to the ultimate new product performance.

More recently, researchers (e. g. Bowersox, Stank and Daugherty 1999; Calantone,
Benedetto and Stank 2005) illustrate the importance of manufacturers’ using “lean
launch” methods to support new product launch success. Lean launch methods emphasize
the use of a ﬂexible supply chain system to react quickly to customer needs and market
demands. Such a ﬂexible supply chain system requires extensive communication and
information-sharing with retailers. As an illustration, the VF Corporation, a manufacturer
of jeans and lingerie, works with Wal-Mart to develop an artificial intelligent software
program that automatically adjusts the assortment models (as cited in Ganesan 1994). If a
particular Wal-Mart outlet sells, for example, more large sizes than the norm, VF's
software changes the assortment at that particular store accordingly. Without point-of-
sale (POS) information, suppliers cannot monitor sales ﬂuctuations to adjust production.
This may lead to products being either out-of-stock causing lost sales or overstocked,
leading to high inventory costs. Nowadays, modern retailers equipped with advanced
technologies (e. g. ECR system, RFII) technology) possess sophisticated marketing
research abilities that draw on information such as scanner data as well as proprietary
data based on loyalty programs (Fisher, Raman and McClelland 2000). Retailers are also
taking control of their information: they do not pass information automatically to

manufacturers as assumed before (Corstjens and Corstjens 1995). Therefore, the retailer’s

willingness to communicate with the manufacturer regarding the new product through

 

information sharing is an important factor determining the new product performance in
stores.

Commitment. Commitment is an implicit or explicit pledge of the continuity of a
relationship between channel members (Dwyer et al. 1987), and is based on one party’s
desire to maintain a valued relationship with another (Andaleeb 1996). Ganesan (1994)
refers to such a commitment in a relationship as long-term orientation and indicates that
long-term relationships anchored by commitment result in two independent parties
working together for the betterment of both. The more committed partners are to the
relationship, the greater the chance for each ﬁrm to achieve their individual and mutual
goals without the overshadowing risk of engaging in opportunistic behavior (Fontenot
and Wilson 1997). Committed partners are willing to invest in valuable assets speciﬁc to
an exchange relationship and to work together to serve customer needs better and thus
increase mutual proﬁtability (Anderson and Weitz 1992). Retailers who are committed to
suppliers can help new product performance by providing better cooperative advertising,
offer special displays, providing information on market and competitive activity, or
training salespeople to sell such products (Ganesan 1994). For example, RadioShack was
committed to a 10-year agreement with Sprint for receiving the exclusive right to sell
Sprint’s residential telephone products, and a non-exclusive right to sell its wireless
digital PC service. In order to promote the product and create consumer awareness,
RadioShack participated in a multi-million dollar co-op advertising campaign with Sprint
and invested in extensive training for its sale associates on new products and services. As

a result, customers who walked through RadioShack’s door daily were introduced

45

immediately to an area dedicated to Sprint products and Sprint products immediately
received large leads over its competitors’ (Rangan, Moon and Bell 2000).

Based on the above arguments, we posit:
H10: Retailer collaboration with the supplier is positively related to new product

performance.
New Product Performance and Retailer Performance

The effect of new product performance on retailer performance. New products are
important contributors to the growth and proﬁtability for manufacturers (Crawford 1994).
The NP literature has examined the role of new product success in driving the business
performance of manufacturers. However, few studies are dedicated to the role of new
product performance in driving retailers’ business performance. New product success is
crucial to retailers’ performance; otherwise, retailers would not allocate scarce resources
to new products that are known to have high failure rates. Just like successful new
products bring beneﬁts for manufacturers, they also bring beneﬁts to retailers. One study
reports that new products are the most proﬁtable items in the supermarket product mix
(Heeler, Kearney and Mehaffey 1973). Another study conﬁrms that new products made a
substantial contribution to total retail sales; about one sales dollar in eight was due to new
products (Montgomery 1975). The successful market performance of new products helps
to increase customer satisfaction, drive category growth, and build a positive store image.
For example, P&G's introduction of its Scentstories air freshener, which was launched in
June 2004, helped to drive sales of the scent holder category in stores. As such, new

product performance contributes to retailer business performance.

46

Increasingly, studies of business performance are adopting a two dimensional
view of the performance construct to reﬂect its complexity. A firm’s market performance
is differentiated from its ﬁnancial performance (Homburg and Pﬂesser 2000). Business
market performance refers to the relative effectiveness of a firm in market domain,
whereas business ﬁnancial performance indicates the relative effectiveness of a ﬁrm in
ﬁnancial measures. In our study context, retailer market performance is related to such
variables as customer satisfaction, market share, and positive store image (Homburg,
Hoyer and Fassnacht 2002), and retailer ﬁnancial performance is associated with
profitability and sales growth. Consistent with the literature, we posit that new product
performance contributes to both retailer market performance and ﬁnancial performance.
H11: New product performance is positively related to retailer market performance.
H12: New product performance is positively related to retailer ﬁnancial performance.

The effect of market performance on ﬁnancial performance. The literature in
general supports the notion that market performance leads to improved ﬁnancial
performance (Anderson, Fomell and Rust 1997). Anderson and Sullivan (1993) and
Fomell (1992) ﬁnd a positive link between customer satisfaction and ﬁrm proﬁtability.
Similarly, Homburg, Hoyer and Fassnacht (2002) conduct an empirical study of clothing
and furniture retailers in both the US. and Germany and ﬁnd that retailer market
performance leads to higher profitability. Further, empirical studies have shown that
retailer market performance such as customer loyalty increases ﬁrm ﬁnancial
performance through the absence of acquisition costs, lower operating costs, repetitive
buying, referrals, and higher price tolerance (Loveman 1998). Therefore, we posit:

H13: Retailer market performance is positively related to retailer ﬁnancial performance.

47

Control Variable

Buyer age has been found to have a signiﬁcant inﬂuence on new product buying
behavior (Marr and Thomas 1999; Silva and Davis 2002). For example, Silva and Davis
(2002) find younger, less experienced buyers tend to use more “objective” criterion in

their decision-making. Thus buyer age is used as a control variable in the model.

48

CHAPTER 3

METHODS

This chapter is organized into six sections. The first section justiﬁes the context
for this study, and the second section presents background information about retailing in
China. The third section introduces how the survey instrument was developed. The fourth
section discusses measures for the constructs, and the ﬁfth section covers data collection.
The last section describes the sample characteristics.

Study Context

We chose China as a research context to study retail new product evaluation and
new product performance for the following reasons:

1) China is one of the largest developing countries in the world, and its retail
sector is growing rapidly and represents one of the best retail opportunities in the world
(Retail News Letter 2005). With its average economic growth ranked among the highest
in the world in the past two decades, China is becoming one of the most important
markets in the world (Luo, Sivakumar, and Liu 2005). An investigation of Chinese retail
buying behavior is thus an important and worthwhile means for extending our
understanding of retailers and the consumer product market in China.

2) Retail buying behavior in China resembles that of the West (Stemquist and
Chen 2007). Similar to the situations in the US. and the European countries, channel
power also rests with retailers in the Chinese distribution channel. This suggests that
China is an appropriate country to test our model because factors affecting new product
evaluation and new product performance in China are probably consistent with those

found by studies in the West, though there may be some differences in strength.

49

3) The Chinese retail market is a buyers’ market characterized by strong
consumer spending growth (Luk 1998). Chinese ﬁrms are generally active in new
product development (Li and Atuahene-Gima 1999). Despite being plagued by resource
deﬁciency in general, the growth rate of new products reached 14.8% and the firms in
China developed a total of 68,633 new products in 2003 according to the National
Statistics Bureau of China (as cited in Li, Liu and Zhao 2006). NPD in China has been
well studied by researchers (6. g. Atuahene-Gima and Li 2000; Beverland, Ewing and

Matanda 2006; Calantone, Schmidt and Song 1996).

Background Information of Retailing and New Product Development in China

Before 1979, the year when economic reform was initiated, China was a shortage
economy constructed by central planning and political and administrative allocation (Luk
1998), known as fenpei system. Consumer goods were produced as commodities by state-
owned units, distributed through a central, three-tier distribution system to state-owned
stores in city districts or co-operatives in the countryside, and the price and the amount
were controlled by the central government (Luk 1998). Market forces were restrained and
suppressed; retailers and manufacturers needed not, and could not, respond to changing
consumer needs. Because demand exceeded supply to a large degree, retailers and
manufacturers had virtually no competition. As a result, retailing was perceived by the
Chinese government as a nonproductive business activity with no value-added function
(Stemquist 2007).

During 19803, the Chinese government started introducing market competition
(Wing 1996). A dual distribution system was formed under which manufacturers could

produce at their own discretion according to the market needs and sell products to state-

50

owned stores at market price after fulﬁlling the production contract placed by the
government (Luk 1998). Consequently, the undersupply situation was gradually
eliminated. In addition, foreign manufacturing ﬁrms were allowed to set up joint ventures
in China after the economic reform.These joint ventures provided upgraded production
technology and product design. Meanwhile, competition from these joint ventures also
provides impetus for state-owned manufacturers to change from product oriented to
market oriented (Stemquist, Huang and Chen 2003). As a result, increasing competition
over products and quality prompts new product development. Jefferson et al. (2003)
report that new product sales accounted for 24.5 percent of total sales of China’s large
and medium-size manufacturers in 1999. Although there are no actual data available in
China, Ozer and Chen (2006) report that the new product success rate in Hong Kong is
44.91%, which is comparable to that of 59% in the US. (grifﬁn 1997).

Competition over product quality and efﬁcient distribution among manufacturers
also prompted changes in the retail system (Stemquist 1998). The retail sector in China
has been dramatically transformed after the reform and especially after foreign
investment in retailing was approved by the Chinese government. Foreign investment
was allowed ﬁrst in the form of joint ventures in 1992 and then foreign wholly-owned-
subsidiaries in 2004. The anival of foreign retailers provided a major impetus in the
modernization of the retail system of China through “superrnarketization”—replacing old
style traditional retail formats with modern retail formats (Gale and Reardon 2004). The
first supermarket emerged in Guangdong province in 1990, but by 2003, Chinese
supermarkets skyrocketed to approximately 60,000 stores, according to the China Chain

Store and Franchise Association (see Hu et al. 2004). The intense competition forces

51

retailers to develop a differential advantage. A study by Sinclair, Lyer and Anderson
(1998) reports that supermarkets in Shanghai introduce 10% to 30% new products each
year. Between 1999 and 2005, China’s total retail market grew 157 percent in current
terms to USD832 billion, indicating an annual growth rate of 19.68 percent (China
Statistical Yearbook 2006). A recent report ranked China as the most attractive place for
retail expansion (Retail News Letter 2005).

Survey Instrument

A questionnaire was used to collect data in this dissertation. The development of
construct measures in the questionnaire followed Churchill’s (1979) recommendation.
The questionnaire was developed in three stages. In the ﬁrst stage, previous research and
relevant literature were reviewed. Scales in previous studies were collected and
categorized so that well-established scales appropriate to this study could be used. New
measures were developed if there were no existing scales in the literature. Items were
generated based on in-depth face-to-face interviews with 16 Chinese retail buyers from a
previous project (see Stemquist and Chen 2007) and discussions with academic experts in
retailing. The ﬁrst draft of the questionnaire was then reviewed by two US. and two
Chinese academic experts in retailing and changes were made based on their
recommendations.

In the second stage, the questionnaire was translated into Chinese using the back
translation method. A pretest was conducted with 19 retail buyers in Beijing. Four
changes were made based on this pretest including (1) rewording items because of
ambiguous and double-barrel questions, (2) deleting reverse-worded items, (3)

redesigning the order of items and the format of the questionnaire, and (4) ﬁtting the

52

scale of one item to the habit of Chinese usage—changing from “50 to 100” to “60 to
100” because Chinese custom.10 The second draft of questionnaire was thus formed.

In the third stage, a pilot study of the second draft of the questionnaire with a
sample of 56 retail buyers in Beijing was Conducted. The pilot study did not reveal any
major problems, and the subsequent analyses of data demonstrated good construct

validity and reliability.

Measures

Overall, constructs were measured by seven-point Likert-type scales with multiple
items. The unit of analysis is individual retail buyers evaluating a new product.
Consistent with the retailer’s deﬁnition and with previous studies (e.g. McLaughlin and
Rao 1991; White, Troy and Gerlich 2000), a new product was deﬁned as a single stock-
keeping unit (SKU) that had not previously been carried by the retailer. Consistent with
supermarket industry sources and previous studies (e. g. Desiraju 2001), new products are
classiﬁed into four categories: new concept for the category, brand extension, line
extension, and upgrade or replacement items. In this dissertation, we focus on new
consumer products. Table 3.1 provides a summary of constructs and measurement items.

Product innovativeness was measured by ﬁve items selected from Lee and
O’Connor (2003) and Calantone, Chan and Cui (2006). New product advantage consists
of two dimensions: product advantage to consumer and product advantage to retailer.
Three items for new product to consumer were selected from Langerak, Hultink and
Roben (2004) and the other three items were newly developed. Items for new product

advantage to retailer were developed based on ﬁeld notes with retail buyers and

 

'0 In China, scores over 60 is considered “pass”, and scores under 60 is considered “fail”.

53

discussions with industry experts. The resulting scales for product advantage to consumer
and product advantage to retailer contain six items and ﬁve items respectively.

Supplier reputation was measured by four items from Ganesan (1994). Past
experience was operationalized by one item: “How many years have you been doing
business with this supplier?” Personal relationships between buyers and salespersons
was measured by four items based on field notes with retail buyers

Retailer trust of the supplier was measured by seven items from Morgan and Hunt
(1994). Retailer cooperation with supplier was measured by three items from Anderson
and Narus (1999). Retailer communication with supplier was measured by four items
from Anderson and Weitz (1992) Retailer commitment to supplier was measured by ﬁve
items from Anderson and Weitz (1992).

Retail new product evaluation was operationalized by one item asking
respondents to give a score from 0 to 100 to the new product relative to other products
s/he has evaluated in general. Sixty is the pass score indicating that a new product was
accepted in stores and 100 is a perfect score.

New product performance was measured by seven items asking the respondents to
rate how well the new product has performed on each of the performance indicator. Items
were adapted from Langerak, Hultink and Roben (2004). Consistent with prior studies,
we use perceptual measures of new product performance.

Retailer performance was operationalized by asking how well the retailer
performed compared with its competitors over the last year. In particular, retailer market

performance was measured by four items from Homberg, Hoyer and Fassnacht (2002),

54

and retailer ﬁnancial performance was measured by two items from Homberg, Hoyer
and Fassnacht (2002).
Data Collection

Collecting data in an international setting is a trade-off between cost, time,
availability, and response rate. The nonresponse issue is a major concern in cross-cultural
survey research (Couper and De Leeuw 2003). Following Jobber and Saunders’s (1988)
suggestion of collaborating with researchers in foreign countries, we collaborated with
China Retail Research Center (CRRC) for data collection.

Since a national list of retailers in China was not available, the CRRC generated a
list of consumer products retailers in China based on information collected from multiple
sources including the China Chainstore and Franchise Association (CCFA), the National
Bureau of Statistics of China, and business directories. 300 retailers were randomly
chosen from the list. The ﬁrst packet included two questionnaires in Chinese and a letter
from the CRRC with the endorsement of a reputable university in China asking each
president /contact person to forward the questionnaires to the company’s buying
department. The buyers in the buying department were asked to complete the
questionnaires. Two weeks after the mailing, a personal telephone call was made to
conﬁrm that the questionnaires had been forwarded to individual buyers. However, no
complete questionnaires were mailed back even after a second phone call after another
two weeks. Although we anticipated the challenge of collecting data in China (e. g.
Calantone, Schmidt, and Song 1996), we were still surprised by the difﬁculties of
collecting ﬁrm—level data in China. This conﬁrms previous studies about suggesting

extremely low response rates from a large sample of mail survey in China (see Lukas,

55

Tan and Hult 2001). Since the data collection through mail survey failed, we had to ﬁnd
another more effective method. Many studies recommended an interview approach (see
Atuahene-Gima and Li 2002; Li and Atuahene-Gima 1997) for effective data collection.
However, this method is most effective and efﬁcient when collecting data from local or
regional areas. Since our intention was to collect data from a national sample, the cost
and time involved in using this method would be tremendous. Therefore the method
needed to be adjusted. We decided that, rather than we go to where every retailer
corporate office is located across the country, we would go to meet many retailers at one
location. The China Chain Store Expo 2006 provided a good venue for our purposes. The
China Chain Store Expo is an annual convention held by the China Chainstore and
Franchise Association (CCFA), a national organization founded in 1997. The theme of
China Chain Store Expo 2006 was “Retail Innovation”. The convention provided retailers
and suppliers with a platform to meet each other and to learn about new trends, new
products and new technologies in the retail industry (CCFA website). More than one
thousand retail buyers/managers participated in the convention. The CRRC randomly
distributed 300 questionnaires to retail participants and 132 questionnaires were returned.
Two blank questionnaires were discarded, which reduced the number to 130. This
represents a response rate of 44%.

In order to improve the validity of the data collected, three actions were taken.
First, an interview approach was used to collect data. The interviewers ensured that the
respondents were either retail buyers or managers who were responsible for selecting

products. Second, the respondents were asked to focus on a new product they personally

56

evaluated and accepted in their stores the previous year (i.e. 2005)l 1, so that the retailer
had carried the new product for at least nine months (the convention was in October
2006), and the respondents could assess new product performance after entering into the
market. The common trial period for a new product in China is 3 months (Stemquist and
Chen 2007). Third, the respondents were guaranteed of anonymity and conﬁdentiality
and were informed that the results were to be published for research purposes only. This
is important because Chinese managers are more willing and more likely to provide
accurate information under the condition of anonymity (Alder, Campbell, and Laurent
1989).
Sample

We compared means of key variables (i.e. new product evaluation and new
product performance) between data collected at the China Chain Store Expo 2006
convention and data from the pilot study and found no significant differences. Therefore,
both data are pooled. The resulting sample size is 186.

Our sample consists of 186 retail buyers from more than 58 retailers located in 28
cities across China. China administers 33 province-level divisions, including 22
provinces, ﬁve autonomous regions, four municipalities, and two special administrative
regions. Our sample represents 18 provinces, four autonomous regions, three
municipalities, and one special administrative region. Table 3.2 lists the number of

respondents is each division.‘2

 

n Respondents expressed no problem with recalling. Rao and McLaughlin (1989) also reported, in their
study, that the buyers were able to recall vividly the circumstances surrounding the introduction of each
new item.

'2 The total number is 173. There are 13 missing values.

57

Among the 186 respondents, 117 (63%) are male and 69 (37%) are female.
Eighty-seven percent of the respondents’ ages range from 20 to 40. The respondents
represent buying managers, central buyers, store managers, store buyers, and department
managers who are responsible for buying consumer products. The average (median) of
the average monthly sales volume that the respondents are responsible for buying is 10
million RMB (US $1.28 million). The average (median) of the number of stores that the
respondents are responsible for buying for is 5. Respondents had spent an average of 7.7
years working in retail. Table 3.3 provides a summary of the respondents’ characteristics.

The companies that the respondents work for are composed of a variety of
retailers with different ownership types. Among these different ownerships, domestic
private ﬁrms account for half the sample. About 47 percent of our sample are among the
top 100 retailers in China. The average (median) annual sales volume for the retailers
reported is 100 million RMB (US $12.8 million). The percentage of sales contributed by
new products during the previous year reported by the respondents ranges from 2 percent
to 80 percent with an average (mean) of 25 percent. The percentage of proﬁt generated by
new products during the previous year ranges from 4.5 percent to 100 percent with an
average (mean) of 27 percent. Fifty—eight percent of the respondents report that their
stores possess a single store format while thirty nine percent report that their stores
possess multiple formats. Supermarket, hypermarket, and convenience store are among
the most frequently reported by the respondents. The average length of retailers’
respective business relationships with their reported suppliers is about 4 years. In terms of
channel power between the retailer and the supplier, 56 percent of the respondents report

that their companies are more powerful than their respective suppliers, while 10 percent

58

of the respondents report that their companies are less powerful than their respective
suppliers. More than 31 percent of the respondents report that their companies and their
respective suppliers have equal power. Table 3.4 provides a summary of the companies’
characteristics reported by the respondents.

New products reported by the respondents comprise items in food & beverage
(49%), home & personal care (17%), consumer durables (15%), and apparel & footwear
(18%). Among those items, 23 percent are classified as new concept, 27 percent are
classiﬁed as brand extension, 33 percent are classiﬁed as line extension, and 15 percent
are classiﬁed as upgrade or replacement items. Table 3.5 provides a profile of the new

products reported by the respondents.

59

Table 3.1 - Constructs and Measurement Items

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Construct Measurement Items Source
New Product This new product
Advantage ‘0 provides higher value for consumers. New
Consumers h h' h l' /. . .
(1=Strongly as a 1g er qua rty price ratio.
Disagree, will meet an unmet need of consumers.
=Strongly is superior to competing products. Langerak,
Agree) has higher quality than competing products. Euggk and
o n
offers unique features for consumers. (2004) (1:072
New Product This new product New
ﬁgﬁaigfsge to would provide us with a high gross margin.
(1=Strongly would potentially achieve high sales volume.
Disagree, would be in high demand.
1=Strongly would generate high store traffic.
gree) would contribute incremental sales; not
cannibalize.
You are asked to give a score to every product (including new
New Product products and non-new products) you have evaluated. Only scores
Evaluation above 60 out of 100 will be accepted in stores (0 is the lowest; 100 is
a perfect score). What score would you give to this new product
relative to other products you had evaluated in general at the time
when you accepted it.
From 0 to 100, I would give this new product a score New
of .
Product Please indicate your extent of agreement at the time when you
Innovativeness accepted the item ......
(1=Strongly Overall, this new product was highly innovative Calantone,
Disagree, compared to other products in the market. Chan and Cui
7A=Strongly This new product was radically different from (300,6)
gree) competitor products. (LN A
The technology this product incorporates was new to Lee and
consumers. O’Connor
This product offered dramatic improvements in (2003)
existing product features. (1:0.84
This product introduced many completely new
features to the market.
Buyer Age Your age?

 

 

 

60

 

Table 3.1 (Cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

concerning their marketing efforts.

 

 

This supplier is willing to let us know their
weakness as well as their strengths.

 

Construct Measurement Items Source and
its Reliability

Reputation Please indicate your extent of agreement at the time when you
(1=Strongly accepted the item ......
Disagree, This supplier has a reputation for being honest. Ganesan
7=Strongly (1994)
Agree) This supplier has a reputation for being concerned 0:032

about retailers.

This supplier has a good reputation in the market.

Most retailers think that this supplier has a

reputation for being fair.
Personal Please describe the relationship you have with the salesperson from
Relationship this supplier gt the time when you accepted the item ......
(l=Strongly I have a very close personal relationship with the New
Disagree, salesperson.
7A=Strt;ngly This salesperson is like a friend to me.

ree
g I am willing to help this salesperson.

I am familiar with this salesperson.
Trust Please indicate your extent of agreement at the time when you
(1=Strong1y accepted the item ......
Disagree, We can count on this supplier to do what is right. Morgan and
7:5”01181)’ The supplier can always be trusted. Hunt (1994)
Agree) This supplier has high integrity. (1:095

This supplier is honest and trustful with me.

I feel that I can trust this supplier completely.

This supplier is truly sincere in their promise.

This supplier treats me fairly and justly.
Cooperation Please indicate your extent of agreement ......
S;Strongly Our company tries our best to help out this supplier. Anderson and
7=lSSat§6::ly This supplier tries their best to help our company 5:121]: (1990)
Agree) 0‘": . . .

Thrs supplier rs very cooperative.
Communication This supplier keeps us well informed about what is Anderson and
(1=Strongly going on in this distributorship and with customers. Weitz (1992)
Disagree, We are quite involved in the marketing and planning (1:0.84
7=Strongly efforts of this supplier.
Agree) This supplier seeks our advice and counsel

 

61

 

Table 3.1 (Cont’d)

 

 

 

 

 

 

 

 

 

 

 

Attaining desired market share.

 

Construct Measurement Items Source

Retailer Compared with your competitors, how well has your company

Financial performed over the last year?

Performance -

(1=Much Attaining desired sales growth. Homberg,

Worse, Hoyer and

7=MUCh Better) Attaining desired proﬁtability Fassnacht
(2002) (1:087

Retailer Market Compared with your competitors, how well has your company

Performance performed over the last year?

(1=Much Achieving customer satisfaction. Homberg,

Worse, Hoyer and

7=Much Better) Attracting new customers. Fassnacht
(2002)

Building a positive store image. a=0.87

 

Commitment
(1=Strongly
Disagree,
7=Strongly
Agree)

Please indicate your extent of agreement ......

 

We are committed to this supplier.

 

We are quite willing to make long-term investment
in selling this product.

 

Our relationship with this supplier is a long-term
alliance.

 

We are patient with this supplier when they make
Mistakes that cause us trouble.

 

We are willing to dedicate whatever people and
resources it takes to grow sales of this product.

 

Anderson and
Weitz (1992)
(1:083

 

New Product
Performance
(1=Very Poor,
7=Very Good)

How well has this product performed on each of the performance

 

 

 

 

 

 

 

indicators?

Met volume goals. Langerak,
Hultink and

Met sales growth goals. Robben

Met market share goals. (2004)

Met profitability goals. a=0.88

Consumer satisfaction.

Consumer acceptance.

Met margin contribution goals. New

 

Past Experience

 

 

How long have you been doing business with this
supplier?

 

 

62

 

 

Table 3.2 — A Proﬁle of the Respondents’ Locations

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Provinces Number of Autonomous Number of
Respondents Reﬁms Respondents

Auhui 3 Guangxi 2

Fujian , 0 Inner Mongolia 3

Gansu l N ingxia l

Guangdong 4 Xinjiang 2

Guizhou 1 Tibert 0

Hainan 0

Hebei l 4 Municipalities

Heilongjiang 4 Beijing 77

Henan 2 Chongqing 0

Hubei 1 Shanghai 12

Hunan l Tianjin l

Jiangsu 6

Ji an gxi 2 Special Administrative Regions

Jilin 2 Hong Kong 2

Liaoning 2 Macau 0

Qinghai 0

Shannxi 2

Shanxi 9

Shandong 13

Sichuan 0

Yunnan 2

Zhejiang 4

 

 

 

 

63

 

Table 3.3 - Characteristics of the Respondents

 

 

Respondents Percentage or Range
Job Titles
Buying manager 36%
Central buyer ' 9%
Store manager 14%
Store buyer 3%
Department manager 24%
Others 13%
Sex Male: 63%; Female: 37%
Age 20-30: 41%
30-40: 46%
40-50: 1 1%
> 50: 2%
Years in retailing Less than 1 year-28 years

(Mean: 7.73; SD: 5.80)

Years personally doing business with the supplier Less than 1 year—30 years
(Mean: 4; SD: 3.29)

Years personally doing business with the salesperson Less than 1 year-17 years
(Mean: 2.13; SD: 2.20)

Average monthly sales volume responsible for buying RMB LOGO—1,500,000,000
(US$ 0.001—192.31 million)
(Mean: $12.85 million;
SD: $33.53 million)

Number of stores responsible for buying for 1 — 1300
(Mean: 36; SD: 126)

 

Note: Totals may not equal 100% because of rounding or missing values.

64

Table 3.4 - Characteristics of Companies as Reported by the Respondents

 

 

Retailers Percentage or Range
Ownership type

State-owned Enterprise 11%

Privatized state-owned 8%

Joint Venture with FDI 11%

Domestic Private Firm 50%

Foreign-owned Subsidiary 13%

Others 7%

Sales volume

Sales accounted for by new products
Proﬁt accounted for by new products

Store format reported by respondentsa

Company ranking on the Top 100 retailers List

1—30
31—100
> 100

Years doing business with the reported supplier

Retailer’s channel power compared
to the reported supplier

RMB 10000—30,000,000,000
(US $ 0.001-3846 million)
(Mean: 87.83 million;

SD: 341.05 million)

2%-80%
4.5%—100%

Hyperrnarket: 73
Supermarket: 105
Department Store: 25
Warehouse Club: 17
Specialty Store: 22
Convenience Store: 61
Others: 9

27%
30%
43%

Less than 1 year—l6 years
(Mean: 3.94; SD: 3.18)

More powerful: 56%
Equally powerful: 31%
Less powerful: 10%

 

Note: Totals may not equal 100% because of rounding or missing values.
a The total exceeds 186 because respondents reported multiple formats.

65

Table 3.5 — A Profile of New Products Reported

 

New Products Percentage or Range
Product category
Food and beverage 49%
Home and personal care 17%
Consumer durables 15%
Apparel and footwear 18%
Classification
New concept 23%
Brand extension 27%
Line extension 33%
Upgrade or replacement item 15%

 

Note: Totals may not equal 100% because of rounding or missing values.

66

CHAPTER 4
ANALYSES AND RESULTS
Initial Data Analysis

First, we examined the data to ensure quality. We checked missing data for each
variable and found that less than 5 percent of the data are missing. A further investigation
using missing value analysis in SPSS 14.0 suggests that the missing data are completely
random. Thus, a replacement of missing data is appropriate (Hair et al. 1998). Missing
values are replaced following the procedure in Wilson and Collier (2000). The
measurement items are grouped in constructs based on the model. If one question item is
not answered, the average score of the answered questions of the scale is used to replace
the missing value.

Second, we checked the normality of the data. The means, standard deviations,
kurtosis and skewness of each item are computed. The kurtosis and the skewness of
almost all items are below 2, falling within the speciﬁed range of univariate normality.
Only one item, which asks the respondents about the number of years they have been
conducting business with the supplier, has a skewness of 4.38 and a kurtosis of 28.7.
After a square root transformation, the skewness of this item reduces to 1.38 and the
kurtosis to 4.78, both below 5. Therefore, the univariate normality of the data is
acceptable.

Third, in order to increase measurement quality and purify the measures, we
conducted an Exploratory Factor Analysis (EFA) for all the items in the hypothesized
model. Items with high cross-loadings and/or low factor loadings were dropped if they do

not contribute to the construct's explanation or provide added insight into the domain of

67

interest. As a result, 5 items were deleted. Cronbach’s alpha was computed for each
construct to assess the reliability of existing construct measurements. Alpha values for all
the constructs in the hypothesized model ranged from 0.72 to 0.89 and are thus well
above the acceptable minimum standard of .60 (Nunnally 1967). See Table 4.1 for each
construct’s reliability and item information.

Fourth, common method variance was tested since our study involved survey data
collection using the single-informant method. If information on the dependent and
independent variables is collected from the same respondents, common method variance
may pose a serious problem. The respondents in our study were asked to answer both the
questions about the situations when new products were accepted and the survey questions
regarding the performance of new products. Thus, the potential problem of common
method variance may occur. Following Podsakoff and Organ (1986), we conducted
Harman’s one-factor test. If common method bias exists in our data, a single factor
should emerge or one general factor should account for most of the variance. We
conducted a principal components factor analysis of all measures, and the result yielded
13 factors with eigenvalues greater than 1.0, with total explained variance of 73 percent.
The ﬁrst factor accounted for only 26 percent of the variance. Further, we used SEM to
test the effect of common method bias among measures. Since our sample size does not
allow us to conduct a full test with all measures, we broke measures into three groups and
conduct a conﬁrmatory factor analysis (CFA) separately. The results showed an
extremely bad ﬁt in all three groups. This suggests that the common method variance
may not be a serious problem in the sample. Next, we use CFA to assess convergent

validity and discriminant validity of measurement scales.

68

Table 4.1 - Construct Reliability and Item Information

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Construct Measurement Items Mean SD.
New Product V1: Provides higher value for consumers. 5.07 1.34
Advantage to V2: Has a higher quality/price ratio. 4.85 1.41
Consumers
a = 0.81 V3: Will meet an unmet need of consumers. 5.02 1.48
V4: ls superior to competing products. 5.10 1.29
V5: Has higher quality than competing products. 5.31 1.50
V6: Offers unique features for consumers. (D) Deleted
New Product V1: Would provide us with a high gross margin. 5.24 1.36
Advantage to V2: Would potentially achieve high sales volume. 5.23 1.24
Retailers
(1 = 0,80 V3: Would be in high demand. 4.98 1.18
V4: Would generate high store trafﬁc. 4.72 1.37
V5: Would contribute incremental sales; not 5.14 1.23
cannibalize.
Product V1: Overall, this new product was highly innovative 4.89 1.31
Innovativeness compared to other products in the market.
a = 0.89 V2: This new product was radically different from 4.48 1.54
competitor products.
V3: The technology this product incorporates was new 4.61 1.54
to consumers.
V4: This product offered dramatic improvements in 4.61 1.46
existing product features.
V5: This product introduced many completely new 4.67 1.49
features to the market.
Commitment V1: We are committed to this supplier. 4.06 1.42
a = 0'85 V2: We are quite willing to make long-term 4.40 1.34
investment in selling this product.
V3: Our relationship with this supplier is a long-term 4.71 1.29
alliance.
V4: We are patient with this supplier when they make 4.25 1.32
mistakes that cause us trouble.
V5: We are willing to dedicate whatever people and 4.27 1.41
resources it takes to grow sales of this product.
New Product V1: Met volume goals. 4.72 1.02
Zerfgrégance V2: Met sales growth goals. 4.73 1.13
V3: Met market share goals. 4.69 1.07
V4: Met proﬁtability goals. 4.82 1.11
V5: Met margin contribution goals. 4.88 1.14
V6: Consumer acceptance. 5.14 1.17
V7: Consumer satisfaction. (D) Deleted

 

 

 

 

69

 

Table 4.1 (Cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Construct Measurement Items Mean SD.
Reputation V1: This supplier has a reputation for being honest. 5.06 1.32
a = 0'87 V2: This supplier has a reputation for being concerned 4.98 1.28
about retailers. I
V3: This supplier has a good reputation in the market. 5.10 1.21
V4: Most retailers think that this supplier has a 5.06 1.15
reputation for being fair.
Personal V1: 1 have a very close personal relationship with the 3.6 1.63
Relationship salesperson.
a = 0.86 V2: This salesperson is like a friend to me. 4.04 1.80
V3: I am willing to help this salesperson. 4.16 1.65
V4: 1 am familiar with this salesperson. 4.09 1.72
Trust V1: We can count on this supplier to do what is right. 4.90 1.26
a = 0'89 V2: The supplier can always be trusted. 4.81 1.03
V3: This supplier has high integrity. 4.87 1.15
V4: This supplier is honest and trustful with me. 4.91 1.18
V5: I feel that I can trust this supplier completely. 4.68 1.28
V6: This supplier is truly sincere in their promise. (D) Deleted
V7: This supplier treats me fairly and justly. (D) Deleted
Cooperation V1: Our company tries our best to help out this 4.12 1.33
a = 0.72 supplier.
V2: This supplier tries their best to help our company 4.70 1.24
out.
V3: This supplier is very cooperative. 4.96 1.10
Communication V 1: This supplier keeps us well informed about what is Deleted
a = 0.76 going on in this distributorship and with customers. (D)
V2: We are quite involved in the marketing and 4.29 1.42
planning efforts of this supplier.
V3: This supplier seeks our advice and counsel 4.60 1.38
concerning their marketing efforts.
V4: This supplier is willing to let us know their 4.38 1.62
weakness as well as their strengths.
Retailer Market V1: Achieving customer satisfaction. 5.26 1.13
Performance ,
a = 0.83 V2: Attracting new customers. 5.27 1.17
V3: Building a positive store image. 5.55 1.16
V4: Attaining desired market share. (D) Deleted
Retailer V1: Attaining desired sales growth. 4.74 1.18
Financial
Performance V2: Attaining desired proﬁtability. 4.75 1.19
a = 0.74

 

70

 

Measurement Validation

Because the number of items is large, fitting all items in a conﬁrmatory factor
model is not feasible. To assess convergent validity and discriminant validity of our
constructs, we follow Hunter and Perreault’s (2007) suggestion and divide our model into
several subsystems (e. g. trust and antecedents), and then use CFA to test items across sets
of scales that constitute the proposed subsystems in the model. As mentioned in Chapter
2, we propose new product advantage as a second-order factor construct with two sub-
dimensions: product advantage to consumers and new product advantage to retailers.
Before we assess the CFA measurement model involving new product advantage, we first
conduct a CFA analysis for this construct alone to assure factorial validity (Byme 2001).
CFA of New Product Advantage

Before conducting a CFA analysis, an EFA is conducted to examine the factor
structure of the construct of new product advantage. Two components are extracted with
eigenvalues larger than 1.0. The rotated component matrix shows that items of new
product advantage to consumers and new product advantage to retailers loaded on the
proposed respective components (see Table 4.2).

Two ﬁrst-order factor CFAs are first conducted. Mardia’s coefﬁcient (Mardia
1970) is used to measure multivariate normality. Mardia's normalized estimate of 12.67 is
above the minimum cutoff point of the multivariate normality of 1.96 (Byme 2001),
indicating multivariate non-normality of the original data at a one-tail .05 level.
Therefore, a robust transformation is used to reduce kurtosis and skewness. The

maximum likelihood method with robust transformation (ML ROBUST) is used and the

71

robust statistics are reported next. We report fit indices such as CFI and RMSEA because
they are least sensitive to sample size (Fan, Thompson, and Wang 1999).

LM statistics are examined for cross-loading items. No cross-loading items are
detected. The model fit is good (S-B x2: 59.43, p: .004, (if = 34, CFI= .952; [Hz .953;
NNFI= .905; RMSEA: .065; 90% CI (.036, .092)). See Figure 4.1 for the CFA
measurement model of new product advantage. Although two items of new product
advantage to retailers (i.e. V1: this new product would provide us with a high gross
margin; V5: this new product would contribute to incremental sales; not cannibalize)
have item-factor loadings of less than .60 (i.e. V1=.57; V5=.59), these items are retained
because Gerbing, Hamilton and Freeman (1994) argue that removing psychometrically
acceptable items may weaken content validity and narrow the domain of the construct.
We contend that in a new product buying setting, a new product proving a high gross
margin and/or not cannibalizing are product attributes that are perceived important by
retailers. Thus, these two items are kept to reﬂect the domain of new product advantage
to retailers.

Convergent validity is assessed using three criteria based on the results of CFA
tests: individual item lambda coefficients greater than .5, a signiﬁcant (.05 level) t-value
for each path, and the loading of each path with greater than twice its standard error
(Anderson and Gerbing, 1988). The CFA results show that all the item-factor loadings
are larger than .57 (see Figure 4.1) with all t-values are greater than 5.59 (p< .01), and
each path loading is greater than twice its standard error. Therefore, convergent validity

of product advantage to consumers and product advantage to retailers is conﬁrmed.

72

Discriminant validity between product advantage to consumers and product

advantage to retailers is examined by ﬁxing the covariance between two latent factors at

1.00. The result shows that the x2 of the unconstrained model is signiﬁcantly lower (p<

.001 for x2 difference) than that of the constrained model, supporting the discriminant

validity between new product advantage to consumers and product advantage to retailers.

Table 4.2 - EFA Analysis of New Product Advantage

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Component Item Factor Loading
Product Advantage V1: Provides higher value for 0.51 0.36
to Consumers consumers.

V2: Has a higher quality/price 0.76 0.30
ratio.
V3: Will meet an unmet need of 0.72 0.33
consumers.
V4: Is superior to competing 0.76 0.19
products.
V5: Has higher quality than 0.66 0.07
competing products.
Product Advantage V1: Would provide us with a 0.38 0.51
to Retailers high gross margin.
V2: Would potentially achieve 0.20 0.82
high sales volume.
V3: Would be in high demand. 0.24 0.77
V4: Would generate high store 0.33 0.68
trafﬁc.
V5: Would contribute to 0.11 0.71
incremental sales; not cannibalize.

 

 

The second-order factor CFA is then conducted and results in the same model ﬁt

(S-B x2 = 59.43, p: .004, df = 34, CFI= .952; IFI= .953; NNFI= .905; RMSEA: .065;

90% CI (.036, .092)). Thus, the evidence conﬁrms the second-order factorial structure of

73

 

the new product advantage construct with two sub-dimensions: product advantage to
consumers and product advantage to retailers. However, in order to test the potentially
differing impacts of these two dimensions, new product advantage to consumers and new
product advantage to retailers are retained as separate constructs in the subsequent

measurement model assessment and model testing.

Figure 4.1 — CF A Measurement Model of New Product Advantage

 
   
   

New Product
Advantage

0.99 0.77

 
  
     

New Product

  

New Product

Advantage to Advantage to
Consumers Retailers
0.62 .61 0.76 0.75 0.6 0.57 0.77 0.7 0.71 0.5'

 

 

 

 

 

 

 

 

 

 

lll Till

S-B x2 = 59.43, p = .004, df = 34, CFI= .952; IFI= .953; RMSEA: .065

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

CFA of Product Innovativeness and New Product Advantage
Mardia's normalized estimate is 16.02, thus the robust statistics are reported. The

model fit is good (S-B x2 = 118.28, p: .0008, (if = 74, CFI= .954; IFI= .954; NNFI=

74

.943; RMSEA: .060; 90% CI (.039, .080)). See Table 4.3 for standardized loadings of

the CFA measurement model. N o cross-loading items are detected.

Table 4.3 — CFA Results of Product Innovativeness and New Product Advantage

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

new features to the market.

 

 

Construct Measurement Items Standar t-
-dized value
Loading
F 1: New V1: Provides higher value for consumers. 0.60 6.29
Product V2: Has a higher quality/price ratio. 0.60 6.04
Advantage to _
Consumers V3: erl meet an unmet need of consumers. 0.78 7.31
V4: Is superior to competing products. 0.74 7.06
V5: Has higher quality than competing 0.67 6.65
products.
F2: New V1: Would provide us with a high gross 0.56 5.34
Product V2: Would potentially achieve high sales 0.79 6.82
Advantage to volume.
Retailers V3: Would be in high demand. 0.75 6.67
V4: Would generate high store trafﬁc. 0.70 6.42
V5: Would contribute incremental sales; not 0.58 5.71
cannibalize.
F3: Product V1: Overall, this new product was highly 0.72 8.12
Innovativeness innovative compared to other products in the
market.
V2: This new product was radically different 0.73 8.93
from competitor products.
V3: The technology this product incorporates 0.74 8.94
was new to consumers.
V4: This product offered dramatic 0.84 10.01
improvements in existing product features.
V5: This product introduced many completely 0.83 9.96

 

Convergent validity of each of the three constructs is conﬁrmed because all the

item-factor loadings are signiﬁcant (all t-values are greater than 5.34, p< .01) and larger

than .56. Additionally, all path loadings are greater than twice their respective standard

errors. Discriminant validity among product innovativeness, product advantage to

75

 

consumers, and product advantage to retailers is assessed by conducting a series of nested
conﬁrmatory analyses (Bagozzi and Philips 1982) ﬁxing all possible combinations of
covariances among these three latent variables (phi matrix) at 1.00, and then employing a
x2 difference test on the values in the unconstrained model vs. each of the constrained
models respectively. In all cases, the x2 of the unconstrained model is significantly lower
(p< .001 for x2 difference) than that of the constrained model, supporting the discriminant
validity of product innovativeness, product advantage to consumers, and product
advantage to retailers.

CFA of Trust, Reputation, and Personal Relationships

Trust and its antecedents are one subsystem in our model. Since past experience is
a one-item measure, this variable is not included in the CFA model. Similarly, Mardia's
normalized estimate of 12.97 is above the cutoff point. The model ﬁt is also good (S-B
x2 = 104.41, p: .0006, df = 62, CFI= .961; [F]: .961; NNFI= .951; RMSEA: .063; 90%
CI (.041, .083)). See Table 4.4 for standardized loadings of the CFA measurement model.
No cross-loading items are detected.

All the item-factor loadings are larger than .69 and signiﬁcant (all t-values are
greater than 6.45, p< .01). Each path loading is signiﬁcantly greater than twice its
standard error. Thus, convergent validity of each of these three constructs is conﬁrmed.

Discriminant validity among trust, supplier reputation, and personal relationships
is also assessed by conducting a series of nested conﬁrmatory analyses fixing one, two,
and all three of the covariances among these three latent variables at 1.00. Again, in all
cases, the x2 of the unconstrained model is signiﬁcantly lower (p< .01 for x2 difference)

than that of the constrained model, supporting the discriminant validity among supplier

76

reputation, retailer trust of the supplier, and personal relationships. The discriminant
validity between past experience and each of these three constructs is examined by
checking the correlations among item variables. In all cases, the correlation between past
experience and each of items is lower than correlations among items that belong to the

same construct. Therefore, discriminant validity of past experience is conﬁrmed as well.

Table 4.4 - CFA Results of Trust, Reputation, and Personal Relationship

 

 

 

 

 

Construct Measurement Items Standar- t-
dized value
Loadin
F 1: V1: This supplier has a reputation for being 0.69 6.46
Reputation honest.
V2: This supplier has a reputation for being 0.84 9.09
concerned about retailers.
V3: This supplier has a good reputation in the 0.88 9.54
market.
V4: Most retailers think that this supplier has a 0.70 8.75

reputation for being fair.

 

F2: Personal V1: 1 have a very close personal relationship with 0.73 8.89
Relationship the salesperson.

 

 

 

 

 

 

V2: This salesperson is like a friend to me. 0.69 6.45

V3: 1 am willing to help this salesperson. 0.92 7.75

V4: I am familiar with this salesperson. 0.85 7.25
F3: Trust V1: We can count on this supplier to do what is 0.69 8.01

ri ht.

Vg2: The supplier can always be trusted. 0.81 12.43

V3: This supplier has high integrity. 0.90 11.64

 

V4: This supplier is honest and trustful with me. 0.76 8.27

 

V5: I feel that I can trust this supplier completely. 0.81 10.14

 

 

 

 

 

 

CFA of Cooperation, Communication, and Commitment
Cooperation, communication, and commitment are proposed as three components

of a second-order factor—collaboration. The measurement scales of three ﬁrst-order

77

factors are first assessed. Again, Mardia's normalized estimate is 12.51, indicating
multivariate non-normality. The model ﬁt is relatively good (S-B x2 = 77.21, p: .0005,
df = 41, CFI= .932; IFI= .934; NNFI= .909; RMSEA: .072; 90% CI (.047, .096)). See
Table 4.5 for standardized loadings of the CFA measurement model. Again, the LM test

does not detect any cross-loading items.

Table 4.5 - CFA Results of Cooperation, Communication, and Commitment

 

 

 

 

 

Construct Measurement Items Standar- t-
dized value
Loading
F1: V1: We are committed to this supplier. 0.72 7.21
Commitment V2: We are quite willing to make long-term 0.75 8.99
investment in selling this product.
V3: Our relationship with this supplier is a 0.71 7.18
long-term alliance.
V4: We are patient with this supplier when 0.76 8.91

they make mistakes that cause us trouble.

 

V5: We are willing to dedicate whatever people 0.71 8.59
and resources it takes to grow sales of this

 

 

 

 

product.
F2: V1: Our company tries our best to help out this 0.63 6.54
Cooperation supplier.

V2: This supplier tries their best to help our 0.73 7.06

company out.

V3: This supplier is very cooperative. 0.74 6.29
F3: V1: We are quite involved in the marketing and 0.75 8.32

Communication planning efforts of this supplier.

 

V2: This supplier seeks our advice and counsel 0.86 9.63
concerning their marketing efforts.

 

V3: This supplier is willing to let us know their 0.65 8.08
weakness as well as their strengths.

 

 

 

 

 

 

Similarly, convergent validity of each construct is conﬁrmed because all the item-
factor loadings are larger than .63, signiﬁcant (all t-values are greater than 6.29, p< .01),

and each path loading is significantly greater than twice its standard error.

78

Discriminant validity among cooperation, communication, and commitment is
also assessed by conducting a series of nested conﬁrmatory analyses ﬁxing one, two, and
all three of the covariances among these three latent variables at 1.00. In all cases, the x2
of the unconstrained model is signiﬁcantly lower (p< .01 for x2 difference) than that of
the constrained model, supporting the discriminant validity of the constructs.

The second-order factor of collaboration is also assessed. The results show that
cooperation, communication, and commitment all load on collaboration with signiﬁcant
factor loadings larger than 0.6. In the next model and hypotheses testing, three scale
scores (i.e. the average of the individual items) of cooperation, communication, and
commitment are used as three indicators of the collaboration construct.

CF A of New Product Performance and Retailer Performance

New product performance and two types of retailer performance—retailer market
performance and retailer ﬁnancial performance—are the ﬁnal subsystem in our model.
Likewise, the multivariate non-normality is indicated by the Mardia's normalized
estimate of 10.57. The results from the CFA model show that the model ﬁt is also
acceptable (S-B x2 = 94.76, p: .000, df = 41, CFI= .922; IFI= .924; NNFI= .895;
RMSEA: .087; 90% CI (.064, .109)). See Table 4.6 for standardized loadings of the CFA
measurement model. Again, no cross-loading items are detected.

Convergent validity of these three constructs is also demonstrated by large and
signiﬁcant item-factor loadings (all item-factor loadings are larger than .61; all t-values
are greater than 7.21 (p< .001); each path loading is considerably greater than twice its
standard error). Discriminant validity among new product performance, retailer market

performance, and retailer ﬁnancial performance is also assessed by conducting a series of

79

nested conﬁrmatory analyses ﬁxing one, two, and all three of the possible covariances
among these three latent variables at 1.00. In all cases, the x2 of the unconstrained model
is signiﬁcantly lower (p< .01 for x2 difference) than that of the constrained model,
supporting the discriminant validity of these three constructs.

Overall, the constructs in our model show good measurement properties.

Table 4.6 - CFA Results of New Product Performance and Retailer Performance

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Construct Measurement Items Standar- t-
dized value
Loading

F 1: Retailer V1: Achieving customer satisfaction. 0.81 11.11

Pigbfrtnance V2: Attracting new customers. 0.80 10.63
V3: Building a positive store image. 0.79 9.95

F2: Retailer V1: Attaining desired sales growth. 0.77 7.21

Financial V2' A . . d . d f' b'l' 0 78 7 66

Performance . ttarnrng esrre pro 1ta 1 rty. . .

F3: New V1: Met volume goals. 0.86 15.01

Product

Performance V2. Met sales growth goals. 0.85 15.26
V3: Met market share goals. 0.77 12.45
V4: Met proﬁtability goals. 0.61 7.78
V5: Met margin contribution goals. 0.64 7.50
V6: Consumer acceptance. 0.67 10.33

The Path Model

Our model is tested using the structural equation model program (EQS 6.1;
Bentler 1995). Due to the small sample size to parameter ratio, testing of the overall
structural model with full scale items as indicators of constructs is not possible.

Therefore, we test the fit of the path model and the statistical signiﬁcance of the

80

hypothesized effects by using the average of the scale items for the latent constructs. In
the model testing, product advantage to consumers and product advantage to retailers are
retained as separate constructs using the average of their respective items. Collaboration
is treated as a single construct using the average of cooperation, communication, and
commitment.

Since the multivariate normality of the data with Mardia's normalized estimate of
4.09 (Mardia 1970) is 16.95, again above the minimum cutoff point of multivariate
normality of 1.96 (Byme 2001), we continue with using the maximum likelihood method
with robust transformation (ML ROBUST) for path model testing and report the robust
statistics. Covariance matrixes for the path model are reported in Table 4.7. Hypotheses
are tested based on the standardized ML estimates using t-tests. Test results are reported
in Table 4.8.
Results

The initial model fit is not satisfactory (S-B x2 = 534.13, p= .000, df = 57, CFI=
.776; [Hz .791; MFI= .693; RMSEA: .114; 90% CI (.094, .133)). Although all
hypothesized paths are significant, the results of the LM test show that several additional
links need to be added to the model. We explicitly considered each suggested link and
only added a link when it could be justiﬁed by theoretical arguments from the literature.
After adding four suggested links, we revised the model as shown in Figure 4.2. We

retest the model and the overall model ﬁt is good (S-B x2 = 77.27, p= .006, df = 49, CFI=

.934; 1FI= .939; MP1: .910; RMSEA: .062; 90% CI (.034, .088)).'3 Although the x2 is

 

'3 We acknowledge that the revised model should be subject to further model validation from another
sample. We include this in the Limitations and Directions for Future Research section.

81

significant, CFI is larger than .90, and RMSEA is smaller than .08. Next, we report the
result of each hypothesis testing.

H1 posited the relationship between product innovativeness and new product
advantage. Specifically, Hla proposed a positive relationship between product
innovativeness and product advantage to consumers, and Hlb proposed a positive
relationship between product innovativeness and product advantage to retailers. The
results of the t-test support Hla (t = 6.24, p < .01), but not Hlb (t = 1.26, p > .1).

H2 posited the relationship between new product advantage and new product
evaluation. H23 proposed a positive relationship between product advantage to
consumers and new product evaluation and H2b proposed a positive relationship between
product advantage to retailers and new product evaluation. The results of the t-test
support both H2a (t = 2.09, p < .05) and H2b (t = 2.01, p < .05).

H3 posited the relationship between new product advantage and new product
performance. Speciﬁcally, H3a proposed a positive relationship between product
advantage to consumers and new product performance and H3b proposed a positive
relationship between product advantage to retailers and new product performance.
Although the result of the t-test does not support H3a (t = .28, p > .1), it supports H3b (t =
2.79, p < .01).

H4 proposed a positive relationship between supplier reputation and retailer trust
of the supplier. The result of the t-test supports this hypothesis (t = 3.23, p < .01). H5
proposed a positive relationship between a retailer’s past experience with a supplier and
retailer trust of that supplier. The result of the t-test supports this hypothesis (t = 2.53, p <

.05). H6 proposed the positive relationships between personal relationships (between a

82

retail buyer and a supplier salesperson) and retailer trust of the supplier. The result of the
t-test does not support this hypothesis (t = 1.85, p > .1).

H7 hypothesized a positive relationship between retailer trust of a supplier and
retailer new product evaluation. The reSult of the t-test supports this hypothesis (t = 2.02,
p < .05). H8 hypothesized the positive relationship between retailer trust of a supplier and
retailer collaboration with that supplier. The result of the t-test supports this hypothesis (t
= 6.00, p < .01).

H9 hypothesized the positive relationship between new product evaluation and
retailer collaboration with the supplier. The result supports H9 (t = 2.70, p < .01). H10
hypothesized the positive relationship between retailer collaboration with a supplier and
new product performance. The result of the t-test supports this hypothesis (t = 4.05, p <
.01).

H11 conjectured the positive relationship between new product performance and
retailer market performance. The result of the t-test supports this hypothesis (t = 4.60, p <
.01). H12 conjectured the positive relationship between new product performance and
retailer ﬁnancial performance. The result of the t-test supports this hypothesis (t = 3.50, p
< .01). H13 conjectured the positive relationship between retailer market performance
and retailer ﬁnancial performance. The result of the t-test supports this hypothesis (t =
4.60, p < .01).

The four added links are as follows:

1) The link between product innovativeness and retailer collaboration with

the supplier (t = 3.85, p < .01).

83

2) The link between product innovativeness and retailer market
performance (t = 3.06, p < .01).

3) The link between product advantage to consumers and product
advantage to retailers ‘(t = 6.77, p < .01).

4) The link between personal relationships (between a retail buyer and a
supplier) and retailer collaboration with that supplier (t = 3.21, p < .01).

Next we will discuss the revised model in light of all hypothesized and newly

suggested links and provide relevant theoretical and managerial implications.

84

cozﬂ>oﬂ pancaam
an umEHHDMH aﬁcmcocﬂom Ecoﬂomummm ”conﬁsmoMHmm—M ”cosﬁonmzoundeuo $853355: “angina EBEBM

9 owS§>u< Hoseoumnmaqd ”£08350 9 omﬁemzod. gunpoinuaam ”ooggotom Hoe—SE uozﬁomnndz ”oocapcotom
ESSEE .EEBMMAE noocmgotom 8355 BoZummZ Moonstomxm ummmumm ”noses—gm 823.5 262ng Home. uozsmneam “Boz

 

mm. $4. 9.. mv. 0N. Fm. wN. n F. VN. 5. mo. CNN 0 _.. Em. mmé DE
vm. 9.... mm. 0N. MN. 0 F. mo. m _.. 5. F 00. w 0N. ONV. _. \tmd mm;
mm. 3. Nm. Nm. Nm. mm. 5. cw. ..mN mo. mvo. _. mod mm"—
mn. 5N. Nm. «.0. Nm. Nm. mm. va v _.. mum. mvé AGO
mm. 9». mN. mm. N0. m ...m o P. NON. F 86 3
mm. mm. NN. Nm. mv. mmN mo. mvm. 006 ~35
vm. mm. ON. RN. NN. No.0 N F. m 5% no.0 De..—
om. ..m. vm. ho. mnN 0o. ham. on...“ m2
3. mm. m ﬁm 5. mmoé mud n...—
no. NV. had No. mmm. vwé .32
mod mo. P omN.m omN m:
NN.mm om. 5mm m ﬁmh ”ma—Z
hm. wmh. mm. F «5
by; mm: mm: HOD :— msm 05m NE A: .32 mn— mmz <m Gm =32

  
     
         
        
     

 

 

 

 

 

 

.252 5.5 e... ..a use: 85:55 u 5. 2.3.

85

    

oocaﬁuotom
SEE uniﬁed

 

oocaccotom
“8.82 ..ozﬁom

 

    

magneto.—
uoauxzm Roz

 

£8252 8%.... All
9:28:38 amour—ma All
9:30:28 HgoEcwaéoZ A .........

   

Enacocﬂom

 

Ecoﬁom

       

oocotoaxm

 

 

 

(

GOSNHODE—OU

5am

       

 

 

 

gag—gm :ozsaom
8: 302 825.980
9 «£2
mmocoégocﬁ
.. “88$
who—88m \
9 $2

 

 

 

 

E52 ‘83: < n 3. 95$..—

86

Table 4.8 — Results of the Path Model

 

 

 

 

. ML
Hypothesrzed Paths Estimate t-value Hypotheses
Product Innovativeness 9 New Product 0.38 6.238** H la supported
Advantage to Consumers
Product Innovativeness 9 New Product '
Advantage to Retailers 0'07 1'263 Hlb not supported
New Product Advantage to Consumers 9
New Product Evaluation 1'82 1089* H2a supported
New Product Advantage to Retailers 9 *
New Product Evaluation 1'42 2'019 H2b supported
New Product Advantage to Consumers 9
New Product Performance 0.03 0.284 H3a not supported
New Product Advantage to Retailers 9 *
New Product Perfo ce 0.25 2.787* H3b supported
Reputation 9 Trust 0.28 3.230** H4 supported
Past Experience 9 Trust 0.03 2527* H5 supported
Personal Relationship 9 Trust 0.11 1.849 H6 not supported
Trust 9 New Product Evaluation 1.42 2.020* H7 supported
Trust 9 Collaboration 0.35 5.998** H8 supported
New Product Evaluation 9 Collaboration 0.02 2.696** H9 supported
Collaboration 9 New Product Performance 0.31 4.047** H10 supported
New Product Performance 9 Retailer **
Market Performance 0.41 4.604 H11 supported
New Product Performance 9 Retailer M
Financial Performance 0'36 3'501 H12 supported
Retailer Market Performance 9 Retailer H
Financial Perfo ce 0.50 4.592 H13 supported
Buyer Age 9 New Product Evaluation -0.00 0002 Control variable (n.s.)

. ML

Non-Hypothesrzed Paths Estimate t-value
Product Innovativeness 9 Collaboration 0.21 3851’”
Product Innovativeness 9 Retailer Market 020 3.062**
Performance
Product Advantage to Consumers 9 H
Product Advantage to Retailers 0'51 6'772
Personal Relationship 9 Collaboration 0.12 3.206“

 

S-B X2 = 77.27, p=.006, df = 49, CFI= .934; IFI= .939; MEI: .910; RMSEA: .062;

90% CI (.034, .088)

 

* p < 0.05
** p < 0.01
Note: n.s.=non-signiﬁcant

87

CHAPTER 5
DISCUSSION

Discussion of the Results

Marketing scholars have paid little attention to examining the importance of
retailers in determining new product success despite the critical role retailers play in new
product introductions (Rao and McLaughlin 1989). In this study, we argue that retailers
not only make buying decisions regarding new products but also inﬂuence the
performance of new products in the market. Drawing theories from different disciplines,
we proposed an integrated conceptual model explaining retailer new product evaluation
and new product success and conducted an empirical study to test the model with data
collected in China. Overall, the model showed good ﬁt properties. Out of the 16
hypothesized paths (including a and b in H1, H2, H3), 13 paths (i.e. 81%) are conﬁrmed.
Therefore, it can be concluded that the results of this study support the proposed model.
New Product Advantage

This study incorporated the construct of new product advantage from NP
literature in the retailer new product evaluation model. However, different from the
unidimensional construct found in NP literature, we propose new product advantage as a
second-order construct with two dimensions: product advantage to consumers and
product advantage to retailers. The CFA results conﬁrmed our proposition of the factorial
structure of this construct. Further, the results of the path model showed that both product
advantage to consumers and product advantage to retailers are important to retail buyers’
consideration of new products. These findings imply that the construct of new product

advantage may need to be redeﬁned to reﬂect its importance to both types of customers:

88

end-consumers and retailers, because it is likely that retailers and consumers may seek
different advantages from a new product. For example, consumers may purchase a new
product because it appears to be of better quality than existing products while retailers
may buy a new product because it proVides an opportunity of obtaining a higher profit
margin. In addition, as suggested by the revised model, product advantage to consumers
is also positively related to product advantage to retailers. For instance, a new, high
quality product may be perceived by retailers as a good opportunity to obtain a higher
proﬁt margin. Thus, the results of this study suggest that the construct of new product
advantage and the interrelationship between its two dimensions are more complex and
more interesting than speciﬁed in literature, and so deserve further investigation.

In NP literature, product innovativeness is usually posited as an antecedent of
product advantage (Calantone, Chan and Cui 2006; Gatignon and Xuereb 1997). Our
results conﬁrmed this relationship through the linkage between product innovativeness
and product advantage to consumers. In addition, the results also suggested that product
innovativeness leads to product advantage to retailers through product advantage to
consumers. The indirect path between product innovativeness and product advantage to
retailers can be explained logically. An innovative product is designed for consumers;
therefore, new product advantages are tailored to meet the desires of consumers
(Montoya-Weiss and Calantone 1994). For retailers, however, only those product
advantages that elicit consumer demand and bring proﬁt to retailers are meaningful.
Therefore, an innovative product itself is not attractive to retailers; it is only attractive to
retailers when its innovative features are perceived as superior by consumers and are thus

likely to beneﬁt retailers as well.

89

Trust

In this study, following Williamson (1991), we examined organizational trust.
Anderson and Weitz (1989) examine relationships between independent sales agents and
their principals and ﬁnd that the reputation of a principal enhances a sales agent’s trust of
the principal. Similarly, Ganesan (1994) conducts a ﬁeld study involving both retailers
and vendors and his ﬁndings conﬁrm that one party’s reputation leads to the other party’s
trust. The results of our study also support a positive relationship between a supplier’s
reputation and a retailer’ trust of the supplier, thus providing further evidence for this
link.

Although theorized in several studies, the link between past experience and trust
has not yet been empirically conﬁrmed (Ganesan 1994; Kumar, Scheer and Steenkarnp
1995). Our study is likely the ﬁrst empirical study that conﬁrms the positive relationship
between a retailer’s past experience with a supplier and the retailer’s trust of the supplier.
Taken together, our results suggest that a retailer’s trust of a supplier can be derived from
the reputation of the supplier and the retailer’s past experience with the supplier.

Though hypothesized in the model, the link between personal relationships
(between a retailer buyer and a salesperson) and the retailer’s trust of the supplier was not
conﬁrmed. This ﬁnding may seem counterintuitive in a Chinese setting because personal
relationships, or “guanxi” as they are called in China, have been advocated for in
international marketing literature because of the belief that such relationships are an
important tool used to cultivate trust in China (Armstrong and Yee 2001; Chang and Ding
1995). However, recently, some researchers have theorized about the decreasing

importance of personal relationships/guanxi in business relationships. For example,

90

Guthrie (1998) argues that personal relationships/guanxi will eventually disappear as the
institutional structure and the rational-bureaucratic system matures. The most recent
empirical evidence also shows that personal relationships are not related to trust as China
is transforming from a planned economy to a market economy. Huang, Stemquist, and
Calantone (2006) conduct a study of retail buyer-supplier relationships in China and ﬁnd
that personal relationships between retail buyers and salespersons are not related to
retailers’ trust of suppliers. Similarly, Chen (2006) also concludes that personal
relationships do not lead to a retailer’ trust of a relatively new supplier at the beginning of
the relationship building process. Therefore, our results add evidence to the recent
argument that personal relationships/guanxi are losing its importance, especially in
business settings.
New Product Evaluation

New product evaluation is proposed as a new construct in our model. Rao,
McLaughlin and Hawkes (1995) suggest that a retailer’ judgment of a new product is of
particular importance to the potential success of the new product. However, no attention
has been given to the retailers’ judgment process per se. Instead, most studies have
focused on the consequence of the judgment process—either the accept or reject decision.
This study is therefore the ﬁrst to examine retailers’ judgments of new products through
retailer new product evaluation. As hypothesized, a retailer’s evaluation of a new product
is based on considerations of both the product factor and the supplier factor. Speciﬁcally,
the results of this study show that the product factor includes product advantage to
consumers and product advantage to retailers, and the retailer factor is comprised of

retailer trust of the supplier. In an industrial purchasing setting, Doney and Cannon

91

(1997) report that trust in the supplier is unrelated to purchase but related to future
purchase consideration. They therefore suggest that trust is only an order qualiﬁer
(meaning a certain level of trust is required before a manufacturer is even considered as a
potential supplier) rather than an order winner (meaning that the extent of trust does
inﬂuence the purchase decision). However, in a retail buying setting, our results show
that retailer trust of the supplier is positively related to retailer new product evaluation. In
the context of China, trust may actually be an order winner because, compared to the
West, China is a low-trust society, and therefore trust is of the highest importance in
organizations (Atuahene-Gima and Li 2002).
Collaboration

In this study, we also examined whether retailer new product evaluation exerts
any inﬂuence on a retailer’s behavior after a new product is stocked. Several researchers
have called for studies of buyer behavior in the buying/purchasing setting. For example,
Rao, McLaughlin and Hawkes (1995) suggest that analysis of retailers’ behavior can
yield valuable insights on channel efﬁciency and management during new product
introduction. Similarly, Lusch and Vargo (1998, p. 585) argue that “in organizational
buying literature, purchasing decisions are typically involved with evaluating relative
importance weights of attributes such as price, delivery, quality. What is missing in this
literature is the empirical investigation of actual perceptions of decision criteria in
relation to intended behavior.” This study answered these calls by analyzing retailers’
collaborative behavior after new products are accepted in stores. Weitz and J ap (1995)
point out that most of the empirical research on channel relationships characterized the

relationships at one point in time but does not provide much insight into the factors

92

leading to the development of the relationship or the effectiveness of the relationships.
The model proposed in this study can be seen as an initial step to solve this problem
because 1) our model consists of two stages: the evaluation stage and post-evaluation
stage; and 2) we explicitly posited the antecedents and consequences of retailers’
collaborative behavior.

As hypothesized, retailer new product evaluation and retailer trust of the supplier
are positively related to retailer collaboration with that supplier. Andaleeb (1995) reports
that a distributor’s trust of a supplier is positively related to the distributor’s cooperation
with the supplier. Anderson and Weitz (1989) ﬁnd that a sales agent’s trust of a principal
leads to the sales agent’s communication with the principal and willingness to continue
with the business relationship. Likewise, Morgan and Hunt (1994) also ﬁnd a positive
link between a retailer’s trust of a supplier and the retailer’s commitment to the
relationship and a positive link between a retailer’s trust of a supplier and the retailer’s
cooperation with the supplier. Therefore, our results are consistent with the findings in
previous studies.

In addition, our results also suggest that product innovativeness and personal
relationships between a buyer and a supplier are positively related to retailers’
collaboration with suppliers. The link between product innovativeness and retailer
collaboration is interesting and may be related to the ﬁnding by Stemquist and Chen
(2006) that in general Chinese retailers think highly of and welcome unique, innovative
products. For example, it is reported that some of the retailers in China may waive the
slotting fee if they think a new product is unique (Huang, Stemquist and Calantone

2006). If a product is truly innovative, a retailer may be more willing to collaborate with

93

the supplier. The positive relationship between personal relationships and retailer
collaboration is a surprising ﬁnding. However, when this ﬁnding is considered together
with the finding that personal relationships were not signiﬁcant in leading to retailer trust
of a supplier, it may suggest that persOnal relationships do have different effects on
buyer-supplier relationships at different relationship stages. For instance, personal
relationships may be more important in driving trust at the relationship maintenance stage
(Chen 2006).
New Product Performance and Retailer Performance

New product performance and retailer performance are consequences in our
model. In general, the results support our hypotheses. Retailer collaboration and new
product advantage to retailers are positively related to new product performance. New
product performance also leads to retailer market performance and retailer ﬁnancial
performance. In addition, the results suggest that product innovativeness is also positively
related to retailer market performance. This probably explains why innovative products
are preferred by Chinese retailers in general. Further, the results did not suggest a link
between product innovativeness and retailer ﬁnancial performance, which seems
consistent with the finding that product innovativeness did not lead to product advantage
to retailers directly. In other words, if an innovative product does not bring any beneﬁts
to a retailer, it is not likely to contribute to the ﬁnancial performance of the retailer.
Theoretical Implications

The theoretical contributions of this study are ﬁvefold. First, this study
incorporates theories from areas of newproduct development, marketing, and

organizational buying and proposes a formal model of retail new product buying

94

behavior. The results of the study demonstrate a good model fit indicating the
applicability of the proposed model. In particular, this study has answered both the call
for an expanded incorporation of distribution in new product research (Rao, McLaughline
and Hawkes 1995; Wind and Mahajan 1987) and the call for developing a theoretical
model in retail buying behavior (Wagner, Ettenson and Parrish 1989). In this way, this
study represents an initial attempt to connect RB literature with NP literature in the area
of retailer new product buying.

Second, this study expands the domain of retailer new product buying from the
previous dichotomous accept/reject decision to a continuous evaluation stage. The results
of this study clearly show that retailers’ evaluations of new products are subject to
retailers’ perceived product advantages of new products and retailers’ trust of suppliers.
In addition, this study suggests that new product advantage, a construct frequently studied
in NP literature, may possess a higher order factorial structure consisting of two
dimensions: product advantages to consumers and product advantages to retailers.

Third, this study incorporates both an evaluation stage and a post-evaluation stage
in the retailer new product buying model. In this manner, we explicitly link retailers’
perceptions of new products during the new product evaluation process to retailers’ actual
behavior after the evaluation and examine the inﬂuence of their behavior on new product
performance in stores. In effect, since the retailer has become the “captain” of the
marketing channel (Lusch and Vargo 1998, p.582), these linkages are particularly
important in the stream of NP research. The results of this study provide evidence of the

importance of retailers in affecting new product performance in the market.

95

Fourth, this study builds on the relationship marketing paradigm and is the ﬁrst
that includes relational constructs in retail buying studies. Speciﬁcally, the results of the
study demonstrate that the retailer’s trust of the supplier exerts an impact on retailer new
product evaluation and the subsequent relationship between the retailer and the supplier.
The results also show that the retailer’s relational behavior, comprised by such attributes
as cooperation, communication, and commitment, inﬂuence new product performance in
the market.

Fifth, this study is the ﬁrst to examine new product performance from the
retailer’s perspective and the ﬁrst to link new product performance to retailer
performance. Although previous studies suggest new products contribute to retail sales
and proﬁts (Montgomery 1975), there has been no empirical evidence until now. The
results of this study provide evidence for this link.

Managerial Implications
Implications for Retailers

Buying new products is not a trivial decision for retailers (Sullivan 1997).
Retailers need to learn to better allocate scarce resources, reduce costs, and thereby
maximize both organizational and channel efficiency. Therefore, retailers need to
understand their own behavior and the consequences of their behavior. The results of this
study suggest the following implications for retail buyers and managers:

1) New product performance leads to both retailer market performance and retailer
ﬁnancial performance. Although new products have high failure rates and
therefore bear risks and uncertainties, in general, new products contribute to

retailer performance. The favorable performance of new products brings mutual

96

2)

beneﬁts to manufacturers and retailers and should be the common goal of both
parties. Further, to the extent that new product success is associated with retailers’
collaboration with suppliers, 3 retailer may be better off developing a partnership
relationship with a supplier during the new product introduction process
(Anderson and Narus 1991) to ensure new product success because new product
performance is a driver of retailer performance as well.

The results of this study show that personal relationships between retail buyers
and suppliers’ salespersons are positively related to retailers’ collaboration with
suppliers. Therefore, although some retailers may have policies disencouragin g or
even forbidding personal relationships between buyers and salespersons (Huang,
Stemqui st and Calantone 2006), managers may need to carefully deﬁne the term
“personal relationships” and differentiate the good consequences of these
relationships from the bad. Further research is needed in this area to give more

explicit guidance to retailers.

Implications for Suppliers

Along with the implications for retailers, the results of this study provide

implications for suppliers as well. New product introduction is expensive for
manufacturers. It is reported that total development expenditures for the Ultra-Pamper
diaper, a product of the Procter & Gamble Company cost $1.5 billion (Rao and
McLaughlin 1989). Ensuring the market success of a new product is important to a

manufacturer. The adoption of a new product by consumers is dependent upon the new

product being distributed by the retailer (Kuester, Gatignon and Robertson 2000).

Therefore, for manufacturers, understanding the buying behavior of retailers is as

97

important as the knowledge of the buying behavior of consumers. The results of this ‘

study suggest the following implications for manufacturers:

1)

2)

Both retailers and consumers are manufacturers’ customers; therefore,
manufacturers need to understand and differentiate two types of product
advantages—product advantage to consumer and product advantage to retailer.
Retail buyers buy the new product before consumers do. Just like consumers, they
need to get excited about the new product (Ottum 1996). Therefore, in addition to
targeting product advantages to consumers, manufacturers also need to develop
product attributes that are likely to bring benefits to retailers. For example, a
manufacturer may put effort into developing a certain unique feature that may
result in consumers’ willingness to pay a premium for it. The manufacturer may
thereby offer the retailer an opportunity for higher markup gained by carrying this
new product.

From a managerial perspective, it is also important for manufacturers to know and
understand the factors that determine retail buying processes. These factors may
be internal or external to the manufacturer and thus either controllable or
uncontrollable. Particularly, manufacturers need to know what variables are
controllable and what variables are not controllable but are inﬂuenceable. For
instance, among the factors inﬂuencing retailer new product evaluation, both
product advantage to consumers and product advantage to retailers are
controllable variables. Manufacturers need to make sure that their understanding
of these two factors is the same as that of retailers’. In contrast, retailer trust of a

supplier is an uncontrollable factor, but it can be inﬂuenced by company

98

reputation. Manufacturers can send “signals” to retailers that they intend to work
together with retailers over the long run by safeguarding their own reputations.
These signals help build the level of mutual trust (Anderson and Weitz 1989).
3) The results of this study show‘that retailer collaboration contributes to new
product performance. Therefore, eliciting retailer collaboration is important to
ensure new product success in the market. Based on our results, manufacturers
can increase retailer collaboration in three ways: through improving product
advantage to retailers, through developing an organizational trust, and through
cultivating personal relationships with retail buyers.
Limitations and Directions for Future Research

Like any empirical research, our study has several limitations. First, we used
subjective measures of new product performance, retailer market performance, and
retailer ﬁnancial performance in our study. Although subjective measures of performance
have been widely used in marketing and NP literature (Homberg, Hoyer and Fassnacht
2002; Langerak, Hultink and Roben 2004), adding objective performance measures may
be more desirable and could further validate subjective performance measures. Therefore,
for further research, it is desirable to incorporate objective performance measures.

Second, as argued by Cannon and Perreault (1999), in any research project,
choices made by the researchers create limitations in interpreting the results. Although
we propose our model as an integrated model of new product evaluation and new product
success, we only investigated the retail buyer’s evaluation of new products accepted in

stores because of the way we asked the respondents to answer questions. Further research

99

needs to investigate the retail buyer’s evaluation of new products that are rejected by
buyers to validate the ﬁrst half of our model.

Third, in our survey study, we asked the respondents to think back to a new
product they accepted the previous year and answer questions relevant to the situations
surrounding their acceptance of the new product. Although the respondents did not
express any problems with recalling situations and previous studies using the same
method did not report any problems (Rao and McLaughlin 1989), it is possible that the
respondents’ memories may not have been precise and thus information about the
situations associated with new product acceptance may be not accurate. To solve this
potential problem, a longitudinal study may be a better design for further research. In
addition, longitudinal studies are also a preferred method to infer causal relationships.

Finally, although we propose our model as a general model of new product
evaluation and success, we only tested the model in China. Further studies need to be
conducted in other countries to validate the model and compare the strength of the
relationships among constructs in different countries.

In addition to alleviating these limitations, there are other fertile avenues for
further research. First, the data for this research are based on the retailer's perspective of
the new product buying process. Readers should keep in mind that suppliers' points of
view might be different from retailers’. For example, suppliers may have differing
perspective regarding whether or not retailers’ collaboration has any impact on new
product performance. In studies on channel issues, dyadic data collection from both sides

is usually encouraged (Siguaw, Simpson, and Baker 1998). A further study using dyadic

100

responses from both retailers and suppliers may reveal any differences in their
perceptions and help to increase channel efﬁciency.

Second, in proposing our model, we have attempted to strike a balance between
parsimony and completeness in identifying theoretically relevant constructs. The scope of
the research is therefore delimited by the constructs we explicitly specify, measure, and
evaluate. However, it is possible that other constructs that are potentially important in the
new product buying process were not included. For example, a retailer’s private label
strategy may have an inﬂuence on retailer new product evaluation because a
manufacturer’s new product may be a direct competitor of some of the retailer’s private
label products (Corstjens and Corstjens 1995). Investigation into the inﬂuence of
retailers’ private label strategy on retailer new product evaluation may prove to be a
fruitful area of research particularly when retail private labels are prevalent nowadays and
will continue to prosper in the future.

Third, Dwyer, Schurr, and Oh (1987) propose a classiﬁcation scheme based on
the stages and processes along which business relationships develop. In the new product
buying context, the business relationship stage that a retailer and a supplier stay in may
be different from case to case. For instance, a new product supplied by a new supplier
indicates that the business relationship between the retailer and the supplier is in the
initiation stage, while another new product supplied by a long-term supplier indicates that
the business relationship stage may be at a maintenance stage. Different business
relationship stages may inﬂuence the relationship and behavior between the retailer and

the supplier (Chen 2006). Thus, the retailer’s evaluation of a new product, and later the

101

post-evaluation behavior, may be contingent on the business relationship stage the retailer
stays in with the particular supplier.

Finally, the model may also be subject to other moderating effects. For instance,
the relative power between a retailer and a supplier may moderate proposed relationships
among constructs. Since only 10% of our sample retailers (i.e. 15 cases) are classiﬁed as
less powerful than their suppliers, our data prevent us from conducting a post-hoe check
of the moderating effect of power balance. However, further research may strive to solve
this problem by asking respondents to think of two suppliers (one who is more powerful
than the retailer and another who is less powerful than the retailer) who provided a new
product in the past, and answer questions for new products and suppliers respectively.
Conclusion

New product development, the process of bringing new products to market, is one
of the most important issues in business research today (Hauser, Tellis, and Grifﬁn 2006).
However, with the explosion in new product introductions and a commensurate increase
in the number of new product failures, a decision on whether or not to carry a new
product is not trivial for a retailer (Rao and Mahi 2003; Sullivan 1997). Previous studies
argued that retailers play the role of “gatekeepers” in new product introductions. Our
study conﬁrms this argument and extends the argument to suggest that retailers play the
role of “co-producers” in new product introductions. Taken together, the results of this
study suggest that the retail buyer derives his/her evaluation of a new product from both
the product factor (i.e. new product advantage) and the supplier factor (i.e. retailer trust of
the supplier). Further, this evaluation inﬂuences the extent of the retailer’s collaborative

behavior, and in turn, inﬂuences new product performance. Due to the critical role

102

retailers play in the potential market success of a new product, this study stands for an
initial step in exploring the importance of retailer new product buying in the NPD

chain—an area that connects RB literature with NP literature.

103

REFERENCES

Akerlof, G. A. (1970). The market for ‘Lemon’: Quality under uncertainty and the
market mechanism. Quarterly Journal of Economics, 84(August), 488-500.

Alder, N. J ., Campbell, N ., and Laurent, A. (1989). In search of appropriate
methodology: From outside the People’s Republic of China looking in. Journal of
International Business Studies, 20, 61-74.

Alpert, F., Kamins, M., and Graham, J. (1992). An examination of reseller buyer
attitudes toward order of brand entry. Journal of Marketing, 56(July), 25-37.

Alpert, F., Kamins, M., Sakano, T., Onzo, N., and Graham, J. (2001). Retail buyer
beliefs, attitude and behavior toward pioneer and me-too follower brands. International
Marketing Review, 18(2), 160-187.

American Productivity & Quality Center (2003). Improving new product
development performance and practices. Houston, TX.: APQC. Available at:
www.apqc.org/pubs/NPD2003.

Andaleeb, S. S. (1995). Dependence relations and the moderating role of trust:
Implications for behavioral intentions in marketing channels. Intematiomll Journal of
Research in Marketing, 12(2), 157—172.

Andaleeb, S. S. (1996). An experimental investigation of satisfaction and
commitment in marketing channels: The role of trust and dependence. Journal of
Retailing, 71(1), 77—93.

Anderson, E. and Coughlan, A. T. (1987). International market entry and
expansion via independence or integrated channels of distribution. Journal of Marketing,
51(January), 71-82.

Anderson, J. C. and Gerbing, D. W. (1988). Structural equation modeling in
practice: A review and recommended two-step approach. PsychologLBulletin, 103(5),
41 1-23.

Anderson, E. and Weitz, B. A. (1989). Determinants of continuity in conventional
industrial channel dyads. Marketing Science, 8(4), 310-323.

Anderson, E. and Weitz, B. A. (1992). The use of pledges to build and sustain
commitment in distribution channels. Journal of Marketing Research, 29 (February), 18-
34.

Anderson, J. C. and Narus, J. A. (1990). A model of distributor ﬁrm and
manufacturer ﬁrm working partnerships. Journal of Marketing, 54(January), 42—58.

104

Anderson, E. W. and Sullivan, M. W. (1993). The antecedents and consequences
of customer satisfaction for ﬁrms. Marketing Science, 12(2), 125-43.

Anderson, E. W., Fomell, C., and Rust, R. T. (1997). Customer satisfaction,
productivity, and proﬁtability: Differences between goods and services. Management
Science, 16(2), 129-145. ‘

 

Armstrong, R. W. and Yee, S. M. (2001). Do Chinese trust Chinese? A study of
Chinese buyers and sellers in Malaysia. Journal of International Marketing, 9(3), 63-86.

Atuahene-Gima, K. and Li, H. (2000). Marketing’s inﬂuence tactics in new
product development: A study of high technology ﬁrms in China. Journal of Product arrg
Innovation Management, 17, 451-470.

Atuahene-Gima, K. and Li, H. (2002). When does trust matter? Antecedents and
contingent effects of supervisee trust on performance in selling new products in China
and the United States. Journal of Marketing, 66(July), 61-81.

Bagozzi, R. P. and Phillips, L. W. (1982). Representing and testing organizational
theories: A holistic construct. Administrative Science Quarterly, 27(3), 459-89.

Banting, P. M. and Blenkhom, D. L. (1988). The mind of the retail buyer.
Management Decision, 26(6), 29-36.

Bentler, P. M. (1995). E08 Structural Equations Program Manual. Encino, CA:
Multivariate Software, Inc.

Besanko, D., Dubé, J ., and Gupta, S. (2005). Own-brand and cross-brand retail
pass-through. Marketing Science, 24(1), 125-137.

Beverland, M. B., Ewing, M. T., and Matanda, M. J. (2006). Driving-market or
market-driven? A case study analysis of the new product development practices of
Chinese business-to-business ﬁrms. Industrial Marketing Management, 35, 383-393.

Bowersox, D. J ., Stank, T. P., and Daugherty, P. J. (1999). Lean launch:
Managing product introduction risk through response-based logistics. Journal of Product
and Innovation Management, 16, 557-568.

Bradach, J. L. and Eccles, R. G. (1989). Price, Authority and Trust: From ideal
types to plural forms. Annual Review of Sociology. 15, 97-118.

Brown, S. L. and Eisenhardt, K. M. (1995). Product development: Past research,
present ﬁndings, and future directions. The Academy of Management Review, 20(2),
343-378.

105

Boyd, R. and Richerson, P. J. (1985). Culture and the Evolutionary Process.
Chicago: University of Chicago Press.

Byme, B. M. (2001), Structural Equation Modeling with E08 and EQS/Windows:
Basic Concepts, Applications, and Programnﬁg. Thousand Oaks, CA: Sage Publications.

Calantone, R. J. and di Benedetto, C. A. (1988). An integrative model of the new
product development process: An empirical validation. Journal of Product Innovation
Management, 5, 201-215.

Calantone, R. J ., Schmidt, J. B., and Song, X. M. (1996). Controllable factors of
new product success: A cross-national comparison. Marketing Science, 15(4), 341—358.

Calantone, R. J ., di Benedetto, C. A., and Stank, T. P. (2005). Managing the
supply chain implications of launch. In Kahn, Kenneth B. (Ed.) The PDMA Handbook of
New Product Develcmment, 2nd edition (pp. 466-47 8), New York: John Wiley & Sons,
Inc.

Calantone, R. J ., Chan, K., and Cui, A. S. (2006). Decomposing product
innovativeness and its effect on new product success. Journal of Product Innovation
Management, 23, 408—421.

Cannon, J. P. and Perreault W. D. (1999). Buyer—seller relationships in business
markets. Journal of Marketing Research, 36(November), 439-60.

Chang, K. and Ding, C. G. (1995). The inﬂuence of culture on industrial buying
selection criteria in Taiwan and Mainland China. Industrial Ngrketing Management, 24,
277-84.

Chen, Z. Y. (2006). Chinese Retail Buyer-Seller Initiation and Maintenance of
Relationships. Doctoral Dissertation.

China Statistics Yearbook (2001). Beijing: China Statistical Publishing House.

Churchill. G. A. (1979). A paradigm for developing better measures of marketing
constructs. Journal of Marketing Research, 16(1), 64-70.

Coase, R. H. (1937). The nature of the ﬁrm. Economica, 4, 386-405.

Cooper, R. G. and Kleinschmidt, E. J. (1990). New Product: The Key Factors in
Success. Chicago: American Marketing Association.

Cooper, R. G. (2005). New products—What separates the winners from the losers

and what drives success. In Kahn, Kenneth B. (Ed) The PDMA HandbooLof New
Product Development, (pp. 466-478), New York: John Wiley & Sons, Inc.

106

Corstjens, J. and Corstjens, M. (1995). Store Wars, Chichester, England: John
Wiley & Sons, Ltd.

Couper, M. and De Leeuw, E. (2003), Nonresponse in Cross-Cultural and Cross-
National Surveys. Cross-Cultural Survey Methods. Hoboken, NJ: John Wiley & Sons.

Crawford, C. M. (1977). Marketing research and the new product failure rate.
Journal of Marketing, 41(April), 51-61.

Crawford, C. M. (1994). New Products Management (4th edition). Homewood,
IL: Irwin.

 

Crozier, R. and Ranyard, R. (1997). Cognitive process models and explanations of
decision making. In W.R. Crozier and O. Svenson (Eds.), Decision making: Cognitive
models and explanations, 2nd edition (pp. 5-20). London and New York: Routledge.

Dawson J. (2000). Viewpoint: Retailer power, manufacturer power, competition
and some questions of economic analysis. International Journal of Retail & Distribution
Management, 28(1), 5-8.

Day, G. S. and Wensley, R. (1988). Assessing advantage: A framework for
diagnosing competitive superiority. Joumzi of Marketing, 52 (April), 1-20.

Delre, S. A., Jager, W., Bijmolt, T. H. A., and Janssen, M. A. (2007). Targeting
and timing promotional activities: An agent-based model for the takeoff of new products.
Journal of Business Research, 60(8), 826-835.

Desai, P. (2000). Multiple messages to retain retailers: Signaling new product
demand. Marketing Science, 19, 381-389.

Desiraju, R. (2001). New product introduction, slotting allowances, and retailer
discretion. Journal of Retailing, 77, 335-358.

Di Benedetto, C. A. (1999). Identifying the key success factors in new product
launch. Joumg of Product Innovation Management, 16, 530-544.

Doney, P. M. and Cannon, J .P. (1997). An examination of the nature of trust in
buyer—seller relationships. Journal of Marketing, 61(April), 35-51.

Doyle, P. and Weinberg, C. B. (1973). Effective new product decisions for
supermarkets. Operational Research Quarterly, 24(1), 45-54.

Dwyer, F., Schurr, RH, and Oh, S. (1987). Developing buyer—seller
relationships. Journal of Marketing, 51(Apri1), 11-27.

Economist (2005, May 5). Makrﬂg a Meal of Pacgkaged Food.

107

Fan, X., Thompson, B., and Wang L. (1999). Effects of sample size, estimation
methods, and model speciﬁcation on structural equation modeling ﬁt indexes. Structural
Equation Modeling, 6(1), 56-83.

Fisher, M. L., Raman, A., and McClelland, A. S. (2000). Rocket science retailing
is almost here—Are you ready? Harvard Business Review, 78, 115-127.

Fontenot, R. J. and Wilson, E. J. (1997). Relational exchange: A review of
selected models for a prediction matrix of relationship activities. Journal of Business
Research, 39, 5-12.

Fomell, C. (1992). A national customer satisfaction barometer: The Swedish
experience. Journal of Marketing, 56(January), 6-21.

Frazier, G. L., Gill, J. D., and Kale, S. H. (1989). Dealer dependence levels and
reciprocal actions in a channel of distribution in a developing country. Journal of
Marketing, 53(January), 50-69.

Frazier, G. L. (1999). Organizing and managing channels of distribution. Journal
of the Academy of Marketing Science, 27(2), 226-240.

Gale, F. and Reardon, T. (2004, November). China’s modernizing supermarket
sector presents major opportunities for US. agricultural exporters. A x orter,
Available: http://www.fas.usda.gov/info/agexporter/2004/November/pgs%204-8.pdf

Gallo, A. E. (2007). First major drop in food product introduction in over 20
years. Food Review. Available at:
http://ﬁndarticles.com/p/articles/mi_m3765/is_n3_v20/ai_20465554

Ganesan, S. (1994). Determinants of long-term orientation in buyer—seller
relationships. Journal of Marketing, 58(April), 1-19.

Garcia, R.and Calantone, R. J. (2002). A critical look at technological innovation
typology and innovativeness terminology: A literature review. Journal of Product
Innovﬁation Management, 19(3), 110-132.

Gatignon, H. and Xuereb, J. (1997). Strategic orientation of the ﬁrrna and new
product performance. Journal of Marketing Research, 34(1), 77-90.

Gerbing, D. W., Hamilton, J. G., and Freeman, E. B. (1994). A large-scale
second-order structural equation model of the inﬂuence of management participation on
organizational planning beneﬁts. Journal of Management, 20(W inter), 859-886.

108

Gerlich, N. R., Walters, R. G., and Heil, O. P. (1994). Factors affecting retailer
acceptance of new packaged goods: Some comparisons across new product types. Journal
of Food Products Marketing, 2(2), 65-92.

Geyskens, I., Steenkamp, J. E., and Kumar, N. (1999). A meta-analysis of
satisfaction in marketing channel relationships. Journal of Marketing Research, 36(May),
223-38. ‘

Granovetter, M. (1985). Economic action and social structure: The problem of
embeddedness. American Journal of Sociology, 91(3), 481-510.

Grifﬁn, A. (1997). PDMA research on new product development practices:
Updating trends and benchmarking best practices. Journal of Product Innovation
Management, 14, 429-458.

Grashof, J. F. (1970). Supermarket chain product mix decision criteria: A
simulation experiment. Journal of Marketing Research, VII(May), 235-242.

Guthrie, D. (1998). The declining signiﬁcance of Guanxi in China’s economic
transition. The China Quarterly, 154, 254—282.

Hair, J. R, Anderson, R. E., Tatham, R. L., and Black, W. C. (1998). Multivariate
Data Analysis. New Jersey: Prentice Hall.

Hamel, G. and Prahalad, C. K. (1991). Corporate imagination and expeditionary
marketing. Harvard Business Review, (July-August), 81-92.

Hansen, T. H. and Skytte, H. (1998). Retailer buying behaviour: A review.
International Review of Retail,Distribution and Consumer Research, 8(3), 277-301.

Hansen, K. (2001). Purchasing decision behavior by Chinese supermarkets.
Intemtional Review of Retail, Distribution and Coﬁnsumer Reseaih, 11(2), 159-175.

Hauser, J ., Tellis, G. J ., and Grifﬁn, A. (2006). Research on innovations: A
review and agenda for marketing science. Marketing Science, 25(6), 687-717.

Heeler, R. M., Kearney, M. J ., and Mehaffey, B. J. (1973). Modeling supermarket
product selection. Joumal of Mgketing Research, X(February), 34-37.

Hogarth, R. (1987). Judgment and Choice. New York: John Wiley and Sons.
Homburg, C., Pﬂesser, C. (2000). A multiple-layer model of market-oriented

organizational culture: Measurement issues and performance outcomes. Journal of
Marketing Research, 37(4), 449-462.

109

Homburg, C., Hoyer, W. D. and Fassnacht, M. (2002). Service orientation of a

retailer’s business strategy: Dimensions, antecedents, and performance outcomes. Journal
of Marketing, 66(4), 86-101.

Hu, D., Reardon, T., Rozelle, S., Timmer, P., and Wang, H. (2004). The
Emergence of supermarkets with Chinese characteristics: Challenges and opportunities
for China’s agricultural development. Development Policy Review, 22(5), 557-586.

Huang, Z., Li, S. X., and Mahajan, V. (2002). An analysis of manufacturer-
retailer supply chain coordination in cooperative advertising. Decision Sciences, 33(3),
469-494.

Huang, Y., Stemquist, B., and Calantone, R. J. (2006). Chinese retail buyer-
supplier relationship in a transitional economy: Does Guanxi still matter? Working Paper.

Hultink, E. J, Tholke, J. M., and Robben, H. S. J. (1999). Retailers’ adoption
decision of new consumer durables. Journal of Product Innovgrgon Management, 16, 483-
490.

Hunter, G. K. and Perreault, W. D. Jr. (2007). Making sales technology effective.
Journal of Marketing, 71(January), 16-34.

Jefferson, G., Hu, A., Guan, X., and Yu, X. (2003). Ownership, performance, and
innovation in China’s large and medium—size industrial enterprise sector. China
Economic Review, 14, 89-113.

Jobber, D. and Saunders, J. (1988). An experimental investigation into cross-
national mail survey response rates. Journal of Interﬂml Business Studies, 19(3), 483-
489.

Johansson, U. (2001). Retail buying: Process, information and IT use: A
conceptual framework. International Review of Retail. Distribution and Consumer
Research, 11(4), 329-357.

 

J ohansson, U. (2002). Food retail buying processes: a study of the UK, Italy and
Sweden. Intemtional Journal of Retail @d Distribution Management, 30(1 1/ 12), 575-
586.

Johnson, W. J. and Lewin, J. E. (1996). Organizational buying behavior: Toward
an integrative framework. Journal of Business Resear_cl_r, 35(1), 1-15.

Kadiyali, V., Chintagunta, P., and Vilcassim, N. (2000). Manufacturer-retailer

channel interactions and implications for channel power: An empirical investigation of
pricing in a local market. Marketing Science, 19(2), 127-148.

110

Kale, S. (1986). Dealer perceptions of manufacturer power and inﬂuence
strategies in a developing country. Journal of Marketing Research, 23 (November), 387-
93.

Kincade, Doris H., Woodard, Ginger A., and Park, H. (2002). Buyer-seller
relationships for promotional support in the apparel sector. International Journal of
Consumer Studies, 26(4), 294-302. '

Kline, B. and Wagner, J. (1994). Information sources and retail buyer decision-
making: The effect of product-speciﬁc buying experience. Journal of Retailing, 70(1), 75-
88.

Knight, J. G., Holdsworth, D. K., and Mather, D. W. (2007). Country-of-ori gin
and choice of food imports: An in-depth study of European distribution channel
gatekeepers. Journal of International Business Studies. 38(1), 107-125.

Kono, T. and Lynn, L. (2007). Strategic new product development for the global
economy, New York: Palgrave Macmillan.

Kotler, P. (1997), Marketing Management: Analysis, Planning, Implementation,
and Control, 9th ed. Englewood Cliffs, NJ: Prentice-Hall.

Kuester, S., Gatignon, H., and Robertson, T. S. (2000). Firm strategy and speed of
diffusion. In Mahajan, V., Muller, E. and Wind, Y. (Eds), New Product Diffusion
Models. 2747.

Kumar, N., Scheer, L. K., and Steenkamp, J. (1995). The effects of supplier
fairness on vulnerable resellers. J oumalgof Marketing Research, 32(February), 54-65.

Kumar, S. and Phrommathed, P. (2005). New Product Development: An

Empirical Study of the Effects of Innovation Strategy, Organization Leamingand Market
Conditions, New York: Springer.

 

Lane, C. and Bachmann, R. (1996). The social constitution of trust: Supplier
relations in Britain and Germany. Organization Studies, 17(3), 365-95.

La Porte, TR. and Metlay, BS. (1996). Hazard and institutional trustworthiness:
Facing a deficit of trust. Public Administration Review, 56, 341-347.

Langerak, F., Hultink, E. J ., and Robben, H. S. J. (2004). The impact of market
orientation, product advantage, and launch proﬁciency on new product performance and

organizational performance. Journal of Product and Innovation Management, 21, 79-94.

Larsbn, A. (1992). Network dyads in entrepreneurial settings: A study of the
governance of exchange relationships. Administrative Science Quarterly, 37, 76-104.

111

Lee, H. L. (2002). Aligning supply chain strategies with product uncertainties.
California Management Review, 44(3), 105-119.

Lee, Y. and O’Connor, G. C. (2003). The impact of communication strategy on
launching new products: The moderating role of product innovativeness. Journal of
Product and Innovation Management, 20, 4-21.

Li, H. and Atuahene-Gima, K. (2001). Product innovation strategy and the
performance of new technology ventures in China. Academ of Management Journal,
44(6), 1124—34.

Li, T. and Calantone, R. J. (1998). The impact of market knowledge competence
on new product advantage: Conceptualization and empirical examination. Journal of
Marketing, 62(October), 13-29.

Li, Y., Liu, Y., and Zhao, Y. (2006). The role of market and entrepreneurship
orientation and internal control in the new product development activities of Chinese
ﬁrms. Industrial Marketing Management, 35, 336-347.

Loveman, G. W. (1998). Employee satisfaction, customer loyalty, and ﬁnancial
performance: An empirical exarrrination of the service proﬁt chain in retail banking.
Journal of Service Research, 1(1), 18-31.

Luk, S. T. K. (1998). Structural changes in China’s distribution system.
Intemartional Journal of Physical Distribution and Logistics Management, 28(1), 44-67.

Lukas, B. A., Tan, J. J ., and Hult, T. M. (2001). Strategic ﬁt in transitional
economies: The case of China’s electronics industry. Journal of Maﬂgement, 27, 409-29.

Luo, X., Sivakumar, K., and Liu, S. S. (2005). Globalization, marketing
resources, and performance: Evidence from China. Journal of the Academy of Marketing
Science, 33(1), 50-65.

 

Lusch, R. and Vargo, S. L. (1998). Multiplex retailers versus wholesalers: A test
of the total value of purchasing model. International Journal of Physical Distribution &
Logistics Management, 28(8), 581-598.

Mardia, K. V. (1970). Measure of multivariate skewness and kurtosis with
application. Biometrika, 57, 519-30.

Mahajan V, Muller E, Wind J, (2000). New Product Diffusion Models. (Eds),
Boston, MA: Kluwer Academic Publishers.

Marr, N. E. and Thomas W. (1999). The acceptance/rejection of new products in

the retail grocery industry: The inﬂuence of background elements. International Review
of Retail, Distribution and Consumer Research, 9(2), 187-202.

112

McGoldrick, P. J. and Douglas, R. A. (1983). Factors inﬂuencing the choice of
supplier by grocery distributors. European Journal of Marketing, 17(5), 13-27.

McLaughlin, E. W. and Rao, V. R. (1991). Decision Criteria for New Product
Acceptance and Success. New York, NY: Quorum Books.

Metcalf, L. E., Frear, C. R. and Krishnan, R. (1992). Buyer-seller relationships:
An application of the IMP interaction model. European Journal of Marketing, 26(2), 27-
46.

Montgomery, D. B. (1975). New product distribution: An analysis of supermarket
buyer decisions. Journal of Marketing Research, XII(August), 255-264.

Montoya-Weiss, M. M. and Calantone, R. (1994). Determinants of new product
performance: A review and meta-analysis. Journal of Product Innovation and
Management, 11, 397-417.

Moorrnan, C., Deshpande, R., and Zaltman G. (1993). Factors affecting trust in
market research relationships. Journal of Marketing, 57(January), 81-102.

Mooy, S. C. and Robben, H. S. J. (2002). Managing consumers’ product
evaluation through direct product experience. Journal of Product and Brand Mmgement,
11(6/7), 432-446.

Moran, P. (2005). Structural vs. relational embeddedness: Social capital and
managerial performance. _S_taategic Management Journal, 26, 1129-1151.

Morgan, R. M. and Hunt, S. D. (1994). The commitment—trust theory of
relationship marketing. Journal of Marketing, 58 (July), 20-38.

Mowen, J. C. and Gaeth, G. J. (1992). The evaluation stage in marketing decision
making. Journal of the Academy of Marketing Science, 20(2), 177-187.

Nilsson, J. and Host, V. (1987). Reseller Assortment Decision Criteria. Aarhus,
J AI Press.

Nunnally, J. C. (1967). Psychometric Methods. New York: McGraw-I-Iill.

Ottum, B. D (1996). Launching a new consumer product. In Rosenau, Milton D.,
Grifﬁn, Abbie, Castellion, George A., and Anschuetz, Ned F. (Eds), The PDMA

Handbook of New Product Development (pp. 381-392). New York: John Wiley & Sons,
Inc.

Ozer, M., and Chen, Z. (2006). Do the best new product development practices of
US companies matter in Hong Kong? Industrial Marketing Management, 35, 279-292.

113

Patterson, P. M. and Richards, T. J. (2000). Produce marketing and retail buying
practices. Review of Agricultural Economics, 22(1), 160-171.

Pellegrini, L. and Zanderighi, L. (1991). New products: Manufacturers’ versus
retailers’ decision criteria. International Review of Retail, Distribution and Consumer
Research, 1, 149-174.

Phillips, L. W. (1981). Assessing measurement error in key informant reports: A
methodological note on organizational analysis in marketing. Journal of Marketiag
Research, 18(November), 395-415.

Porter, M. E. (1974). Consumer behavior, retailer power and market performance
in consumer goods industries. Review of Economics and Statistics, LVI, 419-436.

Rangan, V., Kasturi, M. Y., and Bell, M. B. (2000). RadioShack. Harvard
Business Review Case 9-500-081.

Rao, A. R. and Mahi, H. (2003). The price of launching a new product: Empirical
evidence on factors affecting the relative magnitude of slotting allowances. Marketing
Science, 22(2), 246-268.

Rao, V. R. and McLaughlin. E. W. (1989). Modeling the decision to add new
products by channel intermediaries. Journal of Marketing, 53(January), 80-88.

Rao, V. R., McLaughlin, E. W., and Hawkes, G. F. (1995). Supermarket buyer
evaluations of hypothetical new products: An empirical analysis, Aﬁgribusiness, 11(1), 21-
33.

Retail News Letter (2005). Retail opportunity ranking. Retail Forward, 531, 2.

Rindﬂeisch, A. and Heide, J. B (1997). Transaction cost analysis: Past, present,
and future applications. Journal of Marketiag, 61(October), 30-54.

Robinson, P. J ., Faris, C. W., and Wind, Y. (1967). Industrial Buying and
Creative Marketing, Boston: Allyn and Bacon.

Rogers, E. M. (1995). Diffusion of Innovations, New York: Free Press.
Rogers, E. M. (2003). Diffusion of Innovations. 5th ed. New York: Free Press.

Rotter, J .B. (1967). A new scale for the measurement of interpersonal trust.
Journal of Personally, 35 (4), 651—665.

114

Schurr, P. H. and Ozanne, J. L. (1985). Inﬂuences on exchange processes:
Buyers’ preconceptions of a seller’s trustworthiness and bargaining toughness. Journal of
Consumer Research, 11(March), 939-953.

Sebenius, J. and Knebel, E. (2006). The Bollingers: Negotiating with Wal-Mart.
Harvard Business Review, 9-907-009.

Shaw, S. A., Dawson, J. A., and Blair, L. M. A. (1992). The sourcing of retailer
brand food products by a UK retailer. Journal of Marketing Management, 8(2), 127-146.

Sheth, J. N. (1973). A model of industrial buyer behavior. Journal of Marketing,
37(4), 50-56.

Sheth, J. N. (1981). A theory of merchandise buying behavior. In Stampﬂ, R. W.
and Hirschman E. C. (Eds). Theory in Retailing, American Marketing Association, 180-
189.

Siguaw, J. A., Simpson, P. M. and Baker, T. L. (1998). Effects of supplier market
orientation on distributor market orientation and the channel relationship: The distributor
perspective. Journal of Marketing, 62(July): 99-111.

Silva, R. and Davis, G. (2002). Assessing the inﬂuence of retail buyer variables
on the buying decision-making process. Euroﬁpean Journal of Marketing, 36(11/12), 1327-
1343.

Sinclair, G., Ananth, L. and Anderson, J. (1998). The supermarket supply chain in
Shanghai. International Food and Agribusiness Management Review, 1(4), 443-450.

Simon, H. (1957). Models of Man. New York: John Wiley & Sons.

Skytte, H. and Blunch, N. (2005). Buying behavior of Western European food
retailers. Journal of Mariteting Channels, 13(2), 99-129.

Song, X. M. and Parry, M. E. (1994). The dimensions of industrial new product

success and failure in state enterprise in the people’s republic of China. Journal of
Product Innovation and Management, 11(2), 105-118.

Standiﬁrd, SS. and Marshall, RS. (2000). The transaction cost advantage of
Guanxi-based business practices. Journal of World Business, 35(1), 21-42.

Stemquist, B., Huang, Y., and Chen, Z. (2003). Market orientation in a
transitional economy. Working Paper.

Stemquist, B. (1998). International Retailing. New York: Fairchild Books and
Visuals.

115

Stemquist, B. (2007). International Retailing (2nd edition). New York: Fairchild
Books and Visuals.

Stemquist, B. and Chen, Z. (2006). Food retail buyer behavior in People’s
Republic of China: A grounded theory model. Qualitative Market Research: An
International Journal. (9)2, 243-265.

Stump, R. L. and Heide, J. B. (1996). Controlling supplier opportunism in
industrial relationships. Journal of Marketing Research, 33(4), 431-441.

Sullivan, M. (1997). Slotting allowances and the market for new products. Journal
of Law and Economics. 40(2), 461-493.

Tellis, G. J. (1998). Advertising and Sales Promotion Strategies. New York:
Addison-Wesley Educational Publishers.

Telser, L. G. (1980). A theory of self-enforcing contracts. Journal of Business,
53(January), 27-44.

Useem, J ., Schlosser, J ., and Kim H. (2003, March 3). One nation under Wal-
Mart. Fortune (Europe). 40.

Vargo, S. L. and Lusch, R. F. (2004). Evolving to a new dominant logic for
marketing. Journal of Marketing, 68(January), 1-17.

Wagner, J. and Ettenson, R., and Parrish, J. (1989). Vendor selection among retail
buyers: An analysis by merchandise division. Journal of retailing, 65(1), 58-79.

Walters, R. G. and MacKenzie, S. B. (1988). A structural equations analysis of
the impact of price promotion on store performance. Journal of Marketing Resear_ch_, 25,
51-63.

Webster, F. E. and Wind, Y. (1972). A general model for understanding
organizational buying behavior. Journal of Marketing, 36(April), 12-19.

Webster, F., E., Jr. (1984). Industrial Marketing Strategy, New York, NY: John
Wiley & Sons.

Weitz, B. A. and J ap, S. D. (1995). Relationship marketing and distribution
channels. Joumai of the Academy of Marketing Science, 23(Fall), 305-20.

White, J. C., Troy, L. C., and Gerlich, R. N. (2000). The role of slotting fees and

introductory allowances in retail buyers’ new-product acceptance decisions. Journal of
the Academy of Mgketing Science, 28(2), 291-298.

116

Williamson, 0. E. (1975). Markets and Hierarchies: Analysis and Antitrust
Implications. New York: The Free Press.

Williamson, 0. E. (1985). The Economic Institutions of Capitalism: Firms,
Markets, Relational Contracting. New York: The Free Press.

Williamson, 0. E. (1991). Comparative economic organization: The analysis of
discrete structural alternatives. Administrative Science Quarterlal, 36(2), 269-296.

Williamson, 0. E. (1993). Calculativeness, trust, and economic organization.
Journal of Law and Economics, 36(1), 453-486.

Willson, D. D. and Collier, D. A. (2000). An empirical investigation of the
malcolm baldrige national quality award causal model. Decision Sciences, 31(2), 361-
390.

Willson, E. (1994). The relative importance of supplier selection criteria: A
review and update. International Journal of Purchasing and Materials Management,
30(3), 35-41.

Wind, J. and Mahajan, V. (1987). Marketing hype: A new perspective for new
product research and introduction. Journal of Product Innovation Management, 4(1), 43-
49.

Wing, C. C. K. (1996). Strategic investment, uncertainty and distribution reform
in the People’s Republic of China: Why a gradual reform strategy works. Journal of
Applied Business Research, 12, 20-8.

Zucker, LG. (1986). Production of trust: Institutional sources of economics
structure. Research in Organizational Behaviour, 8, 53-111.

117

   

1|iﬂﬂtlijijjﬂiﬂilﬂijijji111113131